DEV Community: Hanzala Mehmood

What Duty-Heavy CVs Taught Us About Scoring Judgement, a Signal Almost Invisible to a Keyword Extractor

Hanzala Mehmood — Sun, 02 Aug 2026 13:00:01 +0000

TL;DR

Executive assistant CVs are dense with duties because the role is defined by duties, but the thing being hired is judgement under confidentiality, which rarely appears as a keyword.
When we built the scoring pipeline in CVPilot, duty-heavy CVs exposed a gap: high keyword coverage, near-zero signal about the decisions behind the tasks.
The fix was not a longer keyword list. It was scoring quantified outcomes and discretion cues, such as gatekeeping and prioritisation, on a separate axis from raw keyword match.

We built CVPilot to score CVs the way a UK applicant tracking system does, so we spend a lot of time reading how a parser sees a document. Executive assistant CVs broke our assumptions early. They were keyword-rich and still read as thin, and it took a while to work out why. The role is defined by duties, so the CV lists duties, but the real hiring signal is judgement under confidentiality, and that barely survives tokenisation.

Here is the uncomfortable truth about the executive assistant CV. The job is defined by duties, so the CV fills up with duties, yet the thing an employer is actually hiring is judgement under confidentiality. Judgement is the one thing nobody thinks to write down.

In 2026, that gap matters more than ever. AI now drafts the emails, schedules the meetings and books the flights. What it cannot do is decide which of three warring directors gets the 4pm slot, or whether to interrupt a board dinner with news that cannot wait.

This guide covers the executive assistant CV keywords that UK applicant tracking systems screen for, the tools recruiters expect to see, and how to turn a list of tasks into evidence of decisions. It sits alongside our complete ATS-friendly CV guide for the mechanics of formatting.

What the ATS is actually searching for

Most UK employers hiring at PA and EA level run CVs through an applicant tracking system before a human sees them. The system scores your document against the job description, and missing the exact phrasing costs you the shortlist, regardless of how good you are.

The vocabulary is remarkably specific. Recruiters do not search for "organised admin professional". They search for diary management, inbox management, gatekeeping and board papers, because those are the words in the brief they were given.

Match the language of the sector too. A legal PA CV needs governance and NDA fluency, while an EA in tech leans on stakeholder management and tools like Notion and Asana. Read three live job adverts for your target role and mirror the terms that repeat.

Key Takeaway: The ATS rewards the exact phrasing in the job advert, so mine live UK listings for the terms that keep appearing and use those, not your own synonyms.

Keywords by role level

Not every keyword suits every stage. A team assistant claiming board reporting looks inflated, and a senior EA listing only "answering phones" looks underpowered. The table below groups the terms by where they carry weight, so you can pitch your CV at the right altitude.

Role level	Core ATS keywords
Team assistant / office manager	Diary management, inbox management, travel booking, expenses, event management, meeting rooms, office management, facilities, supplier liaison, onboarding support
PA (personal assistant)	Complex calendar management, gatekeeping, minute taking, travel and itinerary management, expenses reconciliation, stakeholder management, confidentiality, document management
EA (executive assistant)	Executive support at C-suite level, board papers, board reporting, governance, NDA handling, visa coordination, budget management, project delivery, cross-functional coordination
Senior EA / chief of staff	Board and committee coordination, governance and compliance, stakeholder navigation, strategic prioritisation, headcount and team management, budget ownership, business planning, confidential decision support

Layer these naturally into your role descriptions rather than dumping them in a skills block. An ATS reads context, and a recruiter reading behind it wants to see the keyword attached to a real achievement.

Key Takeaway: Choose keywords that match your seniority, then embed them inside achievements so both the software and the human reading after it are convinced.

The tools recruiters expect to see

Software fluency is a hard filter. If the advert names a system and your CV does not, an automated screen can drop you before a person ever reads a word. Name your tools explicitly and by their proper titles.

The core stack is Outlook and Microsoft 365, plus Slack for internal comms. For expenses and finance, list whichever you have used: Concur, Expensify or Coupa. Do not write "expenses software" when you can write the product name the recruiter searched for.

Beyond that, name the workflow and document tools that signal a modern operation: Notion and Asana for project tracking, Salesforce for CRM exposure, Egnyte for secure file management, and DocuSign for contracts. For board-level roles, board portals such as Diligent are a strong differentiator that few candidates think to mention.

Key Takeaway: List every relevant tool by its exact product name, because "expenses software" fails the keyword match that "Concur" passes.

How to quantify EA work

The most common EA CV mistake is describing scope without size. "Managed the CEO's diary" tells a recruiter nothing about whether you handled one calm director or three founders in constant conflict. Numbers convert responsibility into credibility.

You can quantify almost everything. Count the executives supported, the volume of your calendar, the travel booked and the cost saved, the board meetings coordinated, the budget owned, the size of the admin team you managed, and the projects you delivered.

If you have never tracked these figures, reconstruct honest estimates. "Coordinated travel for a 12-person leadership team" is far stronger than "arranged travel", and a defensible estimate beats a vague verb every time.

Key Takeaway: Attach a number to every core duty, because scale is what tells a recruiter whether you supported a quiet office or a demanding C-suite.

Before and after: bullets that earn interviews

Here is the difference between a duty and an achievement, shown with the kind of real numbers a UK EA can defend in an interview.

