An ATS (Applicant Tracking System) is the software most employers now use to screen CVs automatically before a recruiter ever sees them. In 2026, the question isn't whether your ATS resume will hit this filter — it's how to structure your CV so it gets through without losing its impact on the human reader who comes next.
This guide explains how an ATS actually works, what formatting and keyword rules to follow, and where the limits of "ATS-friendly" optimisation sit.
What is an ATS, and why has it become so widely used?
An ATS is software that collects every application a company receives, automatically pulls out the key information (contact details, work history, education, skills) and gives each CV a matching score against the job description. Applications below a certain score are never passed on to a human recruiter.
This shift is a direct response to a practical problem: the number of applications per role has grown sharply with online job boards and Easy Apply-style features, while recruiting teams haven't grown at the same pace. For many mid-sized and large employers, automating the first screening round has become an operational necessity rather than a preference.
In practice, most large UK employers and graduate schemes now run an ATS, using platforms like Workday, SAP SuccessFactors, Taleo, Greenhouse or Lever. Smaller businesses and SMEs use these tools far less consistently, particularly for speculative applications or local roles.
How does an applicant tracking system actually read a CV?
An ATS processes a CV in three stages: it extracts the raw text from the file, identifies standard sections (experience, education, skills), then calculates a relevance score by comparing the content against the job description's keywords.
Text extraction. The ATS "reads" your file to pull out the text. This is the stage most sensitive to formatting: a CV with columns, tables or graphic text boxes can lose or scramble information that's technically still on the page.
Section recognition. The software looks for standard headings. An overly creative section title ("My Journey", "What I Bring to the Table") risks not being recognised as "Work Experience" or "Skills".
Matching score calculation. The ATS compares the terms in your CV against the keywords set for the role. A widely cited Harvard Business School study found that automated screening systems filter out a significant share of applications before any human review — a pattern that hits candidates with non-linear or unconventional career paths particularly hard, since an algorithm struggles to assess context that a recruiter would pick up on instantly.
Newer ATS platforms include natural language processing that can recognise close synonyms ("project management" and "project delivery", for example). This reduces the risk of an unfair rejection, but doesn't remove it entirely — whoever configures the tool still sets a specific list of priority keywords.
What CV format actually passes ATS screening?
For an ATS, the safest format is a text-based PDF (with selectable text, not a scanned image) or a Word document (.docx), laid out in a single column, with no tables or images, in a standard font.
A heavily designed CV built in a general design tool can end up completely unreadable to an ATS, even when the candidate is a strong fit for the role. A few simple rules avoid that problem:
Single column only. ATS software reads left to right, top to bottom; a multi-column layout can scramble dates, job titles and duties during extraction.
Standard fonts. Arial, Calibri, Garamond or Times New Roman, at 10–12pt, are read without errors.
No tables or text boxes. Content placed inside these elements is often ignored or extracted incorrectly.
Contact details in the body of the document. Many ATS platforms ignore headers and footers entirely — a phone number placed only in the header may never be picked up.
No skill bars or rating graphics. A skill shown as five stars or a progress bar conveys nothing an algorithm can use — write "French: conversational" instead of relying on a symbol.
If you'd rather start from a ready-made structure, you can build your CV from a template designed to stay readable by both a piece of software and a recruiter.
How do you use keywords to optimise a CV for ATS?
To optimise a CV for an ATS, pull the exact keywords from the job description — skills, tools, job title — and work them in naturally across your headline, summary, skills section and experience, without forcing repetition.
Read the job description twice. Pick out the 5 to 8 terms that come up most often: required skills, named tools, specific certifications.
Include both the acronym and the spelled-out version. If the listing says "SEO", also write "search engine optimisation" at least once — not every ATS links the two automatically.
Place keywords where they carry weight. Your CV title and skills section are weighted most heavily by the algorithm; your summary and experience bullets matter, but slightly less.
Keep the language natural. A CV that repeats a term excessively (so-called "keyword stuffing") can be flagged as suspicious by more recent ATS platforms, which also assess how coherent the overall document reads.
Two things to avoid entirely: copying and pasting the whole job description into your CV, and hiding keywords in white text on a white background — a trick most current systems detect and penalise.
