your job application was rejected by a human, not a computer

by Alison Green on October 13, 2020

Christine Assaf of HRTact.com recently published a kind of amazing break-down of something I’ve been trying to convince people of for years: that your job application was most likely rejected by a human, not a computer. It’s reprinted here with her permission. (Christine, by the way, is currently looking for an HR, people strategy, or I/O psychology-related role in the greater Boston area. You should probably hire her.)

Recently I attended a presentation where a commonly held belief was repeated and I feel the need debunk this. The speaker stated “75% of applications are rejected by an ATS (applicant tracking system) and a human never sees them…” First, I want to point out that recruiters will tell you this is false. As the main users of ATSs, recruiters have extensive experience and years in talent acquisition, and will tell you they hear this all the time and they cringe upon its utterance. But if you want to know my opinion on why this “myth” has infiltrated the job seeking world, scroll past all the research and jump to the end.

MY RESEARCH:

Secondly, let’s track down the origin of this false statistic. The speaker I heard it from cited topresume.com. So I did some digging:

From topresume.com

That topresume.com article (which includes the same false stat on an infographic) cites this Forbes article:

From Forbes

But Forbes doesn’t cite, well, anyone. But this particular Forbes article gets cited a lot in various places. Here’s one in Zipjob.com:

From zipjob.com

Moreover, cio.com uses a clever quote to make it think that Josh Bersin (of Deloitte) gives the 75% fact, but in reality they just lay it out there as if it’s truth and almost hidden as being given by Bersin/Deloitte:

From cio.com

The scary thing is how often I found similar statements to be uttered in online articles but with no background, link, or citation to the research or statistics behind it:

From careers.workopolis.com

Keep in mind, this is just the front page Google SEO search. So I went a few pages deep into Google search to track down the actual “study” that everyone seems to have read. I found two websites that reference the “study.” The first from careerminds.com:

From careerminds.com

The second from CNBC.com:

From CNBC.com

Both give no links, but instead reference “job search services firm Preptel” so of course, the next logical step is to find this company Preptel. But y’all – it no longer exists:

From rwdigest.blogspot.com

Yes, that’s right. Preptel, the company who provided amazing, job seeker resume-writing and a totally unbiased study went out of business in August 2013. In fact, remember that cio.com article I mentioned earlier that cleverly used Bersin/Deloitte quote to make it seem like the stat was legit? Welp, in this 2012 article look what they said about Preptel:

You’ve been hoodwinked.

FURTHER RESEARCH:

So where does that leave us? With my self-critical nature, ever the skeptic, and despite the cold-trail, I decided that there has to be some research on ATSs out there and rejection rates. Luckily, as I’m in the field of Organizational Psychology, I figured I’d put my free Google Scholar and journal publication library access to work. My keywords/terms were “ATS” and “rejection rate”. I spent entirely too long on this… and after wasting a full hour searching, I gave up. But here’s some interesting findings:

1. Found an interesting article from the Wall Street Journal (2012) that states that ATSs screen out half of all resumes according to John Sullivan, a management professor at San Francisco State University. I emailed him to ask for more info and I’ll update as soon as I hear more.

From WSJ

In fact, that same WSJ article is chock full of sketchy and insinuating language that, if one is not critical, would make a reader think that ATSs are rejecting applications. If you really read it – that’s not what the author ever explicitly says.

From WSJ

From WSJ

They finally say the quiet part out loud further down the article:

From WSJ (2012)

2. This study from 2014 from the Journal of Business Economics basically reviews ATSs and gives an empirical review of how ATSs provide a benefit to the organization (when utilized well). The authors state that ATSs provide better organization of openings and applications, better processes, and, as a result, shorten opening-to-hire time. There is no mention whatsoever of “computers rejecting applications” anywhere. And I cite an article that lacks direct information to emphasize that I wasn’t able to find any research on this premise.

3. I found a 2014 exploratory research article that discusses ATSs by reviewing Zoho Recruit (ATS software) and came to a few conclusions. They suggest that ATSs parse and limit applications and that the ATS will reject those it can’t read. But doesn’t give any experimental discovery or statistical analysis, nor does it explicitly say that the ATS will review and reject a job application.

4. Lastly, here’s a good article from the New Hampshire Business Review (2015) by Bill Ryan, a career services business owner, who talks about how ATSs don’t reject applications but instead what he calls “contextualize” applicants for recruiters.

CONCLUSION & THOUGHTS:

Look, the statement “75% of applications are rejected by an ATS system” is just false. It’s false, because there’s no concrete source data, or research to even back up the statement. In my opinion there’s three reasons why this myth exists and persists:

  1. ATS systems screen applicants. They do (depending on the system) increase the likeliness of ranking an applicant as a match for a particular job. And if that’s true, then conversely it means it will decrease the ranking if the candidate isn’t a good match. So what does that mean? It means that the ATS simply uses keywords, phrases, and “contextualizes” an application based on criteria provided by the employer. What it doesn’t mean is that a computer simply rejects applications. A human rejects the applicants; a computer ranks applications based upon data provided by a human. Ultimately, ATS systems screen applicants – they do not reject them.
  2. Career service industries, resume writers, and other professional career authorities out there really want you to believe that a computer is rejecting you and only they can help. So they perpetuate this statement in an attempt to sell their services. Ironically, sites like jobscan.com are putting them out of a job. These self-service sites provide a way for you to review your resume against aggregate algorithmic data of various ATS systems. In this way you can see what keywords are missing and attempt to revise your resume to better match the ATS system’s screening.
  3. Lastly, the real reason this myth exists and persists is because job seekers really want it to. Looking for a job is time-consuming and an emotional rollercoaster. You’re hopeful one moment, and you’re rejected the next. And the idea that our application was rejected because of “computer error” sounds way better than: we lacked the basic job requirements, lacked experience, or *gasp* neglected to tailor our resume to the job posting keywords.

I’ll once again reiterate that myself, many of my fellow recruiter colleagues, and even ATS companies will tell you that a computer is not rejecting your job application. Typically, the ATS is sorting it and a human is making the final rejection. So please, let’s stop continuing to use outdated and unsupported statements about ATSs. Thank you.

You may also like:
{ 288 comments }

{ 288 comments… read them below }

    1. a username*

      I’ve been on a hiring committee a few times and we receive the “ranking” where the computer tried to guess how well the candidate matches our needs, but we’re basically told to not take it at face value and read every application personally. We’re a niche skill set with a manageable number of applicants to do so, so I can’t speak to Big Generic Corporate Corp. but yeah, my university non-teaching dept is not letting anyone get ruled out without human evaluation.

      1. Pocket Mouse*

        Why receive the ranking at all then? Seems like the only influence it has would be to introduce bias.

        1. Diahann Carroll*

          I don’t know about the bias thing, but if you’re an employer that’s going to read every resume anyway, I’d get rid of the ATS to save money. There’s no sense in paying for something you’re not really using.

          1. Ask a Manager* Post author

            ATS’s do lots of other things than rank candidates though — the way I’ve used them, they’ve been configured not to bother with ranking at all, but they’re useful for tracking candidates through the process, sending rejections, sharing applications with others in the process, etc.

            1. a username*

              Yes, this! The system is managed and used mostly by HR, but we’re the department specific hiring committee to assess compatibility and skill. They pass it along because they have the information anyway, but we don’t really use it, and we would know if they were choosing to withhold candidates from us because of it – we can see that part in the hiring committee portal access we’re given.

              They’re HR for an entire university! So I understand automation helps to make sure no one slips through the cracks of not getting communication, etc.

            2. Diahann Carroll*

              I get that, but you can also do that stuff with cheaper software and email accounts you already have. And I’m mainly talking about in situations like a username mentioned above – smaller, niche employers can possibly do all of that stuff without spending money on an ATS.

              1. Something Something Whomp Whomp*

                You’re right in that at very small companies you can track applicants without an ATS; we did that at a <10-person org. The problem is that it's not an approach that scales well at all. Even at a small org, expanding a summer student program was enough to overwhelm our standard processes. We didn't invest in an ATS, but it wouldn't have been a crazy idea.

                Something that Alison left out is that many companies use ATSs that aren't standalone tools – in fact, they're multi-purpose HRIS products that handle other stuff like payroll and benefits administration. If you're at an organization that doesn't outsource a lot of HR stuff the way that small orgs often do, or is large enough that you can't easily do all of your HR or CRM stuff in Excel or some homespun thing, it becomes a different matter altogether.

              2. RG2*

                Not all ATS systems are that expensive, especially compared to the time involved in processing emailed applications. Additionally, employers are required to retain applicant records/interview notes/etc for several years and it’s a HUGE pain to do this if it’s all via email, especially if general record retention is a shorter timeframe. We use an ATS and it’s just to make sure we’re seeing every applicant and their applications/know what other jobs they’ve applied to at our company/communicating with them in a timely fashion/gathering and retaining interview feedback/retaining records over time in an organized fashion.

              3. SchuylerSeestra*

                I’ve worked at small companies. We still used ATS. It really does make the hiring process much easier. They store all notes, all candidate emails, all resumes. Its worth the investment. Tracking things with spreedshets or whatever is an absolute nightmare.

            3. Hush42*

              This is what we use it for. Our HR department mostly uses it as a way to keep track of candidates in a central system rather than just a stack of Resumes. It also allows them to keep notes on applicants that we liked so that if something else comes up that they’d be a fit for we have that information already. For example, yesterday I did a Phone Screen interview for one of the three positions I currently have open on my team and discovered just in that 1t minute conversation that while he’s definitley not a good fit for my team (Admin type roles) he would likely make a great candidate for the Sales team when we have an opening there. HR can put that information into the ATS so that, when a sales position opens, they can reach out to him to see if he’s interested.

            4. NotAnotherManager!*

              This is how we use them as well. I don’t even know if ours has a ranking system – I’ve never seen it. But it’s very helpful for doing mandatory hiring statistics reports, candidate tracking between recruiters/HR personnel (including emails and contacts, so, if someone’s out, all the notes are there and it’s a seamless experience for the candidate), bulk rejections of uncontacted applicants when the position is closed, etc.

              We could use Excel/Outlook to do this, but not with the same efficiency and ease. The ATS saves a ton of time and is really not that expensive compared to the manual work it saves. Ours is part of a larger, enterprise software package we use for other functions, too, I think.

            5. AcademiaNut*

              Also cases that have come up on this blog, where you get people aggressively over applying, and can filter them out instead of having to weed them out manually. Or when you have a lot of easily filterable unhireable candidates, like a MLIS job with tons of applicants who explain that they really like books.

          2. Sonson*

            Actually in my old job we used to use an ATS for managing the recruiting process as a whole and it was excellent for tracking notes, setting up interviews with candidates etc…
            We did set it to reject applicants who did not have the professional registration for the job advertised – if you did you click yes to the statement you were a professional teapot health checker and enter your professional number you couldn’t continue to complete the online application. Hugely beneficial. Why would someone who is a teapot seller suddenly think they are qualified to apply for a health checker role?

        2. Snow globe*

          I’d suggest the ATS is less likely to introduce bias (the illegal kind). Plenty of real studies have shown that humans reviewing resumes may be biased by information indicating gender or ethnicity. The ATS is ranking candidates based on job-related keywords in their resumes. Having a human review the ranked resumes can improve the accuracy of the results, but the initial ranking can ensure that well qualified applicants are sorted into the top tier.

          1. Indigo a la mode*

            I wonder if an ATS can strip resumes of identifying information before sending on the ranked list. That could be a valuable feature to help avoid bias at all pre-interview levels.

            1. SchuylerSeestra*

              There is seperate software that can remove identifying info. Some can integrate with specific ATS’s but not all.

          2. Ping*

            I’m going to challenge that. Amazon quietly ditched their AI hiring tool because it was systematically discriminating against women. It threw out all resumes that had women’s universities. It also favored action words, which were more likely to be used by male candidates.

            There’s plenty of articles of you want to look it up. The ACLU even addresses it.

            And that shows the key weakness of these systems. They only do what they are programmed to do. The HR person rarely has a deep understanding of the open position (especially tech ones). That means they may used the wrong key words and search terms.

            2. Lisa*

              Absolutely. As the article you posted states, “These tools are not eliminating human bias — they are merely laundering it through software.”

              1. Tau*

                Yeah, I was gonna say. It’s a known issue that AI can reproduce real-world biases depending on what data you give it and how you train it… but I sincerely doubt that most ATS include AI.

                1. Ping*

                  It doesn’t matter. You can inflict bias in ATS simply by the way you write job descriptions and also which key words you use.

                2. Ping*

                  AI is merely a software program, just like ATS.

                  Any program and be biased depending on how you program it. For example, job descriptions and even key words can be tuned to attract and even filter a certain demographic.

            4. LQ*

              The algorithms and machine learning tools are trained by humans with bias. You have to work really hard to peel that out of them because in large part of systemic biases.

              Stripping names off resumes is a good use of tools like this. But you can’t assume computers aren’t at least as biased as humans if humans are the ones who built the tools

            5. andy*

              Their AI system was trained on their hiring data. Basically, it just made bias inherent in their system more apparent.

      2. Washi*

        Idk about others, but when I’ve used an ATS, the ranking is usually crap! Other aspects of using one are awesome, like the ability to track where someone is in the process and send form emails, but I never even thought of the rankings as something a hiring manager would base their whole hiring system off of. Sort of how my phone has suggestions about how I want to complete my sentences, but my phone doesn’t write my texts for me, because the result would be nonsense.

        1. Aitch Arr*

          That’s been my experience as well, in 20+ years of HR. The ATS ranking is crap and we rarely keyword search within the database. Instead, we’ll look back at applicants to similar or the same type of job in the past and reach out to see if they are still interested.

