Why Is My Applicant Tracking System Not Working? (w/Examples) + FAQs

Your applicant tracking system is probably not broken. It is filtering resumes exactly as it was set up to, matching exact keywords instead of judging real skill, which is why more than 98% of Fortune 500 companies now run hiring through one and still miss qualified people.

The stakes are real on both sides of the screen. A job seeker can submit a strong resume and never hear back, purely because the parsing engine could not read it correctly. A hiring team can burn weeks and real budget chasing a setup problem that has nothing to do with the software itself, and the cost figures further down reflect 2026 industry data, so confirm current numbers with your own vendor before you budget.

What "Not Working" Means in Practice

An applicant tracking system is software that centralizes job postings, applications, and candidate messages. It replaces email threads and scattered spreadsheets with one place to search. When someone says theirs "is not working," they usually mean one of three failures. Those are broken parsing, unfair keyword screening, or poor setup, and each one needs its own fix, since the same symptom can come from any of the three.

Parsing failure happens at the document level. The system's parsing engine reads resume text and pulls out education, skills, and work history. It turns that raw text into a searchable profile. Tables, columns, and text buried in a header or footer often break this step, so contact details can vanish before a human opens the file.

Screening failure happens at the matching level, once the text is read right. Most systems score a resume by counting overlapping terms from the job post. A candidate who writes "executive communication" instead of "stakeholder management" can score low, even while doing the exact job. That is a language mismatch, not a skills gap, and it is the single most common reason for a false rejection.

Setup failure happens before a single application ever arrives. A company that skips workflow mapping lets the new system get built around whoever spoke loudest in the room. That gap rarely shows up on day one. Instead it appears months later as duplicate job posts, dirty data, and reports nobody trusts.

These three failures often compound rather than appearing alone. A resume that survives parsing can still lose to a keyword mismatch. A well-matched candidate can still get lost inside a poorly built pipeline, especially at a company that never mapped its own hiring stages. Figuring out which failure applies matters more than blaming the whole system, since a parsing fix looks nothing like a workflow fix.

Which Situation Applies to You?

Not every reader here has the same problem, so the fix depends on which seat you sit in. A candidate needs to fix the document and the words it uses. A recruiter or hiring manager needs to fix the setup and the rules layered on top of it. Read the section below that matches you first, then come back for the mistakes and self-check further down.

If You're the Job Seeker

Your resume is likely being read, but it may be scored against the wrong words. Most parsers look for exact phrasing from the job posting, not a conceptual match. "Lean Champion" and "lean manufacturing principles" can read as two unrelated ideas, even though they name the same person. That single quirk explains far more rejections than most candidates realize.

The fix is mechanical, not creative. Mirror the posting's specific terms somewhere natural in your bullet points. Use a single-column layout with standard section headers, and submit a .docx file unless a posting asks for a PDF. None of that guarantees an interview, but it clears the parsing and keyword hurdles that block a good resume before a person ever sees it.

If You're the Recruiter or HR Lead

Your ATS is only as good as the workflow built inside it. Undefined hiring stages, missing permission rules, and duplicate job requests are among the costliest setup mistakes a team can make. All three happen before the first candidate ever applies, which is why they go unnoticed for months.

Map every step from job request to offer letter on a whiteboard, before you touch a single setting. Then set up the system to match that real process, not a vendor's demo workflow built for a sales pitch. A properly built system should also show its own false-negative rate, since a low interview-conversion number usually points to overly strict screening rather than a weak pool. That single number is worth checking before you assume the candidate pool itself is the problem.

How Modern ATS Platforms Differ From Legacy Ones

Not every applicant tracking system works the same, and treating them all alike is a mistake. Legacy platforms built in the 2000s and early 2010s, the kind people mean by "Taleo-style" software, tend to rely on rigid keyword rules and desktop-only forms. Their reports usually stop at basic applicant counts. One corporate recruiter reported using Workday, Taleo, Greenhouse, Dover, and ADP since 2015, and watched the older platforms fall further behind on mobile and screening quality each year.

Modern platforms close some of that gap with genuine AI screening, though how much varies by vendor. Recruiters active on hiring forums note that newer platforms such as Ashby, Gem, TeamTailor, and Loxo already use AI to screen, rank, and reject candidates. Others counter that this stays rare enough that most applicants will never meet it, and expect that balance to shift within two or three years. Treat any single platform's AI claims as unproven until a vendor demo shows them firsthand.

The practical gaps show up in daily use, not sales copy. A modern platform often has mobile-first forms, since most job applications now start on a phone. It usually adds live links to payroll and background-check vendors instead of manual work. Its reports also go beyond a raw applicant count to show exactly where candidates drop off.