What mistakes get a CV rejected by an ATS?
The most common mistakes fall into two categories — formatting and content. These come up again and again:
Formatting mistakes
Multi-column layouts or tables
Key information placed in the header or footer
CV submitted as an image or scanned PDF, with no selectable text
Icons used instead of text (a phone icon with no number actually written out)
Content mistakes
Non-standard section headings the software doesn't recognise
Missing keywords from the job description
Missing or inconsistently formatted dates
A generic CV sent unchanged to every application
That last one is arguably the most costly: a CV that's never tailored to the role almost always scores lower on the match than one adjusted, even slightly, for each application.
Are AI-powered ATS platforms changing the game in 2026?
Yes — part of recruitment now relies on tools that go beyond simple keyword matching. These platforms try to assess overall fit between a candidate and a role, sometimes inferring skills that aren't explicitly stated based on the rest of the CV.
This shift has two practical consequences for job seekers:
Overall career coherence now matters more than keyword density alone. A well-built CV, where each role logically connects to the next, tends to score better with these tools than one that piles on technical terms with no clear thread.
These systems aren't neutral. Several studies have shown that algorithms trained on historical hiring data can reproduce existing bias — around gender, age, or the assumptions a name might trigger. In the UK, the Information Commissioner's Office (ICO) has published guidance on the use of AI in recruitment and its implications under data protection law, and the EU's AI Act, in force since 2024, classifies automated recruitment tools as "high risk" systems requiring transparency and human oversight for any employer operating across EU markets. None of this changes what you need to do as a candidate, but it explains why more employers are now documenting and auditing their screening tools.
The practical takeaway stays the same either way: keeping your CV clear and internally consistent protects you both against a weak ATS score and against the kind of bias these systems can carry.
Is an ATS-friendly CV enough to get you an interview?
No. Passing the ATS filter is a necessary condition, not a sufficient one: a CV with a strong match score that's vague or unconvincing will still get rejected by the human recruiter at the next stage.
An effective CV in 2026 has to do two jobs at once: be simple enough in structure for software to read correctly, and contain specific, measurable results that make a recruiter want to meet you. Neither replaces the other.
One more thing worth keeping in mind: an application passed on internally by someone in your network often bypasses the standard ATS filter entirely, landing directly with a recruiter. Building and using your network remains one of the more reliable ways to get a human set of eyes on your application quickly.
If you're building a job search strategy from scratch, our guide on how to get your first job after university covers the complementary steps: where to apply, cover letters, and interview preparation.
Key takeaways
An ATS screens CVs automatically before a human recruiter reads them; most large UK employers use one in 2026.
Best format: text-based PDF or .docx, single column, standard font, no tables or images.
Pull 5 to 8 keywords from the job description, including acronyms and their spelled-out forms, without forcing repetition.
A non-standard section heading or contact details in the header can be enough to lose information during extraction.
AI-powered ATS platforms also weigh overall career coherence, not just keyword density.
Passing the ATS filter doesn't guarantee an interview — the content still has to convince the person reading it next.
FAQ — Common questions about ATS and CVs
Do all employers use an ATS? No. Most large employers and graduate schemes use one, but smaller businesses and SMEs rely on these systems far less consistently, especially for speculative applications or local roles.
Can a visually designed CV still pass ATS screening? It mostly comes down to the structure of the file rather than the visual style: if the text stays selectable and the layout holds to a single column, with no information hidden in images or tables, a well-designed CV can still be read correctly.
Do you need two versions of your CV — one for the ATS and one for humans? Not usually. A CV that's simply structured, with standard sections and specific results, generally works for both readings at once.
Do CV keywords need to match the job description exactly? Ideally, yes, for the most important terms — especially on older ATS platforms that rely on exact matching. Using the job description's exact wording, plus a common synonym where you know one, remains the safest approach.
Can an ATS detect a CV written by AI? Some newer ATS platforms include modules that flag overly generic or uniform phrasing. A CV generated with no personalisation at all can score lower as a result — not for ethical reasons, but because the content lacks the specificity that reflects a real candidate's background.
External sources
Information Commissioner's Office (ICO) — guidance on AI and data protection in recruitment
GOV.UK — official guidance on employment rights and recruitment practices