Before (duty)	After (evidence with numbers)
Managed the CEO's diary	Owned complex calendar management for 3 C-suite executives, resolving 40+ scheduling conflicts a week across 4 time zones
Booked travel and hotels	Coordinated international travel and itineraries for a 15-person leadership team, negotiating rates that cut annual travel spend by £22k
Took minutes in meetings	Produced board papers and minutes for 8 quarterly board meetings, circulating actions within 24 hours to 11 directors
Handled expenses	Reconciled expenses across Concur for 20 staff, reducing late submissions by 60% and clearing a £30k backlog in one quarter
Organised company events	Delivered 6 all-hands events and 2 offsites for 120 staff, managing a £45k budget with zero overspend

Notice what the "after" column does. It keeps the ATS keyword, adds a number, and quietly signals the judgement behind the task. Negotiating rates and prioritising conflicts are decisions, not chores.

Key Takeaway: Rewrite every duty into a keyword plus a number plus a hint of the decision you made, and your CV starts reading like a shortlist candidate.

The 2026 shift: evidence judgement, not typing speed

For years EA CVs sold speed and accuracy. In 2026 that framing is quietly obsolete, because AI drafts, schedules and files faster than any human. Selling your typing speed now signals that you have not noticed the change.

What employers screen for instead is the human layer: discretion, stakeholder navigation and the confidence to make a call when the principal is unreachable. These are the parts of the job an automated tool cannot touch.

So evidence the decisions, not the tasks. "Acted as gatekeeper for the CEO, filtering 200+ weekly requests and protecting focus time" shows judgement. Show where you exercised discretion, handled something confidential, or steered a difficult stakeholder to a better outcome.

Key Takeaway: The rare and valuable skill in 2026 is judgement under confidentiality, so give it the space on your CV that typing speed used to occupy.

EA, PA or office manager: label yourself correctly

These titles are not interchangeable on a UK CV, and using the wrong one can misfile you in the ATS. A clear title tells the recruiter exactly what you support before they read a line.

A personal assistant typically supports one principal on both business and personal matters. An executive assistant supports leadership at C-suite level with a heavier weighting toward business, governance and strategic coordination. An office manager runs the workplace itself: facilities, suppliers, budgets and often a small team.

The progression runs from team assistant to PA to EA to senior EA, and for some into chief of staff or business manager. Name your current rung accurately, then let your achievements argue for the next one.

Key Takeaway: Pick the job title that genuinely matches your remit, because an inflated or vague label confuses both the software and the hiring manager.

Putting it together

Start with three live UK job adverts for your target role and highlight every repeated keyword. Map those against the tables above, then rewrite your bullets so each one carries a keyword, a number and a trace of judgement.

Run the finished CV through an ATS checker to confirm it reads cleanly and scores well against the description. CVPilot flags the missing keywords and the duties that still read as tasks rather than decisions, so you can fix them before you apply.

The candidates who win EA roles in 2026 are not the fastest typists. They are the ones who prove, in specific and measurable terms, that they can be trusted with the room when it matters. Let CVPilot help you show it.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/executive-assistant-cv-keywords-ats?utm_source=devto&utm_medium=organic&utm_campaign=executive-assistant-cv

If you have built text scoring, how do you model a signal the author almost never states out loud?

Full guide: https://cvpilot.pro/blog/executive-assistant-cv-keywords-ats

What building ATS keyword extraction for regulated sectors taught us: score evidence, not sentiment

Hanzala Mehmood — Sat, 01 Aug 2026 13:00:01 +0000

TL;DR

High-volume regulated hiring, and UK adult social care is a textbook case, runs on keyword-matching ATS software, so extraction quality decides who a recruiter ever sees.
The hard part is not parsing the document, it is scoring evidence (framework plus metric) above sentiment words like "compassionate" and "hardworking" that carry no signal.
We built CVPilot to flag the exact regulated terms a draft is missing, and the care sector made the sentiment-versus-evidence problem impossible to ignore.

When we started building CVPilot, an ATS optimisation tool, I assumed the interesting work would be document parsing. It was not. The interesting work was deciding what a keyword is actually worth, and no sector made that clearer than UK adult social care. A care CV is dense with genuine skill, but the terms that carry regulatory signal, "MAR charts", "mental capacity", "safeguarding", sit right next to terms that carry none, "caring", "hardworking", and a naive matcher weights them the same.

You finish a twelve-hour shift, administer a full medication round, de-escalate a resident in distress, update three care plans, and log an incident before handover. Then you go home and write a CV that says you are "caring and hardworking". The gap between what care workers do and what care workers write down is the single biggest reason good applications get rejected before a human ever reads them.

Large care providers and agencies receive hundreds of applications per vacancy. To cope, they run applications through an Applicant Tracking System, software that scans your CV for specific keywords before a recruiter sees it. If the words the software is looking for are not on your CV, your application is filtered out automatically.

This guide maps the exact vocabulary UK care roles are screened against in 2026, grouped by role level, with a before and after table showing how to turn everyday tasks into evidence that scores.

Why care applications get screened by software at all

Care is one of the highest-volume recruiting sectors in the country. A single care home group might advertise the same care assistant role across forty sites, and a domiciliary agency can carry a permanent vacancy simply because demand never stops. High volume plus large employers equals automation, and automation means keyword screening.

The systems are not clever. They match text. When a provider needs someone with the Care Certificate and enhanced DBS clearance, the software searches for those literal phrases. A candidate who wrote "fully trained and checked" is invisible, even if they hold both.

This is where most care CVs quietly fail. The work is done to a high standard, but it is described in warm, general language that no scanner can score.

Key Takeaway: Care roles are screened because they are high volume and employer-heavy. The scanner matches exact phrases, so vague competence loses to precise vocabulary every time.

The contrarian truth: compassion does not score

Every care worker is taught to lead with compassion, and rightly so, because it is the heart of the job. But compassion is unscoreable, and the ATS cannot measure kindness. When your opening line is "a compassionate and caring individual", you have spent your most valuable CV space on words the software ignores and every other applicant also used.