Knowing your platform's generation matters for one plain reason: it tells you where to spend your energy. A team stuck on a legacy system should focus on workflow discipline over waiting for an AI upgrade, since discipline is free and a new platform is not. A team already on a modern platform should instead audit its AI screening rules for the bias risks covered later in this guide. A legacy system can still work well if someone has truly mapped the workflow inside it.

A Worked Example: What a Bad ATS Rollout Costs

Cost is where an ATS problem stops being an idea and starts showing up on a spreadsheet. Picture a 150-employee logistics company whose aging ATS is holding back hiring. It switches to a new platform, expecting the subscription fee to be the main expense. The subscription turns out to be the smallest number in the whole project.

Cost research on setup finds that total first-year cost often runs three to five times the annual license fee. That multiple comes from migration, connections, and training stacking on top of the subscription itself. This sample rollout shows exactly where those extra dollars go.

Data migration for the 150-employee company runs $3,000. That figure sits inside the $500 to $5,000+ range one industry source tracks for this step, though other guides quote it in staff hours instead of dollars, so treat the number as a planning estimate, not a fixed quote. Linking the new ATS to the company's HRIS, payroll platform, and background-check vendor adds three connections at roughly $2,000 each, another $6,000. Before the software has screened one resume, the company has already spent $9,000 in cash costs on top of the subscription.

That upfront number is not the end of the bill. Companies typically see a 15% to 25% jump in time-to-fill during the first 60-to-90-day window after launch. Recruiters need that window to learn new screening rules, and candidates need time to adapt to a different application flow. A team that fills 50 roles a year loses several extra weeks of search time across the transition, the real, less visible half of the bill.

Staying on a broken system carries its own hidden price tag. A company that keeps limping along on a poorly built ATS keeps losing good candidates to false-negative rejections, month after month, with no line item to show for it. Fixing the workflow instead of swapping the platform often costs far less than the $9,000 rollout above, or the ongoing cost of losing good candidates for free.

Where Applications Genuinely Break Down

Cost problems and parsing problems eventually trace back to specific moments where a real person gets filtered out for the wrong reason. The three situations below are not the same failure wearing different names. Each teaches a distinct lesson about where an ATS truly breaks.

Daniel's Exact-Phrase Rejection

Daniel led three lean-manufacturing transformations and holds a Six Sigma Black Belt, yet a screening tool still flagged his resume as missing "lean manufacturing principles" and "key performance indicators." One job seeker with a nearly identical résumé reported the same problem: the system flagged missing "lean manufacturing principles" despite his Six Sigma Black Belt and three lean transformations. A corrective reply explained the mechanism directly: the parser never equates "Lean Champion" with "lean manufacturing principles," since it only matches exact wording. The fix is to mirror the posting's wording somewhere natural in your bullets.

Priya's First-Come, First-Served Wait

Priya applied to a warehouse-operations role four days after it posted, assuming a strong resume would be judged fairly no matter the timing. A recruiter explaining how these systems truly queue people put it bluntly: many platforms simply sort candidates on a first-come, first-served basis. A resume submitted late may never reach a human if the recruiter already found a fit by candidate eighty-five. Priya's real mistake was timing, not her resume, and the fix is applying within the first day a role posts, before the queue fills up.

GreenLeaf Retail's Configuration Failure

GreenLeaf Retail rolled out a new ATS without first mapping its hiring stages, so recruiters built the pipeline around whoever spoke loudest in the kickoff meeting. Within two months the system had duplicate job posts, splitting the same candidate pool across two listings. Employers swamped by pay-to-play auto apply bots face a related problem: in a typical flood of applicants, only 30 to 50 are truly qualified, which is why they raise the drawbridge and screen harder instead of more openly. GreenLeaf fixed this by mapping its real workflow on a whiteboard and naming one owner to approve every new posting, which cut its duplicate rate to nearly zero within a single hiring cycle and restored trust in its own reporting.

Run This Free Self-Check Before You Blame the Software

Before assuming the software itself is the problem, run a check that costs nothing and takes about ten minutes. Open your resume, or the job posting you are screening against, in a plain-text editor rather than a word processor. This single step reveals more about parsing failure than any paid scoring tool ever will, and it needs no account, no download, and no waiting on a result you have to pay to see.

Copy your resume's text and paste it into a bare-text notepad application. Watch for sections that merge together, work history that appears out of order, or a phone number and email that vanish entirely. That is exactly what a parser sees when it reads a table-based or column-based layout. Fix the layout to a single column with standard headers like "Work Experience" and "Skills," then repeat the test until the text comes through clean and in the right order every time.

For the keyword side of the check, place the job posting and your resume side by side. Circle every skill, tool, and certification the posting names, then check whether your resume uses those exact terms, not only different wording for the same idea. One frustrated user reported a first match score of 80% on one pass, then only 50% on a second pass of the very same resume. That swing undercuts the idea that these tools give a stable, honest number worth trusting on its own.