    2. Miss Demeanor*

      Yes. My office is part of a larger system, but we still screen every applicant. There is some computerized stack ranking as noted in the post, but the HR system doesn’t send an automatic rejection, they just tell me who has what filled out. It’s on me as the hiring manager to determine if it matters. I will read every application as a matter of due diligence, but I’m also able to focus on certain applicants above others depending on what I hire for.

      1. Diahann Carroll*

        That makes sense – the rankings basically give you a list of candidates that you should absolutely screen right away because they have a high probability of fulfilling your immediate need, but then you also have another pile of resumes that you can go back to later if you don’t find what you’re looking for in the first batch.

        1. Something Something Whomp Whomp*

          …and we often end up hiring from that second pile because our top picks from the first pile either become unavailable or get screened out early for some (sometimes alarming) reason.

      2. Nicotene*

        I hate when I fill out a tedious database application (especially when it duplicates what was already in my resume in different cells) and then they STILL don’t ever send a rejection. C’mon, that should be easy! It’s not like anyone is painstakingly writing it out! You’re going to waste an hour of my time and can’t be bothered to send a simple template to all applicants??

        Ahem whew sorry apparently I’ve been sitting on that one for a while.

      1. Jack Be Nimble*

        My record, in my old HR role was 350 resumes, of which ~70 passed the initial screen. I had a list of four must-haves from the hiring manager (they had to have a portfolio, had to have a certain number of years experience, had to have done two specific kinds of work), which knocked out almost 300 candidates. There were a few who were grossly overqualified or had a salary requirement 2-3x more than the budget, but most just didn’t have the specific skills we needed.

        It is an incredibly manual process. There are some things that just can’t be automated with current technology; hiring is one of them.

  2. Teekanne aus Schokolade*

    Absolutely spot on! Any recruiter that has to spend time doing cold-calls is not going to fully trust a system that might just miss someone we need desperately. Our system essentially sorted candidates into pools of most likely to qualify. I spent so much time even going through the pool of those who had checked some clearly disqualifying question just to see we might be able to pay for the certificates the candidate lacked or to find out if they might fit a different role. I only ever let the system throw out those who had answered less than three questions (those who were likely applying as part of unemployment).
    Construction engineering in remote locations for the military has very specific licensing requirements and we seldom had a large pool of “overly qualified candidates”, so I was not about to accidentally throw the baby out with the bathwater!

    1. Where’s the Orchestra?*

      The ATS my job uses is similar to that. The only applications that are automatically discarded are those that are not completed within 72 hours.

  3. mmb*

    No offense, but it seems like this whole post just spent a lot of time telling us that it’s not just a computer that’s rejecting a resume, but a person who input some data into a program that’s REALLY rejecting the resume. Not much of a difference from where I’m standing.

    3. D3*

      There’s a difference. The fake data is being used to craft a narrative to sell services that are based on a lie.
      Lots of advice given to job seekers is bad advice, because it depends on this lie as well.

    4. Slinky*

      Well, yes, a person is rejecting the resume. I can only hire one person for the role, which means that all of the other applicants will be rejected at some point. As a human reading/screening resumes, I’m rejecting them based on how well they fit our required criteria. People too often buy into ideas like putting keywords in white text to “trick” the APS into thinking they’re a good match so they don’t get rejected. This is a terrible idea. Just write your resume and cover letter assuming a human audience.

    5. Snarkus Aurelius*

      The computer ranking is only as good as the human being behind the keyboard.

      I’m certainly not saying computers do a better job, but I’ve heard of stories where the human who input preferences royally goofed up the hiring process. For example, requiring five years of software experience on a piece of software that’s only two years old.

    6. Fuzzy Pickles*

      I can see where you got that. However, in order to accept that conclusion, you have accept that the person simply followed said rankings from the system. We don’t have info on that. So really, with such a lack of info, my understanding is that the myth could be true or false, worse or better. The point is that we can’t begin to suspect it to be fact without more days.

      2. SchuylerSeestra*

        Not all ATS use the ranking systems. I’ve used 3 seperate systems. I still had to review each resume I received. I could manually search via keyword, but the ATS did rank anything.

    7. AVP*

      Honestly, that was my initial reaction, too, mmb.

      I think it depends on the user though – so goes back to the human. An iffy or too-busy recruiter may let the ATS sort for them and may not read the bottom pile, a good recruiter will likely at least glance at them all, you never know what you’re gonna get, etc etc. Do I feel better as an applicant? Not really. Was it interesting to learn that this particular oft-repeated stat isn’t true? Sure.

      1. That Girl From Quinn's House.*

        Same here. The ATS reads your resume for keywords and sorts it to decide if it is a match or not. Then the human who gets the report has the choice of deciding whether they should take the easy way out and just read the 35 resumes the ATS flagged as matches, or the hard way out and read all 120 of them including the ones the computer flagged as not a match.

        And that’s before you get to Very Big Companies- I’m familiar with a few places where HR only releases the “match” resumes to the hiring manager for review.

        1. Something Something Whomp Whomp*

          Honestly, even at Very Big Organizations hiring managers still have some leeway when the “match” resumes don’t produce a shortlist they’re happy with. That can either mean asking to review the non-matches or reposting the position with a revised job description. The role of HR and ATSs as gatekeepers is really overstated.

        2. Hillary*

          I used to work for a manager who asked HR for all the resumes they’d screened out – it was hilarious and showed HR had done a good job with their screening. We were hiring for a manufacturing planner, but we had applications from both a city planner and a wedding planner.

    8. Wintermute*

      This was my takeaway.

      The difference between “the computer assigns a (low) score and a person reads the low score and rejects you” and “computer rejects you” is academic at best and saying there’s a difference is disingenuous at worst.

      1. MissGirl*

        No, the fact that she’s disputing is that 75% of applicants are rejected by a computer with no human eyes ever seeing it. That statistic is made up. She’s not saying that every resume gets a human eye but it’s a lot more than you think.

        This false statistic also leads people to feel they have no control of the hiring process or have to do weird things to “beat the system.” When the reality is your resume and cover letter still matter the most.

      2. Jack Be Nimble*

        I mean, it’s an academic difference only if:

        a) every single ATS has a ranking function (they don’t)
        b) every single ATS that has a ranking function ranks in the same way (they don’t)
        c) every single recruiter that uses an ATS blindly follows the recommendation of the ATS without injecting their own judgment into the process (they don’t)

        I’ve hired and hired recently. There’s not a computer that skims 75% of applicants out of the pile and whisks them away, untouched by human eyes. *That* is the myth that gets repeated, and that is what is being debunked.

    9. Something Something Whomp Whomp*

      It’s not, and it’s tiring to hear this retort even though we understand the frustration that job seekers have with the process. Lots of people here, whether they’ve been in HR, served on hiring committees, or have been hiring managers, know that this is kind of BS because we have lived experience on the other side.

    10. MissGirl*

      No, the fact that she’s disputing is that 75% of applicants are rejected by a computer with no human eyes ever seeing it. That statistic is made up. She’s not saying that every resume gets a human eye but it’s a lot more than you think.

      This false statistic also leads people to feel they have no control of the hiring process or have to do weird things to “beat the system.” When the reality is your resume and cover letter still matter the most.

      2. mae*

        Right but I agree with mmb, to me this reads like a fine distinction that matters more to HR types than to job seekers. My concern typically isn’t whether 50%, 75%, or 99% of jobs were rejected by a computer, or whether a person had to hit Enter to delete the bottom whatever-%, it’s whether I need to worry about including enough keywords to get through an automated system. Sounds like a resounding yes.

        1. MissGirl*

          No, keywords aren’t what’s keeping you out. Focusing on that isn’t going to solve your problem.

        2. Forrest*

          But keywords are what you should be including to catch a human recruiter’s eyes too! If they’ve put, “client-focussed team player with a five years experience in firework design”, those are the keywords they’re scanning for. Someone whose CV includes those words will simply stand out more than someone who hasn’t changed their CV header from, “track record in pyro industry customer service”.

          1. une autre Cassandra*

            I suspect part of this is people (unconsciously?) assigning a different value to “keywords” than they do to “the stuff on my résumé” when really they’re basically the same thing.

            1. Mints*

              I think this is true. There are viral stories or unsubstantiated tweets or whatever about somebody getting rejected because they say something like “Experienced in spreadsheets, including pivot tables, VLookups, etc…” but don’t mention “Excel” by name. But like, that’s one bullet point. The rest of the resume should still be a high match unless you somehow did that on every point of the job description, but that seems impossible

      3. BRR*

        Yeah I think the main point to take away from Christine’s article is there’s more human involvement than applicants think. I strongly agree with her second and third points at the bottom that this idea won’t go away because “job search assistance” businesses promote the idea and applicants want it to exist.

        People then spend all their time thinking an algorithm is the reason they’re not getting interviewed/hired and ignore that it could be their resume/cover letter or that while they could hypothetically do the job, a hiring manager’s job is to hire someone they think can do the job well.

    11. Massmatt*

      I was thinking something similar. I appreciate the work the writer did to really dig down into this myth, but IMO an ATS system rejecting applicants vs: an ATS that ranks applicants, the lower tier of which an HR person then rejects, seems to be a distinction without too much difference.

      I get that some HR folks don’t use ATS systems for ranking candidates, and several here have said they do but they also look at that bottom pile just in case, but at least for some positions that attract large numbers of applications, it would be odd if companies DIDN’T use software to winnow the pile.

      When I was on a hiring committee years ago we would get 10-40 resumes per opening for jobs that required some industry licenses. On this scale we could go through and quickly toss resumes that were clearly unsuitable, but this would be time consuming if we had hundreds of applicants.

      Resume spamming is definitely a thing; may I ask HR folks that handle jobs that attract large numbers of resumes (i.e. several hundred plus) whether they are looking at each one they reject vs: spot checking the lowest ranked pile as ranked by an ATS?

      1. fhqwhgads*

        I sort of disagree about the “distinction without difference”. The reactions I generally see to the 75% myth is people assuming they submit, the computer evaluates, and if they don’t “score” high enough, they get an autorejection immediately. That might happen sometimes, but nowhere near 75% of the time.

        What I have seen at my last several employers is more like: there is one opening. 400 applications are received. System sorts them. Hiring Manager reads them in the order they’re sorted. If after reading the first 50 or so they have half a dozen candidates who look good and worth interviewing, they interview those. If they don’t find what they’re looking for there, they read another 50. If by doing so they find the sorting was clearly out of whack, and what they wanted ended up lower, and unsuitable candidates ended up high, they reconfigure the sorting criteria in hopes that it spits it out in a more desirable order. They’ll keep going through the pile, very unlikely to go through all 400 if they’re finding good candidates sooner. But that’s always going to be true when the number of applicants is that huge for a single role. Which it always is.

        Then again, there’s also a bit of timing involved. Some positions are open to applications until X date, and then everyone reviewed after it’s closed. Others, they review them as they roll in. So depending on how many come in daily, and if the HM is checking daily, they might end up reading them in small enough chunks that the order didn’t really matter anyway.

        If knowing all that about a process still feels the same as “rejected without human eyes ever seeing it”…then I agree with the author’s hypothesis about why this myth persists.

  4. Yet another Alison*

    Great write-up.

    Google “(SYSTEM NAME) Applicant Screenshot” or something similar. You’ll find what the HR-side looks like. Applicants are ranked, not rejected, although a 0% or lower might as well be a rejection.

    I get it, it is FRUSTRATING as a job-seeker, but here’s my pet peeve: you can’t blast out 500 generic resumes and then complain about the existence of ATS and ATS ranking you very low (or “rejecting you”). Some jobs get 2000+ applicants, most that are completely unqualified. Resume blasting is crippling job candidates.

    1. Diahann Carroll*

      Yup. I didn’t start getting anywhere as a jobseeker until I started actually applying to very specific roles and tailoring my resume and cover letters to the job ads. It was way more time consuming this way, but I got much better results.

      1. Something Something Whomp Whomp*

        +1. There are fields where this strategy works particularly well (I suspect we’re both in those kinds of fields) and it’s a little bit more difficult for entry-level folk to pull off, but it’s worth it.

      2. Dan*

        I work in one of those fields. The amusing thing is that even when I apply to jobs that I am “well qualified” for, I only have like a 50% callback rate. Don’t get me wrong, in the grand scheme of things, that’s super high. But for a niche field where my background is NOT a dime-a-dozen, I’m always wondering what it was about my resume that made somebody go, “naw, this guy isn’t worth a lousy phone call.”

        But it also makes me feel bad for people who are “reasonably” qualified for lots of jobs, where it’s really hard to stand out. It has to be maddening to toss out 100 resumes and not hear a peep.

        Which is why quality over quantity is the right play here. IMHO, if you can’t write a stand-out cover letter for a job, you probably are wasting your time (e.g., if you can ship the same cover letter to more than a dozen jobs, you’re either not selecting your jobs very well, or you need to write a better cover letter). There’s a chance your blast could work, but when I apply to a job where I can write a stand-out cover letter, I take the time to do it, and it’s worth the effort. So I save myself the clicks and get picky about my applications.

        1. Forrest*

          THis is the advice we give students: applying to fewer jobs that you really want and putting significant effort into each application will have a better success rate than, “I’ve applied to a hundred jobs this week and not heard back from ANY of them!” What’s staggeringly frustrating is that many candidates on Jobseeker’s Allowance are required to apply for orb a large number of jobs a week: way more than I think you can do serious, good applications for. It’s horribly counter productive for both applicants and employers.

          1. Fried Eggs*

            When I went on unemployment (in Germany) I was worried about this, because I’d heard there was a requirement you reply to a certain number of jobs a month. I met with my caseworker, and it was 5. Not a typo. 5 applications/month.

            I averaged 10/month and had a 33% call-back rate. Spending a lot of time on fewer applications really works!