The CVs that get through evidence the regulated, technical side of care. This is the part carers perform every single day and rarely think to write down: medication administration and MAR charts, safeguarding, mental capacity assessments, moving and handling, person-centred care planning, risk assessment.

None of that makes you less caring. It proves you are a trained professional operating inside a regulated framework, and that is precisely what the scanner, and the CQC-inspected employer behind it, needs to see. You can be warm in the interview. On the page, be technical.

Key Takeaway: Compassion belongs in the interview. On the CV, lead with the regulated skills you already use daily, because those are the terms the software is built to find.

Care assistant CV keywords by role level

The keywords a scanner searches for change as you progress. A team leader application filtered for "person-centred care" alone will miss you if the real requirement is supervision and rota management. Match your vocabulary to the level you are applying for, not the level you are leaving.

Role level	Qualifications and frameworks	Skills and regulatory terms
Care Assistant / Support Worker	Care Certificate, NVQ / QCF Level 2 Health and Social Care, enhanced DBS, moving and handling, basic life support, first aid	Person-centred care, personal care, care plans, safeguarding adults, dignity and respect, MAR charts, incident reporting, activities of daily living
Senior Carer / Healthcare Assistant	RQF Level 3 Diploma in Adult Care, medication administration, PBS, dementia care, end of life care, tissue viability	Medication rounds, care plan authoring, risk assessments, mental capacity (MCA), DoLS, duty of candour, supervising junior staff, CQC compliance
Team Leader / Deputy Manager	Level 5 Diploma in Leadership for Health and Social Care, Skills for Care registered manager pathway, safeguarding level 3	Rota management, supervisions and appraisals, audits, care plan reviews, complaints resolution, CQC inspections, staff mentoring and retention, budget oversight
Setting-specific (all levels)	Learning disabilities training, complex needs, PEG feeding, catheter care, epilepsy awareness	Residential, nursing, domiciliary, supported living, respite, live-in care, complex and continuing care

Use the exact spelling the sector uses. Write "person-centred", not "patient-focused". Write "safeguarding", not "protecting vulnerable people". The scanner is matching the job advert, so mirror the language of the advert you are answering. A quick pass through a tool like CVPilot will flag which required terms are missing from your draft.

Key Takeaway: Vocabulary is level-specific. Feed the scanner the qualifications and regulatory terms that match the advert in front of you, spelled the way the care sector spells them.

Quantify care work honestly

Numbers survive automated screening and lodge in a recruiter's memory. The worry is that care feels unquantifiable, because you cannot put a figure on holding someone's hand. You do not need to. Quantify the operational reality around the care, not the emotion inside it.

Count what you already track. Residents supported per shift, care plans authored or reviewed, medication rounds completed, years without a medication error, staff mentored, complaints resolved, client retention on your domiciliary round, your contribution to a "Good" or "Outstanding" CQC outcome.

Here is how that looks in practice, taking bullets straight off a real care CV and rebuilding them.

Before (invisible to the scanner)	After (keyword-rich and quantified)
Looked after residents and helped with daily needs	Delivered person-centred personal care to 14 residents per shift in a 40-bed residential setting, supporting all activities of daily living
Gave out medication safely	Completed medication rounds for 20 residents twice daily, maintaining accurate MAR charts with zero medication errors across 3 years
Wrote up notes and helped with paperwork	Authored and reviewed 30+ care plans and risk assessments, applying mental capacity (MCA) principles and duty of candour reporting
Helped train new staff	Mentored 8 new care assistants through Care Certificate induction, improving team retention and shift-cover reliability
Care home passed its inspection	Contributed to a "Good" CQC inspection outcome through consistent safeguarding practice and accurate incident reporting

Every "after" bullet still describes the same shift. The difference is that it names the framework, states the number, and uses the phrase the scanner is hunting for. Nothing is exaggerated, and every figure is one you could defend in an interview.

Key Takeaway: Quantify the operational facts around the care, not the feeling. Residents supported, rounds completed, plans reviewed, errors avoided. Honest numbers beat warm adjectives.

Structure your CV so both the scanner and the recruiter can read it

Keywords only help if the software can parse your document. Care CVs are often rejected on formatting alone, before content is even assessed. Save as a Word document or a simple PDF, use standard headings, and never hide skills inside tables, headers, or text boxes that scanners cannot read.

Put a short professional summary at the top with your headline qualifications, the Care Certificate, your Diploma level, enhanced DBS. Follow with a clear skills section, then experience in reverse date order. Our complete ATS-friendly CV guide covers the layout rules in full.

Show progression clearly. A pathway from care assistant to senior carer to team leader to deputy manager tells an employer you grow within a service and stay, and staff retention is gold to a sector fighting constant turnover. Name each title as it appeared, because "senior carer" and "team leader" are themselves keywords a scanner may be filtering for.

Key Takeaway: A perfectly worded CV still fails if the software cannot read it. Use plain formatting, standard headings, and real job titles so both machine and manager can follow your progression.

Put it into practice

Care work is skilled, regulated, and relentless, and your CV should read that way. Lead with the technical and regulatory vocabulary you already live by, quantify the operational side honestly, and format the document so the scanner can actually read it.

Do that, and you clear the filter that rejects most applicants before a human ever weighs your experience. The compassion that makes you brilliant at the job will speak for itself the moment you are in the room. Run your draft through CVPilot to check which sector keywords you are missing.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/care-assistant-cv-keywords-ats?utm_source=devto&utm_medium=organic&utm_campaign=care-assistant-cv

If you were scoring CV evidence over sentiment, how would you weight a quantified metric against a matched regulatory term?

