LevelUpWith

Explore / Job Hunting in the AI Era: Pass the Bots, Win the Humans / Lesson 2

Lesson 2: How Applicant Tracking Systems Really Work

Learners follow the steps an applicant tracking system (ATS) takes: receiving an application, parsing it into fields, storing it, then searching, filtering and passing it to a recruiter. They also learn which common beliefs about ATS 'auto-rejection' are myths.

Free to read, no account needed. Part of Job Hunting in the AI Era: Pass the Bots, Win the Humans.

By the end of this level you'll be able to

  • Describe, step by step, what happens to an application inside an ATS from submission to recruiter review
  • Tell apart knockout questions set by a recruiter and automatic keyword rejection, and debunk at least three common ATS myths
  • Check one past application of their own for points where parsing or a knockout question could have failed

Why this matters

In Level 1 you saw that hiring now runs on software at both ends. The software employers use most is the applicant tracking system (ATS). In 2025 the CV-tool company Jobscan found an ATS at 97.8% of Fortune 500 companies (489 of 500), and Workday alone was used by more than 39% of them. Most of your applications to large employers will go through one.

An ATS is less like a robot judge and more like a library's cataloguing system. It takes in your application, files it in the right fields and lets a person search the shelves. The person still does the choosing. This level follows your application through that system step by step.

The five steps inside an ATS

  1. Receive. You submit through a careers page, a job board or an email link. The ATS creates a candidate record and usually sends an automatic confirmation email.
  2. Parse. The parser reads your CV and form answers and splits them into fields: name, contact details, job titles, employers, dates, education and skills. Think of a clerk copying a handwritten form into a spreadsheet. If the clerk misreads your layout, the wrong text ends up in the wrong box, or a box stays empty.
  3. Store. Your record goes into a database, often alongside your earlier applications to the same employer. It can be found again for future roles.
  4. Search and filter. Recruiters use the database. They might:
    • sort applicants by date or by job stage
    • search for terms such as "payroll" or "CSCS card"
    • apply rules that they set up themselves, such as knockout questions (explained below)
  5. Pass to a recruiter. A person opens your record, reads the parsed fields and your original CV, and decides whether to move you on, hold you or reject you.

What to do: remember that steps 2 and 4 are where things most often go wrong. Those are the steps you will check in your own task.

Knockout questions and keyword rejection

These two are often confused, so keep them apart.

Knockout questions are rules that a recruiter chooses and sets up. Examples: "Do you have the right to work in the UK?", "Do you hold a full driving licence?" or "Can you work in Leeds three days a week?" If your answer doesn't meet the rule, the system can reject you automatically. The help pages for Workable, Gem and Kula all describe this feature. Workable's documentation says a candidate who answers "No" still receives the normal confirmation email, is moved to a Disqualified list, and "does not know" that this has happened.

Automatic keyword rejection is the idea that the software scores your CV for keywords and silently deletes anything below a threshold. Some systems can be set up this way, but it is uncommon. In a survey Enhancv ran in September and October 2025, 23 of 25 US recruiters (92%) said their ATS does not automatically reject CVs for formatting, content or design. In the same survey, 84% said they use knockout filters for things like work authorisation, certifications or location. This was a small sample from a CV-tool company, so treat it as a sign of how things work, not as proof.

The practical difference:

  • A knockout question depends on your answer to a direct question. You control it, and one careless click can end the application.
  • Keyword matching is mostly something a recruiter searches with. It is rarely an automatic gate.

Three common myths, debunked

  1. Myth: "75% of CVs are rejected by an ATS before a human sees them." This figure is traced to a 2012 sales pitch by Preptel, a company that sold CV optimisation and closed in 2013. It never published a method. No credible study supports the figure.
  2. Myth: "The ATS reads your CV and decides." The ATS files and sorts your application. In most setups, people make the decisions. When an application seems to vanish, the more common cause is volume: a recruiter with hundreds of applicants who never reached yours.
  3. Myth: "A rejection within minutes proves a robot read your CV." A fast rejection more often comes from a knockout answer, an automated email triggered by a recruiter's bulk action, or the job being closed. Silence doesn't prove you passed, either: knocked-out candidates may get only the normal confirmation email.