            1. OyHiOh*

              In the US, it’s common for Unemployment to require 3 to 5 applications a week. Resume blasting at it’s finest.

    2. Washi*

      I’m so glad that I’ve gotten a chance to do some hiring, because it truly drove home the enormous amount of resume-spamming that happens. (And it was pretty clear from the cover letters/lack thereof that the applicants were truly just sending their resume everywhere. So many people would just put N/A when asked to attach a cover letter.) It really drove home how important a decent resume and cover letter are. A cover letter that actually mentioned my nonprofit by name one time (instead of “your company”) would put you in the top 10% of applicants already.

      As I’ve noted elsewhere in this thread, I never used the rankings that the ATS provided because they were absolute crap, but if I could have magically filtered out the applications with no relation to the job + no cover letter to explain a career transition or something, I would have!

      1. BRR*

        Being on the hiring side was a real eye opener and I’m glad you brought up that point. Applicants who have had good interviews and often times the ones we’ve hired have awful resumes and cover letters.

        1. lindrine*

          Hiring manager chiming in. We do use an ATS. I don’t use the rankings at all. I’m looking at the resume, and flagging people who look strong and starting with those first. I hire developers and designers, so I’m looking at websites and portfolios. One thing I’ve learned is – make your resume easy for the hiring manager to scan. Make your certifications, years of experience, and highlights easy to find.

    3. Uranus Wars*

      Yes, and when I was an internal recruiter we would get the same applicant for multiple positions (IT, sales, editorial, marketing, web design). It was the same resume every time. Our APS would let us know that we already determined X was only a candidate for X job so we could pass by his application for VP of IT when all of his experience was in personal training.

      We would review every applicant and update our notes once per year. Or as close to a year as when that applicant applied again. It was very efficient but I still got an alert that it was happening and could “lay eyes” on the resume at any point.

    4. Dan*

      The worst part for many entry level roles where 2000+ resumes may not be unreasonable, is that while most (a majority?) may not be qualified (or otherwise screened out in a few seconds), 500 or so could easily be “hireable”. Getting through those is a chore. Sure, there’s probably 10 or so that are “OMG, we have to interview this person”, but the other few hundred are probably good enough to get the job done.

    5. FormerInternalRecruiter*

      I’ll always remember the time that I had a candidate apply to every open job that my company had in my country (over 200 jobs). I only realized it because I kept seeing his application in jobs that I was hiring for and he was not remotely qualified. I imagine that we weren’t the only company that we did that to. I don’t know what made him think it was a smart thing to do, he automatically went on my do not hire list.

    6. Wilbur*

      That’s a great tip!

      There’s also a corresponding issue on the employer side-incredibly generic job postings. I’ve seen tons of postings that either list so many things it might as well be 5 job postings, or are so sparse you can’t possibly tailor a resume.

  5. Peter Piper Picked a Peck of Pickled Peppers*

    Thanks Alison! I actually found and commented on this article via LinkedIn just a few days ago – it’s really good – we so often just believe statistics and it was super helpful to have someone actually investigate properly and debunk this myth.

  7. KayEss*

    As much as I love comprehensive research, I’m not sure there’s really a difference on the jobseeker’s end between “ATSs reject applications (because they don’t have the right keywords) before they are human-reviewed” and “ATSs rank applications without the right keywords as less of a match, and there’s a match threshold below which humans likely do not review applications if they’re getting hundreds of them.” The end result is the same perceived necessity of keyword-matching.

    1. mmb*

      Exactly, that’s what I was trying to get at above. It seemed like we were going to arrive at a big GOTCHA moment, but it really just boiled down to semantics.

      1. Koalafied*

        I think it does suggest a different scale of rank-based rejection going on, to my mind. 75% is most applications never being looked at. But if you tell me that it’s up to individual hiring managers whether they’ll look at 100% of the resumes personally or whether they’ll only look at resumes above a certain score….I just can’t imagine very many employers deciding to not even glance at the bottom 3/4 of the stack. That just seems crazy high for someone who has much more of a vested interest in making a good hire than they do in making a fast hire. I would estimate it’s something like 0-5% for small/medium businesses with manageable volume, 25-33% for larger companies getting 250+ applicants per 1 role, and maybe 50-66% for megacorps that get multiple thousands of applications for a position.

        So to me the article is saying yes, there is some filtering going on, but much much less than is claimed by companies trying to sell you solutions to filtering.

        1. Dan*

          I have a feeling much of this is role-dependent. If you’re hiring for a job with a common skillset, many people reasonably could do the job, and you’ve got a few hundred applicants, then you really should be able to make a good hire in no more than 50 resumes.

          OTOH, I work in a niche field where skimming 200 resumes for an open role wouldn’t be unreasonable.

      2. Something Something Whomp Whomp*

        Another way of putting this, that’s not explicitly stated, is that everyone who’s done any amount of hiring knows that keyword-matching often has such poor alignment with actual suitability that we often end up going really, really far down a rank list if we even bother using it at all.

      3. SchuylerSeestra*

        But its not a “gotcha”. The point is ATS’s do not automatically reject candidates. It has nothing to do with the decision making process. Hiring managers use it to track the hiring process. Resumes are stored in the database. HM can search resumes in the system via keyword, but thats all done manually.

    2. Combinatorialist*

      Because according to Alison and other people who do hiring here and other places, there isn’t a match threshold below which humans do not review applications. They say they look at all of them. I have never seen someone who actually does hiring say they look at less than all the applications.

        1. Washi*

          At least in my org, the ranking isn’t something we opted into. It was just a random feature of the ATS that we completely ignored.

        2. Combinatorialist*

          So lots of people are saying they don’t even use the ranking — so for a lot of people doing hiring, there is no point.

          The ranking can still have value. If you are sorting your applications into yes/no/maybe — the more strong candidates you have seen, the easier it is to be questionable matches in the no pile rather than the maybe. So if you have a rough sort that filters so stronger candidates are likely to be earlier, then you will have less work to sort through the maybes once you have seen the full pool.

        3. DANGER: Gumption Ahead*

          We used them to sort out what order to read the application materials. It was artificial, but we liked having a reason why Fredrica’s resume was on the top of the stack and Fredrico’s was on the bottom. Then we’d all read through the stack in the same order and come together with the resume’s of the folks we wanted to interview. The ranking had less to do with the candidate and more to organize things

      2. Lynn*

        This is where it feels weird to me to have try to have a quantified number, because I can’t think of any company that freely publishes this type of data, but everyone has anecdotal evidence. I know, for example, of at least 2 companies that do use programs to automatically weed out candidates — and have done so on the scale of thousands of applicants. I think it must likely depends on how many applicants vs how many recruiters are involved.

      3. KayEss*

        If that’s the case, then the takeaway should be that ATSs are purely organization/communication tools (which to some companies I’m sure they are) and the keyword matching is irrelevant because a human reviews every application. But there’s still references to being rejected because you “neglected to tailor [your] resume to the job posting keywords”? Either the ranking matters, in which case the ATS is having a tangible influence on whether an applicant is rejected, or it doesn’t, in which case there’s no need to keyword-tailor your application materials.

        1. PVR*

          But I think there’s a difference between a human looking at a resume and ultimately deciding to reject that applicant (regardless of the rank) vs a human never looking at the resumes that the ATS ranked in the bottom 75% of the applicant pool and those applicants being rejected sight unseen. The question isn’t whether an ATS influences recruiters and hiring managers or changes their processes in some way, it’s whether or not a human sees all of the resumes.

        2. Junger*

          I think that whey say “it wasn’t tailored to the job posting keywords”, they mean that the application didn’t look tailored to the job, and was therefore probably a generic application sent out to many different job postings.
          Which usually means the candidate didnt make sure that the job is a good match, or described their qualifications very well.

    4. Teekanne aus Schokolade*

      The only key words we ever looked for in my company was the required certification titles and I reviewed hundreds of applicants a day.

    5. Trout 'Waver*

      Yeah, this was my impression also. Unless the hiring manager is reading every resume (in which case why would they need an ATS to score them?), saying that the ATS didn’t reject the candidates is distinction without a difference.

      1. Combinatorialist*

        An ATS providing a ranking can still be useful even if you are going to read every resume. Let’s say you have 10 interview slots. If you read the resumes in a (very rough) order of “match”, then it will be faster to sort the top candidates out. If you know the best of your pool, when you get a candidate you can quickly evaluate if they are competitive in your pool

        1. Trout 'Waver*

          The order in which resumes are read introduces a bias, for exactly the reasons you describe. Initial candidates are judged largely in a vacuum. Latter candidates are scored against prior ones. This could be good or bad. But someone scored by an algorithm to be less of a fit and then having their resume read later in the process because of it certainly isn’t being done any favors.

          1. Christina*

            But…you have to start sorting them somewhere, and what’s wrong with using a tool to give you a place to start? If I have 100 resumes to get through, and know I’m going to at least skim all of them, how is it a problem to start with all of the ones that the system knows at least have the word engineer in them, or all the ones that mention a required certification? Sure, I’ll eventually read the ones that don’t include those things, and maybe there will be a gem I missed for some reason, but it seems more efficient to at least have a sense of what I’m starting with. Saying that reading resumes in a particular order benefits the first read and disadvantages the ones after that – how would you even fix this? You can’t read them all at the same time.

      2. Betty Boop*

        I got to the same place. In the highlighted WSJ article it says that the implication is not to remove human screeners but to use the ATS to narrow the field to be more manageable to review. And that they rank the candidates and low scorers don’t make it to the next round. In addition the example from the bank is that those who don’t meet the required minimum requirements are weeded out and sent recommended jobs that would better suit them.

        I understand the need to have minimum requirements but this is also where having a person review your resume would be helpful. What if those people who didn’t have exactly 2-3 years “cash handling experience” had other experience that a person could determine they were capable. It’s nice to know the statistic isn’t 75% but it is still stressful to know if you don’t simply match the keywords or specific experience the computer spits you out before a person can even review it and see if you’re worth giving a chance too. I guess our only hope is to a write a resume exactly tailored to the computers keywords and algorithms to even get a shot, instead of being ranked in a lower tier or rejected entirely??

      3. Washi*

        The ATS is not just there for the purpose of scoring, the ATS is collecting and tracking applicants throughout the hiring process. At least in my organization, the scoring was just a random feature that we did not use, because it was terrible.

          1. Washi*

            I assume on the ATS side, it’s a feature they use to promote their system as being able to save hiring managers time. However, in my experience on the organizational side, we just ignored it. I guess maybe there was a way to disable it but we didn’t bother and just reviewed resumes in the order they came in.

          2. Eeyore's Missing Tail*

            IME, removing a feature like that would be considered a modification and could increase the cost of the system. It’s cheaper to simply ignore a feature than to spend the time and money having it removed.

    6. Ask a Manager* Post author

      Frankly, the keywords thing is over-emphasized in job search lore too. To the extent that an employer looks for keywords, they look for a much broader set of keywords than most people envision. For example, if they’re looking for experience with high dollar fundraising, they’re going to look for “major gifts” and “fundraising” and “donor management” and a bunch of other words, not just “major gifts.” If you have a good, specific resume that shows a track record of achievement in the skills the job description is looking for, a decent keyword search is going to pull up your resume. If it doesn’t, it’s probably because you’re not qualified — not because you didn’t include a laundry list of words from their ad.

      1. MollyG*

        Your own guest poster brought the issue up herself when she wrote “or *gasp* neglected to tailor our resume to the job posting keywords”. This is strongly implying that keywords are so important that it should be obvious that you should write your resume so it will get past the computer. The *gasp* part it it drives my point home.

        1. Ask a Manager* Post author

          Yep, I disagree with her on key words. She’s a guest poster with her own opinions, not a doppelgänger of me. (Although perhaps I should put a asterisk in the post about our disagreement there.)

        2. AndersonDarling*

          Yep, if you are building a creative, stand out, success driven resume, then your specific accomplishments may not have these keywords.
          * Increased revenue 150% during first year of the Save the Lamma event
          * Built relationships with 5,000 new LammaLovers and increased engagement with 40% of current LammaLovers
          These doesn’t have any of the keywords and yet it is impressive. Keywords can be obvious for the recruiter, but not so much for the candidate.

          1. Awkward Interviewee*

            Unless “revenue” “relationships” and “engagement” are keywords – I would argue they should be. I do think the guest poster didn’t quite finish connecting all the dots, though. They debunked the 75% statistic but didn’t really address keywords, which is another huge concern / misconception.

        3. Koalafied*

          Huh, I had parsed that sentence as though she were mocking the importance of keywords. Like if I, a grown adult, were to say, “Our colleague got in trouble because she *gasp* dared to take a lunch break!” I’d be underscoring the ridiculousness of someone getting in trouble for that.

        4. Forrest*

          I read it as about the fact that humans look for keywords too. If you pick up a CV and you put out a job ad saying you want someone with strong writing skills, an interest in marketing and great attention to detail, those are the keywords you’re looking for. If you don’t see the words “marketing”, “writing/written” and “attention to detail/accuracy” on the first skimread, how much time are you going to spend on this candidate?

          1. hbc*

            I think that’s a key point. Like, maybe you get dumped in the bottom 10% ranking if you have no keywords and the reviewer stopped at resume 160 of 200, and the ATS has effectively screened you. But how likely is it that you managed not to hit *any* keywords and you’re actually in the top 10-15 people who a human would interview, never mind the finalist for the position?

      2. Mockingjay*

        Exactly. If my company is looking for another technical writer, there are many more keywords than just “writer”: technical, writing, editing, document management, SharePoint, NAS, deliverable, software, software version description, plan, report, Agile, waterfall, template, style, industry-specific systems and equipment, industry technical standards…

        These are just off the top of my head. The actual job description will have many more, depending on the seniority of the position and the project it will support.