Full guide: https://cvpilot.pro/blog/care-assistant-cv-keywords-ats

Structured credentials in unstructured text: parsing CVs for the trades

Hanzala Mehmood — Fri, 31 Jul 2026 13:00:01 +0000

TL;DR

Trade CVs are dense with high-signal structured tokens (18th Edition, BS 7671, AM2, ECS, NVQ Level 3) that a naive keyword matcher finds trivially and then overweights.
The hard problem is the opposite direction: the qualification is easy to extract, the evidence of applying it is buried in free text or absent entirely.
Scoring credential presence without evidence of use produces false positives at the top of the pipeline and rejects genuinely experienced candidates further down.

Building CVPilot's keyword extraction meant handling sectors well outside the tech CVs everyone benchmarks against. The trades turned out to be the interesting case. An electrician's CV contains a cluster of standardised credential tokens that any regex will find, alongside a description of a decade of work that frequently contains no extractable evidence at all.

That asymmetry is a scoring problem, and it is one that penalises the candidate rather than the parser.

You have got the 18th Edition, the ECS Gold Card, and twenty years on the tools across domestic, commercial, and industrial sites. So why does the application vanish the moment you hit submit? The honest answer is that a piece of software read your CV before a human ever did, and it was not impressed by your card alone.

High-volume trades hiring runs through an Applicant Tracking System, the same as any office job. When an agency posts one maintenance role and gets 180 applications by Friday, nobody reads all 180. The ATS ranks them by keyword match first, and the recruiter opens the top handful.

This guide is written for the person holding the card, not the person selling the software. We will cover the exact qualifications, compliance terms, and sector vocabulary the filters search for, how to quantify work that lives on a site rather than a spreadsheet, and the one habit that costs time-served sparks the interview they had already earned.

Why a trades CV gets screened at all

There is a myth on site that CVs are for office workers and that a good spark gets work through the phone and the reputation. That still holds for a lot of direct hires. The problem is that agency and framework applications almost never reach a human cold.

Agencies use a database. When a contract manager needs three approved electricians for a data centre fit-out on Monday, the consultant types the requirements into the system and pulls matching profiles. If your CV never used the phrase "approved electrician" or "three-phase", you are not in the results, and your reputation cannot save a record the software never surfaced.

Direct employer applications differ slightly. A facilities manager or an M&E contractor may read more carefully, but the large ones still use ATS software to sift the first wave. Either way, the machine is the first gate and keywords are the key.

Key Takeaway: Your trade reputation opens doors in person, but on agency and framework applications a database decides who gets seen first, and it only understands the words on the page.

The contrarian bit: your card is not the evidence

Here is the mistake I see on almost every electrician CV that lands in my inbox. The candidate lists the 18th Edition, the 2391, the NVQ Level 3, the AM2, the ECS card, and then more or less stops. The thinking is reasonable: the qualifications should speak for themselves.

They do not. The ATS finds the card, ticks the box, and then goes looking for evidence that you did the work. When the bullets underneath say "carried out various electrical duties" and nothing more, there is nothing to match against the job spec, and your file scores low.

So a fully time-served electrician with a Gold Card loses to someone less experienced who simply wrote better bullets. The card gets you past the qualification filter. The described work is what actually ranks you, and that is the half most people skip.

Key Takeaway: List the card, then prove it. Qualifications clear the first filter; quantified, keyword-rich bullets are what push you to the top of the recruiter's list.

Qualifications and cards the ATS searches for

Recruiters paste specific terms into the search box, and they use both the formal name and the shorthand. Your CV should carry both so it matches either query. Write "18th Edition (BS 7671)" in full at least once, because some systems search the standard number and some search the plain-English name.

The essential set for most UK electrical roles includes the 18th Edition (BS 7671), City & Guilds 2365 and 2357, the 2391 for inspection and testing, the 2392 for initial verification, NVQ Level 3, the AM2, and your ECS card with its JIB grading. If you hold Approved Electrician status or a Gold Card, name it explicitly rather than leaving the reader to infer your grade from the card colour.

Do not bury these in a paragraph. A clear "Qualifications" heading with each item on its own line reads cleanly for a human and parses reliably for the software. A CV built on a proper complete ATS-friendly CV guide structure gives every card its own parseable line.

Key Takeaway: Spell out both the standard number and the common name for each qualification, and give the card its exact grade so the search matches you either way.

Compliance, testing, and sector vocabulary

Compliance language is where trades CVs win or lose against the job spec, because these are the exact phrases a facilities or M&E advert repeats. If the role is inspection and testing, the words EICR, periodic inspection and testing, and initial verification need to appear in your work history, not just your skills list.

Testing and certification terms worth carrying, where true, include EICR, minor works certificate, Part P, RAMS, permit to work, LOLER, PAT testing, and the scheme names NICEIC and NAPIT plus SafeContractor. Never claim a scheme registration you do not hold; recruiters verify these fast.

Sector vocabulary tells the system which environment you know. Containment, tray and basket, SWA, three-phase, distribution boards, and fault finding signal commercial and industrial competence. BMS, HV/LV, and PLC signal maintenance and controls, while solar PV, EV charge point installation (OZEV), and heat pumps signal the renewables work now driving a large share of new UK vacancies.

Key Takeaway: Match the advert's compliance and sector words inside your actual experience, and only ever claim schemes and registrations you genuinely hold.

Keywords by role level

Not every term belongs on every CV. The words that get an improver shortlisted differ from the words a supervisor needs. Use the table to pick the language that matches the level you are applying for, and lift the exact phrasing from the advert wherever it fits your real experience.