The ranking and "fit score" tools that sit on top of some ATSs are covered in Level 3.

Your task: audit one past application

Pick one application that went nowhere, ideally one sent through a large employer's careers site.

  1. Find what you sent. Locate the CV file and, if you can, the confirmation email or your account on the employer's portal.
  2. Check the parse. If the portal shows your profile, compare each field with your CV. Are your job titles, employers, dates and contact details in the right boxes? Note anything missing or jumbled.
  3. If you can't see the profile, test it yourself. Copy all the text from your CV file and paste it into a plain text editor such as Notepad. Is it in a sensible order? Are any sections, symbols or dates missing?
  4. Recall the knockout questions. List every yes/no or drop-down question you answered: right to work, licence, location, notice period, salary, qualifications. Did you answer any of them hastily or in a way that undersold you?
  5. Write down one likely failure point and one thing you will do differently next time.

Checklist

  • I can name the five ATS steps in order
  • I checked whether my CV parsed into the right fields
  • I listed the knockout questions I answered
  • I found at least one possible failure point
  • I no longer assume that silence means a bot rejected me

Key takeaways

  • An ATS receives, parses, stores, searches and filters applications, and then passes them to a person.
  • Knockout questions are rules the recruiter sets, and they can reject you automatically. Answer them carefully.
  • Automatic keyword rejection is uncommon. Most rejections come from people facing a large volume of applicants.
  • The "75% rejected by ATS" figure comes from a 2012 sales pitch, not from research.
  • Parsing errors and careless knockout answers are the two failure points you can actually check.

A 6-question knowledge check follows this lesson in the app, free and without an account.

Keep going: levels 3 and 4 are free

Create a free account to open the next two levels, save your progress and pick up where you left off. No card needed.

Next: Level 3, AI Screening and Ranking Tools.

Create a free account →See the whole course

More like this

  • Practical AI for Small Businesses: Save Time Safely

    Learn where AI really saves a small business time and money, and where it does not. You will choose tools without overspending, use them in marketing, customer…

    Course · Business · Business & Finance · Capable · 18 levels · 4 free · Professionally curated by levelupwith.com

  • Your AI Rights in the UK: Data, Decisions and Redress

    Learn what rights you have in the UK when organisations use AI to make or shape decisions about you at work, as a customer and as a citizen. You will be able t…

    Course · Business · Business & Finance · Capable · 20 levels · 4 free · Professionally curated by levelupwith.com

  • Your Career Through the AI Transition

    Learn to tell AI evidence apart from hype and fear, check your own job task by task, and pick the skills, reskilling routes and career moves that suit you. You…

    Course · Business · Business & Finance · Capable · 19 levels · 4 free · Professionally curated by levelupwith.com

  • AI Product Go-to-Market: From Early Users to Pipeline

    Make each key go-to-market decision for an AI product, using one running example: an AI compliance tool for UK SMEs. The decisions run from choosing an ideal c…

    Course · Business · Business & Finance · Capable · 19 levels · 4 free · Professionally curated by levelupwith.com

  • Pricing and Packaging AI Products

    Price an AI product where every task costs real money. Working on one case, an AI assistant for small law firms, you will choose a value metric, model your cos…

    Course · Business · Business & Finance · Capable · 19 levels · 4 free · Professionally curated by levelupwith.com

  • Product-Market Fit for AI-Native Startups

    Learn to test whether people really need your AI product. You will choose the right customer and problem, run discovery interviews and prototype experiments, s…

    Course · Business · Business & Finance · Capable · 18 levels · 4 free · Professionally curated by levelupwith.com