        1. Koalafied*

          Yes! Even if a hiring manager does filter out some applications based on rank, I just can’t imagine anyone doing so with 3/4 of the applications in the stack. Hiring a new employee is a kind of exciting time where you’re dreaming about the perfect person you might be able to find and hire, and how much of a difference it’ll make for your team to have someone in that role… in my experience you want to cast a broad net because of the risk of missing that dream candidate over slightly different language.

      3. Lily Rowan*

        Yeah, it’s not about customizing to the keywords, it’s about customizing to the job. I would love to consider people with non-obvious work trajectories, but they have to explain to me why they are the right fit for this position!

      4. Thankful for AAM*

        Alison, I wonder if you think the ATS ranking matters more for entry level jobs, hurts more, just because so many people would qualify/apply?

  8. Anonym*

    Love this analysis! So thorough and helpful. There’s so much job search bunk out there, it’s nice to have some clear information. Especially when it undermines those that prey on job seekers’ anxieties to make money.

  9. Not A Girl Boss*

    Wow, this is an excellent example of the broader issue of how frequently nonexistent “facts” get repeated by what we think of as reputable news sources (WSJ? Really guys?) until no one even knows the truth about an objective and obvious fact like “who screens resumes?”

    But also, thanks, I totally was a believer of this myth.

  10. MissGirl*

    I’ve met people who have internalized this so much they don’t bother to apply online. My last three great, professional jobs all came from online applications.

      1. MissGirl*

        They look at jobs where they have a networking connection. They won’t apply if it’s strictly sending their resume to a computer without that connection. So they might apply online eventually to meet requirements but not cold calling.

        1. Diahann Carroll*

          Ahhhh, I see. Yeah, that’s probably a more prudent way to do it if you have a big enough/strong enough network. I do not, so I usually just end up applying to jobs online and hoping for the best.

          1. MissGirl*

            Maybe. But every job where I tried to leverage my network, I didn’t get and every time I cold applied, I got it. Go figure.

      2. Grapey*

        Word of mouth, which is how I landed my past 2 jobs. (I had to upload my resume as a formality for their system but that was after an initial in person interview.)

    2. Roscoe*

      i wont apply to taleo systems. i’ll apply to simpler ones though. Those ones where you have to completely re-enter your resume are just absurd

      1. MissGirl*

        If you can do that great but if I had refused to copy my resume again after uploading, I would’ve missed out on my six-figure job.

      2. Partly Cloudy*

        I agree that it’s ridiculous to have to do it, but I’ve always literally copied & pasted from the Word version of my resume for this. It’s obviously easier than re-typing everything and doesn’t really take that long.

        1. PersephoneUnderground*

          Protip- copy-paste from a Notepad version, not Word! Word has internal formatting stuff that can copy over into a browser and mess up layouts. If you use Notepad first (or any other strictly plain text editor) it will lose that invisible formatting.

  11. No Tribble At All*

    Interesting, thanks for tracking down the source of that statement! Would you consider filterable yes/no questions part of an ATS? My spouse got an insta-rejection because the web form, at the end, had the question: “Do you fulfill every job requirement of this listing?” Well no, he fulfilled about 80%, but the options were only yes/no. Being honest, he clicked ‘no’, so he instantly got rejected :/

    1. MissGirl*

      We definitely had this set up to auto reject at my last job. We had a question that was do you have experience in X field. If you selected no, it auto rejected you. It doesn’t toss your resume. It it puts you in another pool. If my manager went through the first pool without finding an applicant then he would go to the other pool.

      1. Megumin*

        That is how the ATS works at my university. It doesn’t send an auto-rejection, but it does put the candidate in the “not competitive” pool. So I am able to see every application no matter what, and depending on the position and the department and how generous HR is that day, I can move candidates out of that pool and interview them. I have found some awesome hires that way! But then sometimes HR gets really stingy and won’t let me. :(

    2. Would it were otherwise*

      There is obviously going to be a range of how organizations use an ATS. I am a state employee and if you don’t meet our required qualifications, we are NOT ALLOWED to consider you. Period. So the yes/no questions (“Do you have a _____ degree/certification?”) ARE deal breakers and will get you rejected.
      None of the soft stuff will do you in, but the requirements? Yup. That’s why, when we write our job adds, so little is in the required category and so much is in the preferred category. Sigh.

      1. Not A Girl Boss*

        Yep, I once accidentally clicked “No” to the US Citizen question. In my field that’s an absolute non-starter so I was, obviously, instantly rejected. It was just frustrating that I couldn’t go back and correct it.

        1. Mockingjay*

          The best systems I’ve seen are the ones that allow you to review your answers before submitting, either at the end or to go back a step. I’ve only seen those used by moderate to large companies, likely due to the high cost of the ATS or where a lot of questions are required due to compliance or security needs of that particular industry.

          1. Kelly L.*

            I once somehow must have accidentally clicked “no” to “can we contact your last supervisor.” I loved her and there was no reason I would have wanted them to not do so–it was either a mistake on my part or they misread it on theirs. But they got super confrontational about it in the interview.

            (It was a disaster for other reasons too, in that it ran on for hours and then someone got snippy when I made a joking reference to being hungry, but that’s a story for another day.)

      2. JustaTech*

        Yup, I’ve had this applying to a specific set of jobs with my local government. If you don’t have X hours of classwork in Y field (the narrowest possible version of that field) then you just aren’t eligible. It doesn’t matter if you have work experience that takes you well beyond that, the people who set up the job listing want that specific classwork.

        But that’s not the computer’s fault. That’s the decision of whoever set up the job listing, ie, a human.

    4. Hotdog not dog*

      I’ve had this happen too. From when I was a hiring manager, the ATS sent me everything as long as the applicant answered the “yes/no” questions correctly. When I’ve been the applicant, there have been times when I got an automated rejection within seconds of submitting. I greatly prefer applying to jobs where they ask for a resume and cover letter to be emailed to a human!

      1. orangewater*

        Yes, I’ve gotten those automated rejections as well. They come so quickly, a human being wouldn’t have even had time to open the resume!

        To be fair, I’d rather avoid working at a place that handles applications like this anyway, so no big loss. But if you have enough of these experiences, of course you’re going to be frustrated with the ATS process.

    5. NotAnotherManager!*

      Some human configured those questions in the ATS, and some people are terrible at coming up with screener questions that get to the right result. Asking someone to affirm they meet all job requirements is an awful question and is going to miss a lot of candidates with strong potential – 80% is a great match. I’m not sure I’ve ever had someone match requirements 100%. The job description is a wish list.

      I had a terrible internal recruiter about 10 years ago, and I ended up reviewing all resumes because they didn’t get what I was looking for at all. I wouldn’t let him write ATS weed-out question, ever. My current recruiter gets it, knows that there are only two show-stopper criteria, and could be trusted to use ATS weed-out questions sparingly and appropriately – and she also has a good eye for resumes that reflect the necessary skills, even if it’s not direct experience. People with solid skills can be taught how to apply those to a specific context pretty easily with some pretty straightforward, job-specific training, in my experience.

      I also think that, as someone else mentioned, rejected resumes are pooled and screened to confirm the system didn’t reject anyone unduly. A person actually approves the rejection notices. I’m sure this is not universal, but it’s been a consistent in my decade-plus of management/hiring.

  12. AndersonDarling*

    I can state that I have been booted out of ATS questionnaires when I answered a question “wrong.” So I know there was no way a human received my application because I wasn’t even allowed to complete it.
    I’ve been dealing with this with my husband who is a maintenance mechanic applying at big employers. You fill out the application and then get a series of questions like “Do you have 6 years experience as a building mechanic at a hospital?” Then you think, well, there is 10 years mechanical experience and 8 years of that was at a nursing home and senior living centers, so that isn’t technically a hospital, so no. And then you get booted out of the application. In one case, the ATS blocked the IP address so he couldn’t even apply for another job.
    I know ATS systems that are not thoughtfully set up reject high numbers of applicants. It’s all up to the recruiter and how aggressively they want to cull their applicant pool.

  13. Smithy*

    I think these kinds of statistics are important to debunk because of the questions asked about using class and coursework to count towards years of experience. If it’s just being done to make it past an ATS system – and THEN someone will be reading/paying attention.

    However, I also think that there’s a core truth that remains helpful to hold onto – and that’s once you submit an application, obsessing over it, why you did/did not get not a first/second interview, etc – none of that helps! When I last hired, all of the applications I reviewed had been initially screened by our HR and I have no clue on the exact process that they used before passing along resumes or not. In the last few months, I had a friend hiring – and for a number of “post COVID” reasons she ultimately received over a hundred applications before she told HR to shut down the system because she couldn’t possible review that many.

    There are inevitably going to be reasons that an “open job” may not be equally open to another. Or what it means when there are dozens and dozens of applicants for a job. So while trying to game the ATS system isn’t the goal, I do think that the advice of “don’t get hung up on any single job or application” remains critical.

  14. Trout 'Waver*

    This particular column is nonsense parading as science. There are many ways to find out what percentage of ATS-managed resumes are read by a real person. A personal anecdote and failing to source a dubious statistic are those ways.

    Presenting this argle-bargle as rigorous science misleads readers and contributes to an overall distrust in actual science.

    1. So long and thanks for all the fish*

      Yeah, while it’s useful to have someone try to track down the source of a claim, good lord. Internet sleuthing is not research. This is the kind of “research” that leads people to believe their Dr. Google degree means their kids will be hurt by vaccinations, and I’m honestly disappointed in Alison for printing it like this.

      1. hawk*

        Huh? This isn’t making any claims about the real number. All it’s saying is the 75% statistic is probably fabricated. Common sense would tell anyone that anyway, because it have to would differ wildly between industries / workplaces anyway.

    2. jose*

      Yeah, at best what you can get from this is that there probably hasn’t been a major public study done on ATS use. That doesn’t mean that every other company is doing exactly what Allison/Christine are doing.

      1. Trout 'Waver*

        The column succeeds in proving that some click-bait journalists didn’t properly source a statistic, and such their use of that statistic should be discarded. It does nothing to prove what actual percentage of ATS-managed resumes are discarded before being evaluated by a human.

        1. cosmicgorilla*

          There are also going to be cases where no one actually rejects you, you just aren’t ever looked at. I work for a Very Big Company, and it’s just not possible to look at every resume that comes in. The recruiter may look at the top x amount of candidates, start pushing them to the hiring manager, and go to the next posting. If you’re applicant #50 and 15 solid candidates have already been sent forward, your odds of getting looked at are smaller.

      2. Christine Assaf*

        To your point, Jose. Correct, as far as what I reviewed, there doesn’t seem to be any research or any study that directly addresses “automatic ATS systems reject rates” There’s plenty out there about ranking though.

    4. Ask a Manager* Post author

      The point is, it’s a statement that’s repeated over and over as fact (when people involved in hiring know it’s not true), and she tracked down where it started and showed that it was never based on the sort of actual data people assumed it would be based on it.

      1. Trout 'Waver*

        If that’s the point, make it the column title!

        “The dubious origins of the much cited “75%-rejection” statistic” would be an equally interesting title supported by the research done.

      2. jose*

        Well…we don’t know that. What we know is that the original source is a study done by a now-defunct company that certainly had an obvious bias. But this doesn’t prove that the statistic is wrong, or that there wasn’t a study, or that the study wasn’t based on actual data – all we have for that is anecdotes. (frankly, I’m not sure how people involved in hiring know it’s not true, unless they’ve been involved in hiring for a huge percentage of all companies everywhere).

        Heck, maybe the reason Preptel went out of business is because they wasted all their money funding a huge methodical cross-industry survey instead of focusing on getting more customers ;)

    5. Littorally*

      How is it parading as science? It’s just digging for a cited source and coming up with bias and failed business.

    6. L.H. Puttgrass*

      “There are many ways to find out what percentage of ATS-managed resumes are read by a real person.”

      How would you do that? I’ve had more than my share of experimental and empirial stats training and I’m not coming up with a good answer. Yeah, you could send a survey to a whole bunch of companies and ask, but that would just tell you how many companies admit to using the ATS to reject applicants (or to using an ATS score as an initial screening mechanism, which is effectively the same thing). If you want to know the percentage of applications that are rejected, you’d need to convince a large number of companies to let you see their ATS data and any data they have on how those applicants progressed through the process (assuming the companies even have that data).

      Short of mandatory reporting EEO-style, I just don’t see how it’s even possible to design a study that would do this.

    7. Al*

      A quick search on EBSCOHost found me 245 peer-reviewed articles on “applicant tracking systems,” and while I don’t necessarily want to spend the time sifting through articles that aren’t my field, it’s frustrating to me that the research that this author is looking for likely exists in some form, but hasn’t been pointed to. Certainly, ATSs have internal analytics through which ATS providers can judge what percentage of the received applications are viewed. I also think they likely have a strong financial interest to keep a lot of that information internal and confidential, so the author may not have access to it. That is, the ATS providor certainly wouldn’t want to admit that only 25 percent of their applications are ever seen and a hypothetical <1% of applications result in a job for an applicant. One, the applicant loses faith in the system. Two, we then have to ask ourselves if the proliferation of ATSs have created their own problems: more ATSs mean more opportunities for unfit applicants to apply. I don't know, but it's worth considering.

      If this article were discussing digital rhetoric, the proliferation of unverifiable "statistics" presented as facts, and the tendency of corporate media interests to manipulate those dubious claims so as to further their own agendas, I would probably be into this. Writing that article through the lens of employment/applicant economics as a case study would make a lot of sense. I think this article/post is valuable in that it can help job searchers on this website reframe their expectations about online job applications.

      Ultimately, the author *hasn't* debunked the statistic. They've reached a dead end with it and said, well, since I can't find an original, verifiable source for this, it must not be real. But this is, in my opinion, a different facet of the same problem: unverified research presented as truth.