Role level	Keywords the ATS searches for
Apprentice / Improver	City & Guilds 2365, NVQ Level 3 (working towards), ECS labourer/apprentice card, first fix, second fix, containment, cable pulling, under supervision
Electrician (domestic)	18th Edition, Part P, consumer unit upgrades, minor works certificate, EICR, fault finding, rewires, NICEIC/NAPIT
Electrician (commercial / industrial)	Three-phase, distribution boards, SWA, tray and basket, containment, HV/LV, motor control, RAMS, permit to work
Maintenance electrician	Reactive and planned maintenance, PPM, fault finding, BMS, PLC, PAT testing, LOLER, downtime reduction, first-time fix
Approved Electrician	2391 inspection and testing, initial verification, EICR, AM2, JIB Approved grading, Gold Card, sign-off authority
Supervisor / Project Manager	Team of X electricians, RAMS authoring, site management, CSCS, apprentice supervision, contract value, zero-RIDDOR record, programme delivery

Key Takeaway: Target the keyword set for the exact level you want, not a generic pile of every term, and mirror the advert's wording where it is honestly true of you.

How to quantify work that lives on a site

Office workers quantify with revenue and budgets. Trades work has numbers too, they just sit in different places, and a number in a bullet beats an adjective every time. The value of the contracts you worked, the count of EICRs you completed, your first-time fix rate, the downtime you cut, the sites you ran, and the apprentices you supervised are all measurable.

Compare the two versions of the same real work below. The left column is what most electricians write. The right column is the same job described so the ATS and the recruiter can both see the scale.

Before (vague)	After (quantified)
Carried out testing on various sites	Completed 340+ EICRs across 22 commercial sites, achieving a 100% on-time certification record
Did fault finding and repairs	Raised first-time fix rate from 71% to 89% on a 40,000 sq ft manufacturing site, cutting production downtime by 30%
Worked on a big install project	Delivered second fix and containment on a £2.4m office fit-out, wiring 18 distribution boards to programme
Supervised some apprentices	Supervised 4 apprentices and a 6-strong install team, maintaining a zero-RIDDOR record across an 18-month contract
Responsible for maintenance	Managed PPM schedule for 3 industrial units, reducing reactive callouts by 25% year on year

Notice that the right-hand versions carry keywords and numbers together. "EICRs", "containment", "distribution boards", and "PPM" match the search, while the figures give the recruiter a reason to call. The number is what turns a keyword match into an interview.

Key Takeaway: Every bullet should pair a searchable term with a real figure, whether that is contracts values, EICR counts, fix rates, or the size of the team you led.

Progression keywords from apprentice to project manager

Your CV should signal where you are on the ladder, because recruiters filter by seniority as well as trade. An apprentice CV leading with "sign-off authority" looks wrong, and a supervisor CV that only mentions cable pulling undersells the level you have actually reached.

The natural progression reads apprentice, improver, electrician, approved electrician, then supervisor or project manager. Each rung has its own vocabulary: improvers add "working towards NVQ Level 3", approved electricians add "2391 inspection and testing" and "sign-off authority", and managers add "RAMS authoring", "programme delivery", and "contract value".

If you are mid-jump, name the target level in your profile summary while your history proves you are ready. Writing "approved electrician with inspection and testing sign-off" at the top tells both the software and the reader exactly which shortlist you belong on. A tool like CVPilot flags which of these level-specific terms your draft is missing before you send it.

Key Takeaway: Match your keywords to the rung you are aiming for, and state the target level in your summary so the filter sorts you into the right band.

Putting it together

Start with a tight profile summary naming your grade, your key qualifications, and your sector. Follow with a clean qualifications block, then a work history where every role carries compliance terms and quantified bullets. Keywords earn the read, numbers earn the call.

Before you apply, read the advert twice and note the exact phrases it repeats, then make sure those honest matches appear in your experience. Run the finished CV through CVPilot to see your ATS score and the specific electrical terms you are missing, then fix the gaps and send it with confidence.

Key Takeaway: A trades CV that pairs the right card language with real numbers, tuned to each advert, beats a more experienced spark who let the qualifications speak for themselves.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/electrician-cv-keywords-ats?utm_source=devto&utm_medium=organic&utm_campaign=electrician-cv

If you have built extraction over credential-heavy documents, how did you weight the certificate against evidence that it was ever used?

Full guide: https://cvpilot.pro/blog/electrician-cv-keywords-ats

How AI Chat Deflection Rewired Customer Service Work, and What It Does to CV Scoring

Hanzala Mehmood — Thu, 30 Jul 2026 13:00:02 +0000

TL;DR

Chat deflection routes the easy, high-volume contacts to bots, leaving humans the hard tail of escalations and edge cases.
That inverts what a customer service CV should optimise for: complexity signals rather than raw throughput counts.
ATS keyword matching still runs on literal string overlap, so mirror the advert's metrics (FCR, CSAT, escalation rate) and attach a number to each.

If you have ever shipped a support chatbot, you have watched the ticket mix change in real time. The bot absorbs the password resets and order-status pings, and what reaches a human is the residual hard tail: multi-issue complaints, vulnerable customers, judgement calls. I spend most of my time building CVPilot, an ATS-scoring tool, and the same shift shows up in how these CVs now get ranked. The volume metrics that used to score well are increasingly describing work that automation already covers.

You have handled thousands of tricky customers, calmed complaints that could have gone very wrong, and hit every target on the wallboard. Then your CV lands on a desk and gets 30 seconds of attention, if you are lucky. The problem is rarely your experience. It is that your CV speaks in plain words while the hiring system reads in keywords.

Most customer service adverts in the UK now pass through an Applicant Tracking System before a human sees them. That software scans for the exact language the job description uses. If the advert asks for "first contact resolution" and your CV says "sorted problems quickly", the match is weaker than it should be.