Treat any single tool's percentage as a rough hint, not a final grade. A parser cares about exact wording, not the skill behind it, so your own side-by-side read catches problems a percentage score can miss. Run both checks before you touch a paid service. The free method finds the same issues a subscription tool would flag, at no cost.

Mistakes to Avoid

• Formatting a resume in multiple columns or tables. Parsers read text linearly, so a two-column layout scrambles work history and education into an unreadable mix.
• Burying contact information in a header or footer. Many systems ignore that zone entirely, so a phone number or email placed there never reaches the candidate profile.
• Rewording skills instead of matching the posting's terms. "Executive communication" and "stakeholder management" describe similar work, but a keyword-matching system scores them as unrelated.
• Skipping the mobile experience test on your own careers site. More than 60% of applications now start on a phone, so a broken mobile form quietly costs a large share of your candidate pool.
• Configuring the ATS before mapping the hiring workflow on paper. The system gets built around whoever was loudest in the meeting instead of how hiring truly works, and that mismatch shows up as bad data for years.
• Running a single 2-hour training session at launch. Recruiters typically need 6 to 12 hours of hands-on practice, and a rushed rollout pushes people back to email and parallel spreadsheets within weeks.
• Ignoring the false-negative rate entirely. A low ATS-to-interview conversion number usually signals overly strict knockout questions, not a truly weak candidate pool, and no one checks unless someone is assigned to sample rejected applications.
• Assuming a resume-scoring percentage is a stable, objective grade. The same resume can score differently on repeated passes through the same tool, so treat any single percentage as a rough signal rather than a verdict.
• Skipping a bias audit on AI-assisted screening criteria. Screening rules that encode a biased proxy, like requiring degrees from specific schools, affect thousands of candidates at once instead of one hiring decision.

Do's and Don'ts for Getting Past an ATS

A short list of concrete moves beats a vague sense that "the ATS is bad," whether you are the one applying or the one setting up the system. These tie directly to the parsing, keyword, and setup failures described earlier. Focus on the ones that match your own situation.

Do

• Use a single-column resume with standard section headers. Parsers are built to recognize "Work Experience," "Education," and "Skills," and creative titles break the field mapping.
• Mirror the exact terms the job posting uses. A parser matches phrasing, not concepts, so using the posting's own words for a skill removes an entire category of false rejection.
• Submit a .docx file unless a PDF is explicitly requested. Compatibility with PDF varies sharply by vendor, while .docx parses consistently across nearly every platform.
• Map your hiring workflow before changing any settings. Documenting every stage, decision point, and person involved prevents the system from being built around the loudest voice in the room.
• Test your own careers site from a phone every month. A broken mobile form silently costs a large share of your applicant pool without ever showing up in a dashboard.

Don't

• Don't format a resume with tables, columns, or text boxes. These break the linear reading order parsers depend on and can scramble your entire work history.
• Don't put contact information only in a header or footer. Many systems ignore that zone completely, so your phone number or email may never reach the candidate profile.
• Don't stuff keywords repeatedly or in white text. Modern platforms flag unnatural keyword density as gaming behavior, which can hurt a candidate rather than help.
• Don't launch a new ATS without a bias audit on its screening criteria. An unreviewed rule can create real legal risk regardless of anyone's intent.
• Don't treat a single ATS match score as the final word. The same resume can score inconsistently across repeated passes, so use it as one signal among several.

Pros and Cons of Relying on an ATS

An applicant tracking system is neither a villain nor a fix-all. Weighing both sides sets realistic expectations before you praise or blame the software itself. The pros mostly show up when a team has truly mapped its workflow and kept the screening rules fair. The cons show up equally often when either piece is missing, which is why the same software feels like a lifesaver at one company and a liability at another.

Pros

• Centralizes every application in one searchable place. Recruiters stop hunting through email threads and spreadsheets for a candidate's current stage.
• Cuts time-to-hire when configured well. One 2025 benchmark report, cited by Jobvite, put enterprise time-to-hire at an average of 35.4 days, faster than that year's broader benchmark.
• Standardizes evaluation criteria across a hiring team. Structured scorecards reduce the influence of any one interviewer's personal bias.
• Improves compliance record-keeping. Applicant and hiring-decision records are retained automatically, which matters for EEOC and federal-contractor requirements.
• Surfaces silver-medalist candidates from past searches. A strong runner-up from an old role can be rediscovered instantly instead of the search starting from zero.

Cons

• Creates false negatives through rigid keyword matching. Qualified candidates who describe their skills differently than the posting get filtered out before a human ever looks.
• Adds real implementation cost beyond the subscription. Migration, connections, and training routinely push first-year cost to three to five times the license fee.
• Can encode and scale bias at a large volume. A biased screening criterion applied by software affects thousands of candidates instead of one recruiter's single decision.
• Frustrates candidates with poor mobile experiences. A desktop-first form on a system most people open from a phone drives real application abandonment.
• Requires ongoing maintenance to stay useful. A system set up once and left alone drifts out of sync with how the team truly hires within a year or two.