      The implication is that the rate of computer-rejected applications received through an ATS is *lower*, but we don't actually know that. We could easily infer that there's a much *higher* rate of computer-rejected attempted applications given what we know about rejected applications before the submission step, i.e., the prescreen questions or personality tests that are given before a completed application. Would these be counted amongst rejections? We don't know.

      But what we do know is that this author couldn't verify that 75% statistic.

      I'm not criticizing AAM for posting this. Alison's advice has always been a detailed cover letter and customized resume for job seekers. This article would seem tacitly to support that advice and perhaps give hope or courage to job seekers. But let's not misconstrue an inability to locate a source or produce verifiable independent research as "debunking" a common internet "myth." We just don't know that's true.

      1. lazy intellectual*

        Another side note: The fact that the 75% statistic isn’t verifiable doesn’t mean there aren’t issues with the hiring process (which is the point I think people are trying to make when they use that factoid in the way of trying to sell you stuff.) The hiring process still sucks in a lot of ways (like over-requiring education and degrees), even if it isn’t due to ATS.

      2. Forrest*

        This seems the wrong way around. These are well-respected publications making a claim, and following the reference to the source fails to substantiate it.

        Not every claim can be objectively falsified. You’re going to let an awful lot of Internet myth stand if your standard for falsifying it is not, “the quoted source doesn’t exist or never says that” but the far higher (and often impossible) standard of “here is a study that proves it’s not true.”

    8. MCMonkeyBean*

      How is this presented as “rigorous science.”

      It’s presented as “Why does everyone say this, where did this originate? Oh it originated from a now-dead company claiming that this was a huge problem that they could solve for you for $29/month which seems like a particularly not credible source.”

      No one is claiming the scientific method here, and they’re not trying to prove any particular number is the correct statistic. Just that people should probably stop sharing that 75% statistic like it’s gospel.

  15. Blackcat*

    There is a relevant XKCD for this! It’s titled “Citogenesis”
    I’ll drop a link in the reply.

  16. Desdinova*

    From the applicant’s standpoint, is there really a functional difference between “the ATS rejected your application” and “the ATS scored your application so low that it didn’t meet the threshold for advancing to the stage where it’s read by a human”?

    Like, I can see why the distinction might be relevant from the hiring side, but from the applicant’s side it’s just semantics.

    1. AndersonDarling*

      I was also thinking this. If you have 20 years of related experience and are an expert in your field, you can still be ranked as #245 because you don’t have the exact degree the person setting up the ATS entered. If there are 10 good candidates at the top, then no one will look past them to find the diamond in the rough. They aren’t technically rejected, but they are figuratively rejected because the likelihood of a recruiter spending an hour to root through the low ranking applicants is so low.

      1. hbc*

        Look, if the hiring manager giving information on how to set up this ATS screening is saying, “They need a degree in purchasing,” you aren’t going to get any more of a fair shake by taking the ATS out of the way. I’ve seen that guy hire without help of a system, and he will personally toss the 20 year purchaser resume for the newbie with the right degree.

        At best, you’re trying to woo someone who has no idea how to articulate what they want, so it’s going to be a crapshoot and you’re going to lose out to the person who happened to mention that they can speak Dutch.

    2. I’m a Loner Dottie, a Rebel*

      It also connects with the type of position as well- if our ATS ranks you as a low fit for an entry level position that we hire for on an ongoing basis, you’ll still likely get a phone screen from a recruiter. If the same person applies for a Director level position (which happens… very frequently) their low ranking will likely not get them a second glance. A major thing to think about is that people frequently apply for things that they are WILDLY unqualified for and it’s necessary to be able to filter those people out very quickly.

    4. SchuylerSeestra*

      The ATS hasn’t rejected your application. The recruiter or hiring manager has. Same with application scores. The hiring manager looks at it first.

  17. Snarkus Aurelius*

    I don’t doubt that a BS stat would be used by scammy companies to get people to hire them, however…

    “A human rejects the applicants; a computer ranks applications based upon data provided by a human. Ultimately, ATS systems screen applicants – they do not reject them.”

    A human is still doing something based on what the computer says even if the original ranking information was provided by that same human. For example, AI merely reflects the biases that the creators and users already held. That’s why Facebook will ban content that disparages antivaxxers, but it’ll leave up an event post encouraging people to bring guns and fight in Wisconsin.

    My understanding of the computer ranking resumes is that it’s scanning application materials for words that are repeated in the job descriptions. At least that’s what the KSA portion of the federal government’s application website was all about, according to a former federal government HR professional who wrote a regular column in the Washington Post. That software didn’t care that you explained your qualifications and examples in a specific area, as I did. It was looking for a set of regurgitated words from the original KSA question or job description, which is basically Mad Libs.

    I don’t know about other screening systems so I’m confident my experience isn’t universal.

    I appreciate the article, but my *personal experience* with these automated systems has soured me. In addition to the federal government, I’ve applied to Facebook, Microsoft, Google, the Gates Foundation, and Amazon. All of these applications were submitted late at night (I am a night owl) with an automatic rejection to follow within a matter of minutes or hours. I find it very hard to believe a human is sitting on the other end of a computer, anxiously awaiting my application, and reviewing it at 3 AM local time. In about half of those cases, I was rejected within a five minute window of submitting my application.

    I’m old enough now to know that I simply do not wish to waste my time as I did in my younger years. I honestly have no idea who these companies hire, as they obviously hire people, but I know of no one who has gotten hired through that automated system alone. The people I know of who did get hired almost always knew someone within the company or they’d done previous work or they had someone keep an eye out for their application.

    1. Anonymous Educator*

      Yeah, I think it is important to make a distinction between “a computer rejects” and “a computer ranks and a human rejects,” but I also fully agree with you that the distinction can be a small one. If you get 2000 applications for one position, and the ATS ranks them, you’re likely going to spend most of your time looking at the top-ranked 100 or so applicants. You aren’t going to actually give the time of day to the “bottom” 1900. If the computer is automatically influencing what a human does, the computer is, in essence, rejecting those applicants.

    2. NotAnotherManager!*

      Also, federal hiring is in a class by itself, and there is a very specific method for responding to USAjobs posts to meet KSA requirements and pass screening (ATS or human). If you’re applying to the fed, particularly if you’re not part of a preferential hiring category like veterans, you’ll want to do research on their application process before going to the trouble of preparing and submitting materials.

      1. Snarkus Aurelius*

        I don’t think KSAs are required anymore precisely because they caused a lot of problems, and I suspect no one read them anyway.

        But a hiring process is problematic if it requires a lot of research and effort before you’ve even applied.

      2. Brooks Brothers Stan*

        There’s an entire cottage industry based on assisting people to get federal jobs as USAjobs has a *brutal* auto-filter. Even as a member of a preferential hiring category one slip-up will get your application auto-denied. Going in with the mentality that someone will see your resume with these jobs if going to be soul crushing, and is incredibly misguided. Chances are, a human never will.

      3. J*

        Am a fed, can confirm. Having advice from friends who were already in federal jobs on how to structure my application was key for me finally getting onto a cert and ultimately getting hired. It is not at all obvious from the outside. It’s really tough, especially for non-veterans.

    3. Elaine Benes*

      I agree… have definitely been automatically rejected before and you know when it’s immediate it’s not a human. It’s a little weird that this article assumes job applicants can’t tell the difference, and that all ATS is run exactly the same. I’m sure a ton of hiring managers DO look at all resumes… but I’m equally sure that there are large corporations with tons of resumes that are happy to let a computer weed out the bottom percentile for them.

      1. Snarkus Aurelius*

        Thank you. I couldn’t put my finger on what I didn’t like here, and you said it best.

        It’s pretty obvious when a computer rejects your application, particularly if there’s a lot of information in it.

        I would be shocked if a human being at any of the places I listed even saw my resume.

    4. Amethyst*

      Exactly. When I got laid off 2 years ago from my employer, I immediately reapplied for multiple positions, even ones I didn’t want, just to get back in. I was instantly rejected from them all & strongly doubt that my resume was even looked at. Even when I applied for the position I now hold at my job, I was immediately rejected upon submission. I was lucky that I had a couple ins & texted one of them (the same woman who told me of the position & to apply for it) & she flagged it to our boss, who had to email HR in order for them to even look at my application & push me through to the interview stage. Whatever their ATS, it’s crap & it needs to be fixed/updated. If it weren’t for those two, I never would’ve gotten back into the company.

      2. A drive by commenter*

        KSA stands for “Knowledge, Skills, and Aptitudes”. You have to read a list of KSAs and write for each one whether you can do it, how much experience you have in it, where you learned it, etc.

      3. Snarkus Aurelius*

        Knowledge, skills, abilities.

        They were the required essay questions that accompanied every federal job application, which really belonged in a job interview.

        A sample question would be: what specific experience do you have in project management and internal communications on high profile issues? Your answer and your resume would have to repeat those words frequently because that’s what the software looks for. I remember answering 3-5 of these per job, which can take hours, only to be rejected immediately.

        To make matters worse, because the automated system is so brutal, there are consultants and companies that do nothing but help candidates through the federal application process. Whether the feds realize it or not, they’ve narrowed their pool to people who have enough income to pay for such services.

        Several years ago, I thought Congress banned the practice and only required a cover letter and resume, basically saying HR had to review applications instead of a machine because too many people were getting screened out. I don’t know what the status is now, but I hope they got rid of it.

        1. Jennifer in FL*

          KSAs are still required for some federal jobs (usually the jobs where they have someone internal in mind for the job but legally they have to put it up on USAjobs).

          —“Your answer and your resume would have to repeat those words frequently because that’s what the software looks for.”—-

          And sometimes it’s not even enough to repeat those words frequently. When my husband was looking to get hired by the DoD a friend told him to put whatever was listed word for word, straight from the job listing, as the answer onto a resume that was specifically catered for that job opening, otherwise he would be screened out immediately.

  18. Zephy*

    It feels like the point Christine is trying to make is “Well ACKSHUALLY the ATS is just telling the human hiring manager which applications to reject, the computer is not doing the rejecting itself.” Which, while technically correct, functionally means the same thing – maybe some hiring managers or recruiters do at least skim all of the resumes that come in for a position, regardless of how the ATS ranks or groups them (as Teekanne aus Schokolade mentions above), but I have to imagine that–especially for low-level, high-turnover, high-applicant-volume roles like retail or foodservice, and most especially for roles like that at big national chains, many of which include a “personality test” as part of the application–hiring managers are just taking the top X number of applications as ranked by the ATS and dumping the rest of them. Sure, a human being pressed the button, but that human being didn’t decide which applications made the cut. The point that the people quoting this statistic all over the place are trying to make is that there are things you can do to “game the system” and increase the likelihood that your application ends up in that top 10 (or 20, or 50, or whatever) that the hiring manager will pull off the ATS to look at.

    1. Snarkus Aurelius*

      Thank you. I remember applying for temporary Macy’s holiday help. The application was intense! Because I said I’d been to graduate school, the system wanted to know the title of my thesis. WHAT? Anyway, I had to take a personality test that lasted at least an hour.

      I was rejected before I even left the store.

      I know a human being decides who hires people to work there. Of course. But they didn’t decide that I ranked so low that I didn’t make the cut.

      1. LunaLena*

        That pretty much happened to me too, years ago, when I applied to work at Borders bookstore. I spent almost an hour filling out the application and another hour doing the personality test… and I received a rejection email within 5 minutes. At 10pm. I’m pretty sure no humans were involved in the process.

        In fact, the only time I successfully got a retail position was when I worked at Gamestop as seasonal help, and I suspect that was only because I had been chatting with an employee who told the manager about me after I filled out a paper application. That and I am a lifelong female gamer, and at that time, female gamers were extremely rare and it was unusual to find female employees at a game store. Any time I was told to go online to apply for a retail position, I either never heard back or was instantly rejected.

        I do believe that many employers have humans who look at every resume that comes in (especially since I’ve served on several hiring committees myself now), but that doesn’t mean that there aren’t bots out there who very definitely reject applicants without any human intervention.

        1. Elaine Benes*

          It wasn’t done until after an interview, but Anthropologie has all applicants do a personality test that is so confusingly worded that it was really hit or miss as to who would pass it. It was a lot of questions about timeliness and stealing, which you would think would be straightforward- but not how they worded it! This was eons ago but I remember it would give you a situation of something like “Sally is 5 minutes late for her shift” and you would have to pick what the appropriate response/punishment should be from the company as a multiple choice (and this was required for all positions, so not people trying to go into management the majority of the time). Or you would have to rank how egregious the offense was. So basically the current employees would be crossing their fingers that the applicant they wanted to hire could pass the thing, so many of them failed and you knew they were great people from the interview… it was so dumb.

        2. Jackalope*

          Yeah, I hate those personality tests too. I also had to take the Borders test and never heard back (although later I learned that they were going out of business and in free fall at the time so that might be why). I specifically remember being frustrated with the introvert vs. extrovert questions. I mean, between spending all evening at a party with 50 people I barely know vs. all evening at home reading with the critters, I’ll choose the second option. That doesn’t mean that I don’t like being around people or can’t spend all day selling books to the public, which is a completely different social skill set than the party scenario.

          1. L.H. Puttgrass*

            Wait, you mean wanting to spend all day at home rading books wasn’t the preferred answer for Borders? I think I know now why they went out of business.

          2. Zephy*

            It’s been a while since I’ve done one of the personality tests, but the experience I had was 50-100 questions that are really just variations on “do you think it’s OK to show up late and/or drunk, steal from your employer, sass back to customers, and/or ghost?” asked over and over again, where none of the answers are particularly good. Not really so much MBTI-style personality questions about introversion/extraversion, etc. Especially terrible are the questions where, logically, you know the company has to have a policy about this specific situation, but since you don’t work there yet you can’t possibly know what that policy is, so do they want you to guess what the policy is? defer to authority? show initiative? mind your business? it’s a surprise ;) To say nothing of hiring managers who say they want a candidate who does X in Y situation, but punish the actual employees who do X when Y situation actually occurs.