This guide gives you the real vocabulary of UK customer service roles, how to attach numbers to it, and how the rise of AI chat has changed what employers actually screen for in 2026.

Why keywords decide whether a human ever reads your CV

An ATS does not understand that "de-escalated angry callers" and "complaint resolution" describe the same skill. It matches strings of text, so the closer your wording is to the advert, the higher you rank in the recruiter's shortlist.

This is not about stuffing your CV with jargon. It is about mirroring the language of the specific role you want. Read the advert, list the terms it repeats, and make sure those exact phrases appear naturally in your experience.

Contact centre and service desk teams are a huge, underserved group of jobseekers. That means a CV that speaks the sector's language properly stands out fast, because so few do it well. For the full technical picture, our complete ATS-friendly CV guide covers formatting rules that stop your file being rejected before the keywords even count.

Key Takeaway: The ATS ranks you on how closely your words match the advert, so mirror the exact phrasing of the role instead of describing it in your own terms.

The keywords UK customer service adverts actually use

Sector language falls into a few groups: the metrics you are measured on, the tools you work in, and the quality behaviours managers care about. The strongest CVs weave all three together rather than listing them in a skills box.

Below is a table grouped by role level. Use it as a checklist against the advert in front of you, not as a bucket to empty into your CV wholesale.

Role level	Metrics and behaviours	Tools and systems
Advisor / Contact centre agent	First contact resolution (FCR), CSAT, average handling time (AHT), adherence, quality assurance (QA) scores, call wrap-up, escalation	CRM, Zendesk, Freshdesk, telephony / ACD, knowledge base
Complaints / Resolution handler	Complaint resolution, root cause analysis, regulatory timeframes, goodwill decisions, de-escalation, FCA complaint handling	CRM case management, Salesforce Service Cloud, audit trail logging
Senior advisor / Subject matter expert	Escalation handling, coaching, call calibration, QA framework, omnichannel support (voice, email, chat, social)	Zendesk, Salesforce Service Cloud, workforce management (WFM)
Team leader / Team manager	SLA management, retention, churn reduction, adherence, shrinkage, one-to-ones, performance management, CSAT and NPS ownership	WFM tools, dashboards, real-time reporting, Salesforce
Customer experience / CX lead	NPS, customer effort score (CES), journey mapping, voice of customer, continuous improvement, stakeholder management	Analytics platforms, survey tools, CRM reporting suites

Notice how the tools stay fairly constant while the metrics shift as you climb. An advisor owns their own FCR and CSAT; a team leader owns the team's SLA, retention and churn. Match the level, not just the sector.

Key Takeaway: Group your keywords into metrics, tools and quality behaviours, then pick the ones that match your target role level rather than pasting the whole list.

Numbers turn a duty into a result

Recruiters and the ATS both reward specifics. "Responsible for customer queries" tells them nothing; "resolved 92% of queries at first contact" tells them everything. Every strong customer service bullet pairs a keyword with a number.

You do not need perfect data. Reasonable, honest estimates from your own stats or team reports are fine. A CSAT score, an FCR percentage, calls per shift, an escalation rate or a retention save figure will each lift a bullet from vague to credible.

Here is how ordinary duties become measurable achievements.

Before (weak)	After (quantified)
Answered customer calls and emails	Handled 60+ omnichannel contacts per shift while maintaining a 94% CSAT score
Dealt with complaints	Resolved 85% of escalated complaints within SLA, cutting repeat contacts by 18%
Helped keep customers happy	Achieved 90% first contact resolution, above the team average of 82%
Tried to keep customers from leaving	Retained 40+ at-risk accounts per month, reducing churn on my portfolio by 12%
Supported new starters	Coached 5 new advisors to QA pass rates of 95% within their first month

The "after" column works because it names the metric the sector already tracks and attaches a figure to it. That is exactly the pairing an ATS scores highly and a manager trusts. When you rework your own bullets, running them through CVPilot shows which keywords are landing and which gaps remain.

Key Takeaway: Pair every keyword with an honest number, because a metric plus a figure is what both the software and the hiring manager are looking for.

Progression: how an advisor CV becomes a team leader CV

Moving up is not about claiming grander duties. It is about shifting the subject of your achievements from "me" to "the team". An advisor evidences personal metrics; a team leader evidences influence over other people's metrics.

If you are targeting your first team leader role, look for the moments you already led without the title. Buddying new starters, running side-by-sides, owning the escalation queue, covering for the manager. Frame those as coaching, QA calibration and SLA ownership, which are the phrases team leader adverts repeat.

Add the numbers that leaders are judged on: team CSAT, adherence, shrinkage and retention. A line like "improved team adherence from 84% to 91% over two quarters" signals readiness far better than "helped the team do well".

Key Takeaway: To move from advisor to team leader, rewrite your achievements around team metrics like SLA, adherence and retention rather than your personal call stats.

Customer service is not customer success

These two titles get mixed up constantly, and using the wrong keywords sends your CV to the wrong shortlist. Customer service is largely reactive: you resolve issues, handle complaints and answer contacts as they arrive.

Customer success is proactive and commercial: you own retention, drive adoption, reduce churn and grow account value over time. If you are applying for customer success, your CV needs renewal rates, expansion revenue and onboarding metrics, not average handling time.

Plenty of service advisors have genuine success experience hidden in their history, especially retention saves and account growth. If that is you, pull those achievements forward and use the success vocabulary deliberately. Match your keywords to the job title on the advert, because the ATS was configured for that exact role.

Key Takeaway: Service is reactive resolution and success is proactive retention, so choose your keywords to fit the exact title on the advert.