What to Do Next

1. Identify which failure you truly have: parsing, keyword screening, or workflow setup, since each needs a different fix.
2. If you are a candidate, run the plain-text paste-test on your resume and rewrite any section that comes through scrambled or missing.
3. If you are a recruiter, map your full hiring workflow on paper before changing a single ATS setting.
4. Pull your ATS-to-interview conversion rate and investigate any number that looks unusually low.
5. Schedule a bias audit on your screening criteria if you use AI-assisted ranking, and confirm your obligations under any applicable state law.
6. Test your careers site's mobile application flow from an actual phone, not a desktop browser resized smaller.
7. Bring in HR, legal, or an outside ATS consultant if the fix touches compliance, connection budgets, or a full platform switch.

Frequently Asked Questions

What Are the Problems With an ATS?

The three biggest problems are parsing failures, keyword mismatches, and poor setup. Parsing errors lose candidate data from tables and headers. Keyword mismatches reject people who word the same skill differently than the posting. Setup failures happen when a company never mapped its hiring steps before turning the system on.

How Do You Avoid Getting Rejected by an ATS?

Mirror the job posting's exact words and use a clean, single-column resume format. Parsers match specific wording, not concepts. Restate your experience using the posting's own terms, use standard section headers, and submit a .docx file unless a PDF is requested.

What Are the Challenges of Using an Applicant Tracking System?

The biggest setup challenges are undefined workflows, poor links to other HR tools, and low user adoption. Teams that set up the system before mapping their real hiring process end up with data nobody trusts. Disconnected systems then force manual re-entry, and recruiters quietly go back to email and spreadsheets within weeks.

How Can I Make My Resume Pass an Applicant Tracking System?

Use the same exact wording, tools, and job titles that appear in the posting. Stick to a single column and standard headers like "Work Experience" and "Skills." Submit a .docx file, since this is how parsing engines read documents most reliably.

Why Did the Same Resume Score Differently on Two ATS Passes?

Match scores from third-party scoring tools are not always stable between runs. Small formatting changes, a tool update, or a reweighted keyword list can shift the same resume's score by 20 points or more. That is why one score should never count as a final verdict.

Is a 70% ATS Match Score Good Enough?

A score in the 60% to 75% range is generally considered competitive. Scores are not standardized across vendors, though. A 70% on one platform can reflect very different keyword coverage than a 70% on another, since the score measures wording overlap, not real qualification.

Does Every Employer Use an ATS?

No, but the large majority of bigger employers do. Big companies rely on one almost every time. Many small businesses use lighter tools or none at all, since full platforms cost more to set up and run than a small team needs.

Can a Recruiter Still Find My Resume After an ATS Rejects It?

Yes, in some cases, but you should not count on it. A recruiter can manually search stored resumes by keyword or recognize you from another channel. Reaching out directly after a rejection, especially through LinkedIn, is a fair next step if you believe you are truly qualified.

What File Format Works Best for ATS Parsing?

A .docx file is the safest default unless the posting asks for a PDF. PDF support varies widely by vendor. Some parse it cleanly, while others strip formatting or miss whole sections, so .docx stays the more consistent choice.

Do Modern ATS Platforms Use AI to Screen Candidates?

Some do, but the practice is far from common. More new platforms now use AI for ranking and screening. Many legacy systems, and even some modern ones, still lean on keyword rules, so ask a specific vendor directly rather than guess.

What Is the Difference Between an ATS and an HRIS?

An ATS runs the hiring process, while an HRIS handles staff after they are hired. An applicant tracking system covers job postings, applications, and interviews. A human resources system then takes over payroll, benefits, and staff records once someone accepts a job.

Should I Pay for a Resume-Scoring Tool Before I Apply?

No, not before you have run the free manual checks yourself. A plain-text paste-test and a side-by-side keyword check against the posting catch most parsing and matching problems for free. A paid tool is worth considering only after those manual steps still leave you unsure.

Why Do Auto-Apply Bots Make ATS Problems Worse for Everyone?

They flood postings with weak applications, which pushes employers toward stricter filters. A posting that collects a thousand applications instead of a hundred still holds only a small share of qualified people. Employers tighten their screening rules in response, and that net then catches more real candidates too.

How Long Should a Small Business Wait Before Upgrading Its ATS?

There is no fixed timeline, since the real signal is your hiring volume and error rate, not your system's age. A company still filling roles through email and spreadsheets at 30 or more hires a year has a stronger upgrade signal than any set number of years in service.