      2. Zephy*

        Those personality tests are the worst. The fact that you can find answer keys for them is proof that they exist only to reduce the number of applications a hiring manager has to review. Even the most reliable and valid personality tests we’ve been able to devise really require (1) a professional who is familiar with the test to interpret the results (which costs money), and (2) more context and familiarity with the actual person to draw any useful conclusions from those interpretations. I’ll give you a hint: Macy’s isn’t shelling out for robust personality assessment tools for their part-time seasonal retail staff.

        1. Beth Jacobs*

          Yes. Personality tests shouldn’t be pass/fail, it’s about fit with a specific role and team. I’ve yet to see any company do it well, they should probably stop trying. Similar info can be gained from a well designed situational interview.

          1. Not So NewReader*

            Everyone has personality. To fail a personality test is not possible. Of course each of us has personality there for there is something wrong with the tests to indicate that we have failed./s

            I hate these things with the fire of ten suns. I usually withdraw my application when I encounter one.

  19. Pooper McPoopy*

    I used to be a manager in state government. To my dismay, when I hired my first position, I didn’t get to look at all the applications. An HR staff member did the first screening – a low-level personnel assistant with zero experience in our field and no understanding of nuance in experience levels. I – the person with years of experience who knows what kind of person I want – just got the top 15 ranked applicants and wasn’t even able to see the other names or resumes. It was ridiculous.

    1. Gloria*

      In my experience, it’s pretty normal for the HR recruiter to do the first screening to weed out the applicants who don’t meet the basic qualifications and only share the good applications with the hiring manager. Ideally, the HR recruiter should be trained to understand the basic qualifications of the job though.

      Depending how many applications you get, sometimes it isn’t worth the hiring manager’s time to review every application. I mean, do you really want to review 200 application? I know I’d rather just review the top 15.

      1. Pooper McPoopy*

        For our jobs, we got maybe 60-80 applications. And yes, I would rather have gone through them myself – even if it was 200! – because I knew of highly qualified people who had applied and I would have loved to hire but didn’t make the HR cut. It was incredibly frustrating. Someone with an associate’s degree in human services and two years’ administrative assistance experience should not be making those calls over someone with 15+ years of experience in the field.

  20. Ana Gram*

    Definitely true where I work. We have a couple questions that you need to check off (depending on the position, they’re things like, are you 21, do you have a HS diploma/GED, etc) but that’s the only way for a computer to reject your application. Otherwise, several people have reviewed it and made that decision.

  21. L*

    Why would hiring managers want to lose his candidates because of some algorithm?

    The only place this is remotely plausible is in extremely oversaturated fields where there are thousands of applicants per job.

    2. jose*

      Why would hiring managers want to lose good candidates because of racism, sexism, classism, etc? Well, they probably don’t, but they also probably don’t think they’re actually losing good candidates because of those things – they think they’re acting rationally and using the information they have to filter for the best possible candidates. Same thing here.

      1. Anonymous Educator*

        Yeah, sometimes people don’t act in their own self-interests. Or they have multiple competing interests and prioritize the wrong ones.

  24. jose*

    As others have said, this really feels like hair-splitting/semantics. In addition to the point made above that one source being wrong doesn’t immediately mean the stat is incorrect (the real conclusion here is that we don’t know – someone should do an actual study!), it’s also ignoring the point that unless every employer is actually reading every single application and resume, there’s functionally no difference for the applicant between “the ATS system rejected you” and “the ATS system put you in the bottom X% of candidates and we don’t even look at those”. That’s not to say that the answer is “try to game the ATS” vs “write the best resume/application you can”, but it doesn’t feel helpful to split hairs and act like these systems have zero impact on applicants either.

    1. Anonymous Educator*

      “75% is a made-up number” is not the same as “humans click a button to reject you based on what a computer recommends.” I fully believe 75% is a made-up number, but that doesn’t say anything about what the real number is.

  25. MollyG*

    From the applicant perspective, the important aspect is “Does a human actually read your resume and cover letter?” From this posting, it seems that that answer is still “Sometimes”. So the computer will rank the applications, and then the HR person will only review the “top” ones and click “Reject” on the rest without reading them. This is functionally equivalent to a computer rejecting it. The last comment she wrote “or *gasp* neglected to tailor our resume to the job posting keywords” is very telling.

  26. anon e mouse*

    There’s an old expression, “the absence of evidence is not necessarily evidence of absence” that seems to apply here. The truth is that many companies get too many resumes for open positions and resort to silly, arbitrary heuristics to narrow the pool. Whether that’s done via the ATS rankings, yes/no screener questions, or simply looking at your degree title/institution while ignoring your experience, it happens a lot. I agree, ATSs are misunderstood, but as a lot of other commenters are saying, it’s sort of beside the point from the applicant experience side. Applicants are frustrated because they can’t even get screeners for jobs they are highly qualified for and I don’t know that they’re wrong to find that frustrating! Just my 2c.

  27. Lora*

    Based on my questions to HR on the general topic of, “Extremely Qualified Person I Know Applied and their resume was not in the pool you sent me, how come??” the answers were a mix of the following:

    -HR person is a generalist, not a technical recruiter familiar with industry-specific terminology, and didn’t understand equivalent terms.
    -Manager person who input the job description wrote a terrible job description so there will be no excellent fits and HR is searching fruitlessly for this purple squirrel of a candidate. Most frequently there is little to no distinction between “must have” vs “nice to have” vs “what you will realistically get”.
    -Something else happened during the job description input part, occasionally computer-based (e.g. autocorrect changed important keywords, some pre-set keywords were used that they thought would be similar enough but in fact are not, some pre-set nonsense everyone has to fill out per company policy about lifting 50 pounds or whatever that makes it sound absurd), occasionally human. This is typically the part where the criteria of “must have 10 years of experience” and “must have experience in (5 year old) programming language” somehow get conflated to “must have 10 years of impossible experience.”
    -Some connection or database maintenance issue happened and a whole pile of resumes are missing from the database generally, they don’t know what happened, IT will get around to working on it the 12th of Never
    -A formatting / layout type of thing messed up the scanning process and now it looks all stupid and unreadable with strange line breaks and tabs, nobody could figure out what happened so they decided “eff it, move on, there’s other candidates.”
    -Hiring manager went on vacation / was out sick, HR had a question and contacted someone else in the department who screwed up and gave them the wrong answer. Don’t ask me why HR does this, it’s not a life or death situation, and god forbid the random person they ask is actually the one you were planning to fire and replace with the new hire…but it’s happened to me more than once.

    There are so many things that can go wrong with these solutions to “you’re gonna get 200 applicants and 80% will be grossly unqualified”. It definitely doesn’t have to be the software, that is just one of many things. Applicants have zero control over any of this and it’s super frustrating. I have had the most luck with technical recruiters who specialize in the industry, because they’ve been hired by the hiring manager to do the hard work of really looking and they know all the words, plus they have typically discussed ahead of time with the manager if their salary range is unrealistic or if they are looking for a purple squirrel. I get the impression that general HR people don’t have time for that kind of sit-down hand-holding work to get to a good place with the job descriptions and competitive salary ranges with a sense of the market.

    1. Diahann Carroll*

      HR person is a generalist, not a technical recruiter familiar with industry-specific terminology, and didn’t understand equivalent terms.

      This reminds me of when I was in a trainee program in insurance, and my manager – the program’s director – ended up essentially “firing” our internal recruiter and requesting a new one because said recruiter kept sending her resumes that didn’t remotely match what my manager was looking for. The job ad itself wasn’t vague or in anyway unclear, so we had no idea why she kept passing along bad fits. It could have also just been the pool itself too…

  28. Tangerina Warbleworth*

    Warning: goofy tangent.
    When I first saw the URL, I didn’t read it as “Top Resume”, but rather as “To Presume”. And thought, “Why would a website announce itself as making presumptions instead of relying on factual research?” Which is exactly what it does.
    Mind. Blown.

  29. Elbie*

    As a former hiring manager, I personally considered ALL applications that HR sent to me, but I suspect that I was not always given all of the applications that were submitted for the positions I was hiring for, because there were times when the resume of someone that had been specifically recommended to me never made it to my desk unless I asked for it.

    On a personal note, just over a year ago, I was in my own job search, and hoping to leave a toxic position. I saw a job at a different company that looked like a “perfect” fit to me even though it was a career change. I tailored my resume and cover letter, highlighted the ways in which my skills were transferable, submitted it online, and did not hear a thing. Weeks later the position was still posted. I reached out to a former colleague who worked tangentially with the hiring department and asked if she would be comfortable putting in a good word for me and passing along my resume and cover letter. Shortly after, I was invited for an interview and was subsequently offered the job about a week later! I later found out that the director had never received my resume from HR and that I had been screened out before my director even had a chance to screen me herself.

    Fast forward one year later, I am still loving my job, my team, and my company. I feel happier and more fulfilled professionally than I have in years. I have no regrets and am thankful for the ways in which it all worked out!

    1. Anonymouse*

      I think this comes back to one solution: networking and putting your self out there. It all comes down to “who you know”. Which I realize can create areas for inequality. But that seems to be our best avenues – doing our best to get around the computer so our resume/application gets looked at by a human and making connections so you can have someone put in a good word for you and highlight your resume/application.

  30. No Tribble At All*

    As interesting as it is to hear that the “75% of all resumes are rejected by computer!” stat is completely made up, this comment thread will be taken over by humans complaining about opaque hiring processes. There’s a lot of bad hiring processes out there, and we don’t know what % of resumes are ranked so low by the ATS as to be instantly discarded by the hiring manager. There are also ATS with auto-reject systems! Sorry, I don’t think this post will be quite as encouraging as you wanted it to be.

    Really, the lesson to take from this is that Someone should do a real study on how businesses use their ATS. If 75% of hiring managers say “yeah I start at the top third as ranked by the ATS and don’t look at the lower ranked ones unless no one gets back to me”…. then humans are basically rejected by the ATS.

    2. Littorally*

      Right! Everyone’s taking the study to say (and I think Alison’s headline unfortunately encourages this) that no ATS ever auto-rejects applicants, which is pretty blatantly false. It’s saying the statistic is made up by a company with a blatant motive in pushing the numbers, and perpetuated by people who haven’t bothered to do any research on it, and that many ATS systems do not auto-reject. That’s all!

  31. Amy Farrah Fowler*

    We do a LOT of hiring at my company and easily get 500 applications a week when we’re in a busy recruiting season. The ONLY thing that our applicant tracking system pre-screens out for is the question “do you feel comfortable with XYZ” (which is basically 90% of the core job requirement. Think “do you feel comfortable using exercise equipment” if hiring for personal trainers).

    If you answer no to that, we will not see your application. Everything else, we screen personally. The ATS we use helps to make it clear who is in which stage of the hiring process because we hire on a rolling basis, so while some people are being invited for a 1st phone screen, others are moving to a second interview or into training. With the volume, I don’t know how we’d track it all and make sure no one fell through the cracks without a service like the one we use.

  32. Phony Genius*

    Is anybody else having trouble seeing a couple of the images of the referenced articles? Specifically, the CNBC and CIO articles?

  33. Mockingjay*

    I’d love to see the job industry adopt universal standards for ATSs that:
    – use the same terms or fields for standard info (name, phone, email…)
    – align with Federal and State EEOC policies regarding data submitted
    – can provide ‘blind’ output of applicant data to mitigate race, sex, and other discriminatory factors that might influence hiring managers
    – safeguard applicant data and provide clear remediation steps for data loss
    – and other stuff I can’t even think of to make hiring easier in the US. Given COVID unemployment numbers, we need this standard NOW.

  34. dustycrown*

    What I experienced in a state university bureaucracy is that the people who reviewed resumes did it with a very narrow approach. So, for example, if the job description said you needed 10 years of experience driving a blue truck, and you had ten years of experience driving a red truck, you weren’t considered unless you specifically spelled out in your cover letter or resume that your red truck driving experience was directly transferrable to a blue truck. The folks in HR would not make that leap on their own. (Obviously, not every HR department works this way. But this one did.)

    1. That Girl From Quinn's House.*

      I’ve applied for jobs at a state university and they set up their screener questions for years of experience the same way. The job description will say, “3-5 years’ experience as a llama trainer.” And the screener question will say, How many years’ experience do you have as a llama trainer? a) 1-2 b) 3-5 c) 6-7 d) 8 or more.

      You’ll have to select b) 3-5, or your resume will be filtered out into the bin HR does not pass on to the hiring manager. Even though there’s really no reason why someone with 2 or 6 years’ experience as a llama trainer could not be successful in that role.

  35. anon73*

    I just want to stop getting emails from recruiters who’s robot told them I was a match when I’m CLEARLY not qualified. Every time I update my resume, I get loads of emails for bullshit jobs that if you actually LOOKED at my resume you would know that I am either not interested in/not qualified for.

    1. VARecruits*

      I would also like for systems to stop sending your unqualified resume to me as a suggestion telling me you’re super qualified and then when I make the point to read your resume to double check it is clear you’re not.

      Frustration is so real on both sides I get your pain. Some people just trust the systems (often things like ZipRecruiter) that give them a list and you can just send out a mass email to everyone. It is less time consuming that way because you’re only getting emails back from people who are actually interested – but that doesn’t mean they’re qualified. And, in your case, I can completely understand why if you’re looking for a job and you get an email about a potential opportunity it is so frustrating to then realize you aren’t actually qualified.