What AI chat deflection changed in 2026

Here is the shift most jobseekers have missed. AI chatbots now deflect the easy, repetitive contacts: password resets, order tracking, opening hours. That means "handled 100 calls a day" is now a weak claim, because the bot handles the easy volume and the human handles the hard cases.

Employers in 2026 screen for evidence of complexity, not throughput. They want proof you resolved the cases the bot could not: the multi-issue complaint, the vulnerable customer, the angry escalation that needed judgement. Volume tells them nothing; difficulty tells them you are worth a salary.

So rewrite your headline numbers accordingly. Instead of raw contact counts, lead with escalation resolution rates, complex case handling, vulnerable customer support and de-escalation outcomes. Show the value you add above what automation already covers, and you become the candidate the sector genuinely needs.

This is where a quick keyword check pays off, and CVPilot flags whether your CV still leans on volume language that no longer impresses.

Key Takeaway: With AI handling easy volume, lead with evidence of complex, escalated and judgement-based cases rather than raw contact counts.

Putting it together

Start with the advert, list its repeated terms, and match your experience to them level by level. Pair every keyword with an honest number, frame complexity over volume, and pick service or success language to fit the title.

Do that and your CV clears the software, reads well to the manager, and reflects the real skill you have built on the phones and the queues.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/customer-service-cv-keywords-ats?utm_source=devto&utm_medium=organic&utm_campaign=customer-service-cv

If you have instrumented a deflection funnel, what share of contacts now reach a human, and has that mix changed what you would screen for in a support hire?

Full guide: https://cvpilot.pro/blog/customer-service-cv-keywords-ats

What Building ATS Keyword Extraction for High-Volume Retail Roles Taught Us About Parsing and Scoring

Hanzala Mehmood — Wed, 29 Jul 2026 13:00:02 +0000

TL;DR

Retail vocabulary (shrinkage, ATV, planogram, footfall conversion) sits nowhere near a standard tech skills taxonomy, so generic extractors miss the terms that actually matter
Scoring has to reward advert-to-CV term overlap and role-level weighting without punishing plain, honest phrasing, which pushes you towards synonym maps rather than exact matching
The highest-volume inputs are the hardest, since a single advert can pull hundreds of CVs overnight, so the parser has to be fast, layout-tolerant and cheap per document

When we built CVPilot, our early test corpus skewed towards software CVs, because that is what we had to hand. Retail broke our assumptions almost immediately. The roles are non-technical but intensely measured, and the vocabulary lives a long way from any standard skills taxonomy. This is a short write-up of what parsing and scoring high-volume, domain-specific CVs taught us, using retail as the worked example.

You have run a till on a Saturday when the queue snaked past the fitting rooms. You have hit a footfall target during a heatwave when nobody wanted to shop. Retail is one of the most measured jobs in the country, yet most retail CVs read like a list of chores rather than a record of results.

Here is the part that costs people interviews. Retail CVs get rejected for lacking numbers, when the numbers already exist on every shift report you have ever seen. Your KPIs are tracked hourly. Nobody puts them on the page.

This guide fixes that. It gives you the exact keywords UK retail adverts use, how to quantify shop-floor work, and how the applicant tracking system reads your application when it is one of four hundred that landed overnight.

Why retail CVs get filtered before a human ever reads them

Most large retailers and their agencies use an applicant tracking system, or ATS, to sort applications. The software scans your CV for words that match the job advert, then ranks you against everyone else who applied.

High-volume retail roles attract enormous application counts. A single Christmas temp advert can pull hundreds of CVs in a weekend, so the filter is aggressive by necessity. If your CV does not echo the language of the advert, it can be parked before a recruiter opens it.

The good news is that this is a language problem, not a worth problem. You already do the work. You simply need to name it the way the system expects, which you can check for free with CVPilot before you apply.

Key Takeaway: The ATS is not judging your character. It is matching words, so the fix is to speak its language while telling the truth about what you did.

The retail keywords UK adverts actually use

Open five retail job adverts and the same vocabulary appears again and again. These are the terms the ATS is primed to find, because the person who wrote the advert used them first.

Read this list and tick off the ones that describe your real experience. Only claim what you have genuinely done, because a good recruiter will test it at interview.

Shrinkage and stock loss: the shortfall between recorded stock and actual stock, whether through theft, damage or admin error.
Loss prevention: the tags, checks and habits that keep shrinkage down.
Footfall: the number of people entering the store, the raw traffic your team converts.
Conversion rate: the share of that footfall who actually buy.
Average transaction value, or ATV: the typical spend per sale, lifted by upselling and add-ons.
KPI: the key performance indicators your store is measured on weekly.
Planogram: the layout map that dictates where each product sits on the shelf.
Visual merchandising: the displays and windows that pull shoppers in and guide them around.
Cash reconciliation: balancing the till and safe against recorded sales at close.
Rota: the staff schedule, which supervisors and managers own.
Seasonal peak: Christmas, sales events and back to school, the trading periods that make or break the year.
Click and collect: the online order fulfilled or handed over in store.
Mystery shopper: the anonymous audit of your service standards.

Notice how specific these are. A CV that says "shrinkage" and "cash reconciliation" reads as someone who lived the job, while "worked on the shop floor" reads as someone filling space.

Key Takeaway: Mirror the advert's own words. If the posting says "conversion" and "ATV", those exact terms belong on your CV, assuming you have earned them.

Keyword table by role level

Keywords are not one-size-fits-all. A stockroom assistant and a store manager are measured on different things, so the ATS expects different vocabulary from each. Find your level and borrow upwards to signal ambition.