    2. Aitch Arr*

      I once got a job match alert email from a job board that said “this seems like a fit for you!”
      It was for a ‘Pet Waste Technician’. I am an HR Business Partner / Manager.

  36. Antrobus*

    Part of what I think people miss here is that ATSs are not sentient. They do what they are asked to do, so even IF it’s even partially true that an ATS is ranking an application low, or filing it to a “does not meet requirements” pile or whatever, it’s because a PERSON gave it the parameters and told it what to do. And lots of people (as we see here all the time) do not know how to hire well.

    I’ve done hiring, and I’ve seen this firsthand. HR added a degree filter to a job posting (per a TERRIBLE HR policy) – so anyone who did not have a degree in for example chocolate teapot design was automatically removed from the applications presented to the hiring committee. The job didn’t require teapot design, just familiarity with chocolate teapots. In fact, of the people who already had that job title, four didn’t have the degree! But HR had decided that anyone working in the Design Department needed a Design degree, regardless of experience or actual job duties (in this case, customer service and scheduling for the Design Department.) We missed out on so many candidates that could have been great for the position with their similar work in different fields because they didn’t get this particular degree within the last decade.

    The point being – the ATS wasn’t at fault for keeping us from seeing the otherwise qualified candidates. Terrible hiring policy that was practiced via the ATS was.

  37. TomatoTomahto78*

    I call bull.

    I have 20 years of success in my field – a resume worth being proud of – but only have an associates degree. I’ve been straight-up asked by application systems – yes/no – “do you hold the minimum education required: bachelor degree?”

    I have to answer no. Never once have been called for one of those jobs, despite my track record. Explain to me why that is if a human is even glancing at my resume.

    1. howaboutalittleinfo*

      The system my institution uses will absolutely filter your application into the trash based on a lack of degree (if the job requires one). No one will even know you applied. You also won’t be given any feedback by the system so you won’t know that your application was rejected before a human ever saw it.

    2. lazy intellectual*

      Technically speaking, a human could have looked at your resume and still rejected you. The takeaway from this article is that the ATS simply filters applications, but it is a human that ultimately rejects them. It is also possible they didn’t read your application, but they still rejected you themselves. (Although idk, there are some comments on here describing systems that do drop applications.)

      I also agree it’s total BS to reject someone who doesn’t have a Bachelor’s when they otherwise meet the requirements, but that is a separate issue from how ATS works.

  38. Triplestep*

    This is a perfect example of a distinction without a difference. Did anyone really think that there wasn’t a human deciding what key words the ATS looked for in our resumes?

    Once I started weaving words from the job description into my resume (I used a word cloud generator to make the most used words easy to spot) I got a much better rate of return on my applications. I am not saying I got called about jobs for which I was not qualified, but using the same “jargon” as the person who wrote the job description got me interviews. For example, if the post said “utilization”, but my resume says “use”, I simply changed my resume and the phone rang much more predictably.

    @Chrstine Assaf: When people talk about an ATS rejecting them, they know a human set up the parameters. What we really want to know – and what you may be able to answer – is how your industry seems to have come the collective conclusion that it’s OK to ghost us after we’ve prepared and shown up for (sometimes multiple) interviews?

    1. Ellen Ripley*

      “This is a perfect example of a distinction without a difference.”

      +1million. OK, the 75% stat that’s been repeated ad nauseum in articles is shenanigans, fine, but many people’s experiences with ATS screening are similar to what that stat suggests, which is why it continues to be repeated ad nauseum. Most job seekers don’t care about the inside-baseball stuff about how HR works and how ATSs are used in companies; we just want to get a fair consideration in front of a hiring manager, and ATSs generally get in the way of that, which is why we resent them.

    2. That Girl From Quinn's House.*

      “word cloud generator”

      You are a genius and I’m stealing this. That is brilliant.

  39. Am I Wrong? Or am I right? Or am I just Misinformed?*

    OK, I’ve always been SUPER confused on the tailoring the resume to to the job posting. The way I’ve always envisioned it (and that could be the problem), if a job posting makes a big deal about email marketing for example, and I have some experience with that (but it wasn’t my primary responsibility in prior roles), I write my resume to glorify that experience. But it seems dishonest to me to make my work experience be so much about something in reality it wasn’t. It seems like it’s practically lying.

    I guess one could argue that maybe I’m not the best fit for that particular role, but in my field (marketing), it’s common to make tasks (such as email marketing) seem like such a big deal and hard to understand, when in reality, it’s not that hard, and as long as one has a curiosity to test different messages/creative/subjects/lists, etc. etc. and apply what they learn, than they’re doing a job better than most.

    I could be overthinking this (it wouldn’t be the first time), but it’s always boggled me and everyone I’ve ever asked about it so I wonder if it’s boggled anyone else here.

    1. SuperDiva*

      Tailoring doesn’t mean exaggerating! In this example, it could mean adding a line about your email marketing work to the resume or cover letter, if you weren’t already including it. So for other positions it might not be worth mentioning because it’s not directly relevant and you don’t have extensive experience, but for this position it’s very relevant, so you want to highlight whatever experience/accomplishments you do have. Then the employer knows you’ve done some email marketing and can ask you more about it during the interview process. But if you omit it, the hiring manager will likely assume you have no email marketing experience, and that might be the different between bringing you in for an interview or not.

    2. Parcae*

      I think of it more as taking OUT the things that don’t pertain to a certain job. My “complete” resume would be like five pages long. That’s not helpful for the person reviewing it, so I trim the parts that aren’t relevant.

      In my most recent job search, I had two basic versions of my resume– one emphasizing my spout experience, the other focusing on handles– but they were both true descriptions of my work as a Teapot Designer. Then for individual jobs, I sometimes tweaked a line or two based on the job description, like playing up my teapot painting internship for a role that involved cross-collaboration with the painting department. It didn’t take long at all– the cover letters (sigh) were the hard part.

    3. MCMonkeyBean*

      You don’t have to glorify your experience, just make sure you include it if they have noted it is something they would be particularly interested in!

      Many moons ago I applied for a job at a bank. I did not include my time as a cashier at Target on my resume because it was my oldest job (just a few months in high school) and I thought it was less impressive than other things I had done. But when it came up in the interview they were like “oh, that’s good to know that you have had a job that deals with handling cash before.”

      So you don’t have to go out of your way to make it sound like you did more than you did. But maybe for another job the little experience you had in that would be irrelevant so you would leave it off entirely, but for this job you would make sure to include it.

  40. Didyoumeantodothat*

    Our University system will kick out applications based on minimum education (same application system used for faculty and staff). For example, if the job lists a Bachelor’s degree as a minimum requirement, then a checkbox appears on the application asking if you have at least a Bachelor’s degree. If you don’t check it, your application gets filtered out automatically (but the system doesn’t tell you that’s what’s happening). But here’s the really fun part, if the ad doesn’t say a degree is required, but that checkbox appears, I know someone didn’t set up the application form correctly on their end and it’s going to reject anyone who doesn’t check the box. I’m sure there have been plenty of qualified applicants over the years whose applications were cyber trashed by this software before a human ever saw it.

  41. e.g.a.*

    As an HR person/internal recruiter I can tell you that I often look at 50-100 resumes for a particular job. We do sometimes set up our system to filter out candidates who don’t meet a very strict requirement (for instance, US citizenship is required to gain a gov’t security clearance), but even then, it doesn’t reject them, it just puts them in a folder so the recruiter or hiring manager doesn’t need to look at candidates that we literally cannot hire for a role.

    These are also self reported, not from the ATS parsing the resume in a particular way.

    1. VARecruits*

      “Self-reporting” is the key here. I think the problem is that people are upset with an ATS when in reality they need to be upset with hiring requirements. People don’t consider there are legal protections built into hiring practices. So for instance when people are complaining about how a Bachelor’s degree is required but they only have an associates + experience so it isn’t fair that their resume gets filtered out…well…I have clients that for better or worse REQUIRE a bachelor’s degree and according to their legal department’s CYA procedures they can’t hire someone with an associates if the advertised job description says Bachelor’s. Same with your citizenship question, or age requirements in certain fields. If you tell me you don’t have x, that is you telling me you aren’t qualified – whether it is a form on a computer or me to my face, one thing just saves me some time.

  43. lazy intellectual*

    This article is very insightful, but I’m righteously indignant that the career services industry got away with blatantly lying about how the hiring process works for so long. Job seekers, from students to experienced professionals, have seen this “fact” on mainstream career advice platforms and it is very misleading. I also don’t appreciate the condescending attitude some career advicegivers have towards people applying via channels provided to us by companies, as if we were being lazy and complacent, and not trying hard enough to find a “back door” method.

    Although I will say, this “fact” didn’t really impact my job searching methods when I was job hunting, other than to target more small companies that didn’t use ATS. Whenever I would read “the ATS automatically rejected your application” in career articles, my takeaway was “welp, nothing I can do about that”. I just continued applying where I could.

  44. Former ATS Employer*

    I used to work for an ATS at a global company. In most cases you are correct that a recruiter evaluates. That said the software has assessments and you can have a question that, if answered incorrectly, will automatically status the candidate to a negative status and send a rejection email.

  45. Laura H.*

    Firstly, I’m kinda put off by the over abundance of salty in the comments on this. I understand it’s a touchy topic but one of the rules is be kind (to guest columnists or Alison and to other commenters).

    Getting salty over semantics isn’t exactly that.

    Secondly, and more on topic, I really like the perspectives of the hiring managers and those in recruiting and HR who actually use the systems. That’s stuff we all need to keep in mind. Whether or not we think there is an actual difference or not, the ATS and human look throughs are someone’s job.

    And we all want to be the best we can be at whatever job we have, right? We all have tools to use and have a responsibility to use them well.

    1. Disabled trans lesbian*

      I do not think most of us who are criticizing this column are salty. Rather, people are pointing out that from an applicant’s perspective, the difference between ‘rejected by computer’ and ‘rejected by human who uses computer ranking’ is basically semantics. Applicants, especially people from marginalized communities, know damn well that they’re not on an even footing with corporate HR, which this column basically glossed over.

    2. CW*

      It seems like people really want the statistic to be true. Debunking it just makes them double down on it with, “Okay maybe it’s not true, but it basically still is so whatever.”

  46. Disabled trans lesbian*

    Those automated systems make it really easy and convenient for companies to reject marginalized people behind the guise of “does not meet requirements”. Amazon’s incredibly sexist “AI” has already been mentioned in these comments, and I am certain there are other companies out there doing the same thing, not just gender-based, but also against BIPOC, LHBTQI people and disabled people. I am honestly rather disappointed that this does not even merit a mention in the OP, and I really believe this topic needs to be addressed by marginalized people rather than corporate HR.

    2. Christine Assaf*

      Hi – I completely agree that bias in recruitment is a disaster and many ATS systems exacerbate that issue. You’re right that my original post doesn’t address ATS rejection based on race, gender, disability, sexuality, etc. For the sake of this post (and its length) I was addressing the one single statement and researching the particular robo-ATS claim.

      For that subject matter though I’ll highlight some peers like Janine Dennis: https://www.linkedin.com/in/janinentruitt420/ or and Katrina Kibben: https://www.linkedin.com/in/katrinakibben/ who are well-versed and thought leaders in their own right.

    3. disabled genderqueer lesbian*

      You have a great point and it’s totally true – you may have read it but Ruha Benjamin has a great book on the topic of AI and marginalization, in case anyone else wants to educate themselves on this issue. I came to the comments to mention a different marginalization.

      I want to point out that basically a lot of this post is inaccessible to people who use screenreaders and other alternative reading technologies. I think it would be most inclusive to use text quotes from the articles in line alongside the screenshots – but there are other options such as image descriptions.

      I really appreciate how much work the author put into researching this and I think it would be relatively simple to make it more accessible to everyone.

  47. OwlEditor*

    I can only speak from my own experience, but I think some companies do have something like an ATS. Years ago after I lost my job, I was frantically applying for any and every job, even just something to bring in $$. I kept applying to work for a fancy chain hotel as a booking agent. I would get the email saying the position was open. I would apply online and my inbox was immediately filled with a rejection email. I kid you not. This happened once or twice. I found it funny as they had asked me to apply! The rejection email was always something about my not fitting the position or something. Not a big deal as I didn’t want the job that badly. Still, an interesting experience.

    1. lazy intellectual*

      Based solely on comments on here where people describe ATS they have used, it does seem like there are some systems that do automatically reject job applicants. But the “75%” statistic is false.

  48. Bette*

    I have been reading AAM for quite some time, but never commented before. I am so grateful for the advice in this column, so I don’t want to come down too aggressively on this particular post. That being said, this article left a bad aftertaste for me, for a couple of reasons.

    First, it’s way too long. Way, way, way too long. The author is stating that the “75%-80%” claim is unsubstantiated, and includes (by my count) 12 different articles. That’s eight too many. If her point is that this is fake news, that’s not really a surprise to anyone on the internet.

    Second, as many other commenters have already pointed out, she’s really splitting hairs here. The author’s grand takeaway is that, and I’m quoting the comment by Jose, above:
    “unless every employer is actually reading every single application and resume, there’s functionally no difference for the applicant between “the ATS system rejected you” and “the ATS system put you in the bottom X% of candidates and we don’t even look at those”. ”

    It’s not a very helpful article. I’m not sure why it got such prominent space in this column today.

    Truly, I love this column. Maybe that’s why I feel extra disappointed in this guest post. I don’t think it fits the tone of the community. I feel like I just got a condescending and ill-informed lecture by my bank about why overdraft fees are necessary.

    ;) That’s my two cents.

  49. Raccoon*

    If Christine is lurking here: you should get in touch with BBC’s “More or Less” radio show about this research! Debunking widely-repeated statistics in an entertaining way is their whole premise. Maybe you’d get an interview out of it. :)

    1. Christine Assaf*

      Hi! Just started reading around here and that’s great – I’ll look into it! Thanks.

  50. goducks*

    Thank you for posting this. As a person who has used ATS for years, I’ve never once used any computer ranking. Setting it up seemed like a giant PITA, and the fear that it would include/exclude based on key words out of context was real. It has always seemed so much simpler/faster to just screen the applications myself.

    There is one thing that is worth mentioning, though. It’s entirely possible that there is a stat about how many applications never get seen by humans that has nothing to do with ATS filtering. If I’m hiring, and I’m flooded with resumes, I’ll stop reading them once I get a manageable number of solid candidates, maybe 10. If I have 10 people who all look very strong, I’m going to move on to phone screens/interviews. Only if I get through those 10 and don’t have sufficient candidates will I go back to the well of unread applications. I haven’t been hiring during this current period of high unemployment, but back in 2008 it was very, very common to have 200 resumes in an hour or so, and 1000 in a day. At some point, continuing to read tons of very well qualified resumes isn’t going to increase your likelihood of a good hire. So yes, there are a lot of resumes that simply never get read. Maybe there’s just a ton of applications, maybe the resume came in after a couple of very strong candidates had been identified, all sorts of reasons.

  51. Raccoon*

    Something that was missing from this article: the basic fact that using a blanket figure like “75%” for this makes no sense. Really? A computer is going to reject 75% of applicants no matter what? No matter the industry, the company, the number of applicants, the number of qualified applicants? It’s meaningless and highly suspicious /on its face/ even before you get into where it comes from or whether it’s true.

  52. Bobboccio*

    One datum is my office, which uses an ATS, and the system itself screens out the majority of applications we ever receive, with no human review whatsoever. HR can go in and look at specific rejected emails, but rarely ever do that. Many job advertisements get >10k applications, so we are kinda stuck. I’d call us an employer of choice, but not for our HR skillz that is for sure.

    Now what percentage am I talking about here? I couldn’t say, and would be interested in a study, and I am not saying we are like the majority of employers.

    1. AnonyMouse*

      Ya, I think this is a very valid point. The size of company and quantity of applications really matter here. Even without the ATS, the majority of resumes will not be reviewed for a position that gets 10k applicants. The manager will probably review enough to get a good candidate pool going and toss the rest out of sheer necessity.

      I think people are upset by an arbitrary and unfair system, but this whole thing has very little to do with ATS and a lot to do with the popularity of the roles and availability of substitutes (aka other qualified candidates).

      If you have a specialized skill set or are applying at less popular employers, then your resume will probably be reviewed. Otherwise, luck will play a bigger role. It’s never nice to feel like a number, but I don’t really see the point of agonizing over this reality…

      1. Bobboccio*

        You make a good point. We received too many applications to review them all (before we had this ATS), and I suggested we draw names out of a hat and only look at those.

        My boss (probably rightfully so) said if it ever got out, it would do us too much PR damage.

        So instead we used a random number generator to draw a few resumes for further consideration. Result was the same of course, but better optics.

  53. Christine Assaf*

    1. Sorry for all the typos, y’all. Yikes!

    2. To some comments that this was an exercise in science. I agree – it was not. This was literally trying to see if a statistic I heard (which is basically an urban myth/legend) had any documentation to back it up. I spent a couple of hours on Sunday afternoon, (where I should have been working on my actual research project). Anyone who wants to challenge the accuracy of what I’ve written is more than welcomed! That’s how research works.

    3. One of the biggest discussions I’m seeing is that my writing didn’t address the ranking of ATS systems (as it concerns rejection). While I touched on it, I didn’t write more on it for two reasons. a) I haven’t had enough technical experience (or research) in them to feel like I’m well-versed in that subject matter. b) That would have derailed the intentional topic and made my writing much longer. This article was about the statement/myth – not how ATS ranks. Is that just nit-picky semantics? Maybe. I’d argue (as I did) there’s a difference. But I was more concerned with finding any study that proved the statement. That said, I do think many ATS systems have issues in ranking (and therefore rejection), especially as it concerns bias and disparate impact. But I’m glad this topic churns a discussion.

    4. Addressing keywords vs no-keywords. I’m not an advocate of “keyword loading” to beat the ATS system. But I’m also an advocate for making sure you do your research on keywords for your resume. Phrasing changes as careers morph. Ex: HR people now have titles like “People Adviser” and “Talent Leader” or “Chief Human Officer” so sometimes it does help to “play the synonym” game. Alison and I may differ on that topic.

    5. Lastly, this article is a good example of when we’re all faced with a long-held belief (especially one that may affect us personally) it may be hard for us to accept. In the past, I myself had thought the robot-ATS-rejection stat was true, but had recruiters and TA folks assure me otherwise. That curiosity led to this post. Curiosity helps us to challenge ourselves, others, and in the end makes for better breadth of knowledge. I’m glad many of you are challenging my writing. So, thanks.

  54. EBG*

    However, usajobs.gov uses an ATS. If applicants do not meet the certifications, they are never put forward to be screened by a human. And non-standard qualifications, such as a Master’s degree from Oxford University, for example, can result in an applicant being considered as not having met the certifications if the HRO who advertised the position did not input the requirements broadly enough. In many US Government job applications, there can be 200 applicants and only 20 make it to the HRO for further screening, meaning 90% of applicants don’t make it to a human. Applicants must meet every requirement in order to be further screened and make it to a human.

    1. Something Something Whomp Whomp*

      …how the hell does the federal government have an ATS that doesn’t understand what an M.Phil is?

  55. Manic Pixie HR Girl*

    To add to this, I am in government, and we don’t have an ATS, mostly because a good portion of our roles are internal-only and/or require standing on an eligible list. But we do have a good number of jobs (not the majority, but not insubstantial, either) that don’t require that. Some are extremely specialized, but many aren’t as specialized and we’ll get hundreds, sometimes thousands, of applicants. Since our rules surrounding min quals are quite strict (any wiggle room is actually built into the quals and is actually not wiggle room at all), in these instances we do a quick 15 second scan of the resume. If we can’t say the person meets the quals with that scan, they go in the no pile. There’s simply no other way to effectively screen these and get candidates to hiring managers in a timely fashion.

    This isn’t to say we won’t go back to the well – In fact, we often do. But the initial scan of the obviously qualified gets candidates in hand. I can also say too that a scan that quick means we aren’t even looking that closely (if at all) at names when we do it. It’s probably not all that different than an ATS.

    Also, I want to note, if the 75% number was substantiated … it’s probably because that many people who apply aren’t actually qualified for the role.

  56. Esmeralda*

    I’ve chaired a number of hiring committees over the years. As chair, I read every application — I take on the initial screen for minimum requiresments so that my committee members do not have to waste their time on it (no need for three to five people to come to the same conclusion about years of experience, minimal education requirements, etc)

    Now, I read them pretty quickly, since I’m screening for minimums. But I do look at every one and evaluate them.

  57. SchuylerSeestra*

    Thank you for this. As a recruiter the myth of the ATS is one of my biggest pet peeves. A lot of my friends come to me for job hunting advice and have asked how to beat the ATS. One even bragged about the old keyword at the bottom of the resume trick. I explain to them that the ATS is more like a project managment system recruiter/hiring managers use to track the entire hiring process. It does not make any sort of hiring decisions, a human does. I’m job hunting myself and will call out any posts that gives wrong advice about the ATS.

  58. Analyst Editor*

    I thought about it a bit, and the main point – that ultimately, the reason your resume wasn’t accepted is a human-driven thing, not something due to bad software, is well-taken and stands up.

  59. College Career Counselor*

    The ATS that current and previous institutions used had certain parameters set to sort applications. Based on what the candidate put in, it would either pass along an application for review (and if I’m not mistaken, this was set to the candidate needing the minimum credential for the position) or place the application in a “to be dealt with later” section. At no time did the ATS auto-reject a candidate (though it did sort them initially). Typically the hiring manager or the chair of the search committee’s job was to look at all the resumes not immediately put through to make sure that there weren’t sorting /category errors. Based on that person’s review, they were either forwarded to the rest of the committee for review, or marked as “did not meet requirements” and a letter was then generated by the ATS to that effect.

  60. regularreader*

    I appreciate this article and the thought and care of the author. Would just add that mythical stats like these are rampant – the other one I think is pretty baseless is “X% of jobs are never posted” or “found through networking.” This may be true in some fields but it’s not something I’ve ever experienced or observed, and like the ATS, fosters desperation and annoying forms of “gumption.”

  61. PrgrmMngr*

    Based on my current job search, I’m confident that real people read my resume and pass it along. I’ve been in a niche for nearly 2 decades that I’m ready to move on from. Hiring managers have to have an open mind to interview me for the jobs I’ve been applying for, and I’ve received plenty of interviews and will be starting a new position at the end of the month). All employers I’ve interviewed with have hundreds to thousands of employees (millions if you want to count the federal jobs as a single employer).

    On the flip side, the services that claim to match your resume to jobs have provided laughable results – science and technology jobs that require technical skills nowhere near my non-profit / social services skill set.

    When hiring, I’ve also received resumes that attempt to get around these filters by inserting duties from the job description; if they’re obvious, they undermine an application. Your retail job didn’t require underwriting loans and your food service job didn’t have any duties that included reporting to my niche non-profit title. It’s not a good way to start your relationship with a hiring manager.

  62. Rocky Mountain Higher-er*

    I’ve been in corporate recruiting for 20 years, and have used a variety of ATS’s in that time. They all have varying levels of customizability regarding ranking, filtering, rating, etc. It’s entirely possible that one of the options was to set up auto-reject for candidates, but I never worked at a company that had such a feature configured. Neither I nor anyone I have worked with have made a practice of rejecting candidates based solely on an auto-ranking/rating assigned by an ATS. It is fully possible to “game” those ratings to get an unqualified resume a high rating, but that absolutely does not mean that the human recruiter who reads it will not recognize an unqualified candidate. Just as that same human recruiter will recognize a qualified candidate even if the ATS rates them at 0%. There are many reasons that someone may not get moved forward on a job they think they’re perfect for, but one of those reasons is not because a robot/Skynet/whatever rejected them.

    There are certainly terrible recruiters out there that give this profession a bad name, as it seems to be a field that has a high rate of people failing their way into it. I’ve been on the receiving end of those terrible recruiters, and on a few occasions I’ve had the distinct displeasure of having them as coworkers. However, the vast majority of
    corporate (as opposed to 3rd party/staffing agency) recruiters I’ve known and worked with have always done their best to do right by candidates as well as hiring managers.

    I’d also like to note that there is a difference between a professional Recruiter and an HR person who does some hiring stuff on occasion. Frequently you’ll find the latter in small to medium businesses that don’t hire at a volume or frequency to necessitate a dedicated hiring resource, and that is likely the root of many complaints. Recruiting, when done right, is not complicated but it’s also not easy. There are equal parts art and science to doing it well, and it’s not something you’ll excel at if you’re not dedicated to it.

    1. Rocky Mountain Higher-er*

      I have also been involved, including within the last 6 months, in selecting and configuring ATS solutions at my companies. The focus of these searches has always been to find a system that meets requirements for regulatory compliance, ease of navigation for both recruiters and candidates, and ability to provide reporting on a variety of tracking metrics within the hiring process. The ability of an ATS to screen out applicants has never been one of the factors in selection. This is simply not a feature recruiters want. Our job is to fill jobs, and anything that would prevent us from seeing a candidate that would get that done is a thing we want to burn to the ground.

      That being said, an ATS is an Applicant TRACKING System, so the primary purpose is what it says on the label–tracking where candidates are in the process. What they are not, to the frequent frustration of many stakeholders in the hiring process, is a Customer Relationship Management tool. Some of them attempt to incorporate CRM functionality, but it’s generally pretty blah no matter what the sales pitch says. So then it generally falls on the recruiter (again, an individual human subject to the same frailties as every other human) to devise their own way of keeping track of who they need to follow up with and when and why.

  63. Jake*

    It’s definitely easier on the psyche to assume you don’t have the necessary magic words on your resume than to acknowledge the company looked at your qualifications and decided you weren’t worth talking to.

  64. AnonymousFed*

    I work for the federal government. We still use keyword checkers! Yes, a real person will look at your resume, but they seem to rely on the number of keyword hits as a way to rank applicants. And if there are “enough” people with a lot of keyword matches, then the “lower-ranked” people are not forwarded on to the hiring manager.

    Our HR is understaffed/overworked. They are doing the best they can. Fortunately things seem to be looking up.

  65. Thankful for AAM*

    I loved tje research and just want to say I was rejected by an ATS system at a large university once without human eyes seeing my application.

    When I did not hear back, the person who recommended me checked with her contact. My application did not make it out of the ATS. The application asked for 10 years of job history. I gave more than that but some of the jobs were part time. I had no idea that 1 year of part time work was adjusted to 6 months of job history. I had submitted the equivalent of 9.5 years of just history and was rejected. So I had to resubmit with more history. I did not get the job.

  66. Solar*

    The ATS probably isn’t rejecting you.

    That said, at larger companies in the tech field, it probably *is* some clueless HR person who is rejecting you, which is why it helps to have a referral! This is frustrating from the hiring side, too :-(

  67. parsley*

    I’m saving this for the next time a client demands I “””optimise””” their CV for ATS. If you have the experience, you have the experience. I can’t magically deduce what the employer’s ATS is going to be looking for.

  68. Jessica*

    Hiring manager here. We read all 50-150 resumes for each posting. And in fact, if a resume is a very good fit and the candidate neglected a cover letter, we’ll reach out to say, ‘your cover letter didn’t attach in our system correctly. Can you send it along?’ rather than rejecting a potential candidate outright.

Comments are closed.