Role level	Core keywords	Metrics to quantify
Sales assistant	Customer service, upselling, add-on sales, till operation, cash handling, click and collect, stock replenishment, product knowledge	Upsell rate, till accuracy, items per transaction, sales against target
Stockroom and stock control	Stock replenishment, deliveries, stock accuracy, shrinkage, stock loss, planogram, health and safety, goods in	Stock accuracy %, shrink %, delivery turnaround, availability rate
Visual merchandiser	Visual merchandising, planogram, window displays, brand standards, seasonal campaigns, floor layout, footfall	Footfall conversion, sales uplift per campaign, compliance to brand guidelines
Supervisor and team leader	Rota, cash reconciliation, key holder, opening and closing, KPI, coaching, loss prevention, mystery shopper	Team conversion rate, mystery shopper score, hours managed, shrink reduction
Store manager	P&L, sales targets, footfall, conversion rate, average transaction value, shrinkage, recruitment, seasonal peak, budget, compliance	Annual turnover managed, year on year growth %, headcount, ATV growth, shrink %

Use this as a pick list, not a script. Choose the terms that match the advert in front of you and drop the ones that do not apply to that particular role.

Key Takeaway: Match your keywords to your level and the specific advert. Borrowing one tier up shows you are ready for promotion, as long as you can back it at interview.

How to quantify retail work when you think you have no numbers

This is where most retail CVs leave money on the table. You are surrounded by data every shift and you have been trained to ignore it. The till tells you your average transaction value. The weekly review tells you your conversion. The stock count tells you your shrink.

You do not need exact figures audited to the penny. A fair, honest estimate is enough to transform a bullet point, and it is what every strong retail CV does.

Before (vague)	After (quantified)
Served customers at the till	Processed 200+ transactions per shift with 99% till accuracy on cash reconciliation
Helped increase sales	Lifted average transaction value by 12% through consistent add-on selling
Responsible for stock	Cut shrinkage from 2.1% to 1.4% over two quarters through tighter stock control
Worked on displays	Delivered seasonal window displays that grew footfall conversion by 8% during peak
Managed the team	Scheduled a rota of 14 staff across a 7-day trading week, hitting labour budget every month

The right-hand column is not fancier writing. It is the same job with the evidence attached, and the ATS rewards it because each line carries the keywords a recruiter searches for.

Key Takeaway: Till accuracy, upsell rate, shrink percentage, footfall conversion and hours managed are your five headline metrics. Pull them from data you already see and put a number on every claim.

Progression keywords: assistant to supervisor to manager

If you are applying for a step up, your CV needs to prove you already think at the next level. The keywords should shift from "I did" to "I led".

A sales assistant talks about service and upselling. A supervisor talks about the rota, key holding and cash reconciliation. A manager talks about P&L, budgets and shrink across the whole store.

When you want that supervisor role, salt your assistant experience with early leadership language. Words like "trained new starters", "acted as key holder" and "covered team leader shifts" tell the ATS and the recruiter you are ready.

For the polish that turns a matched CV into an interviewed one, our complete ATS-friendly CV guide walks through formatting, section order and the mistakes that quietly sink applications.

Key Takeaway: To win a promotion, use the vocabulary of the role above yours. Show the leadership work you already do informally before you have the title.

How retail experience translates to other sectors

Retail is not a dead end, and your CV should not read like one. The skills you built on the shop floor are the same skills other sectors pay well for, once you rename them.

Your conversion and upselling become sales and account management. Your rota and team leadership become operations and workforce planning. Your loss prevention and stock accuracy become risk, compliance and inventory management.

Customer service in a busy store is high-pressure stakeholder and complaint handling, which hospitality, call centres and client-facing office roles all want. The trick is to keep the measurable results and swap the retail label for the target sector's label.

Key Takeaway: Your retail metrics travel. Keep the numbers, translate the job titles, and a store role becomes evidence for sales, operations or compliance work.

What the ATS does with high-volume retail applications

Here is the reality of applying for a popular retail job. Your CV lands in a pile of hundreds and the software ranks it in seconds. Speed of filtering is the whole point of the system.

The ATS scores you on how closely your words match the advert, whether your job titles are recognisable, and whether it can read your file at all. A clean layout with standard headings and no text buried in images or tables beats a designed CV every time.

So keep it simple. Use plain section headings, save as a Word or standard PDF, and load it with the honest keywords from this guide. Then check your score with CVPilot so you know exactly where you stand before you hit apply.

Key Takeaway: In high-volume hiring the ATS ranks first and humans read second. A readable file plus advert-matched keywords is what gets you into the shortlist.

Your next shift starts with the CV

You already have the results. Every metric an employer wants is sitting in a shift report you have read a hundred times. The only task left is moving those numbers onto the page in the language the system rewards.

Pick your role level, lift your five headline metrics, and mirror the advert. That is how a shop-floor career reads like the record of results it truly is.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/retail-cv-keywords-ats?utm_source=devto&utm_medium=organic&utm_campaign=retail-cv-keywords

If you have built keyword extraction for a non-technical domain, how did you handle synonyms the taxonomy had never heard of?

Full guide: https://cvpilot.pro/blog/retail-cv-keywords-ats

Fingerprinting the ATS: what an application form's URL tells you before you type a word

Hanzala Mehmood — Tue, 28 Jul 2026 13:00:02 +0000

TL;DR

The URL domain fingerprints the ATS. myworkdayjobs.com, greenhouse.io, lever.co and workable.com each name the parser sitting behind the form.
Workday makes you re-key your CV into typed fields, and that text is what gets scored, not the PDF you uploaded.
Three confirming tells remove any doubt: the parse spinner, the upload limit, and the confirmation email sender domain.

When we built CVPilot's CV parser, the first surprise was how little the pretty PDF mattered on some platforms. Upload the same file to ten Applicant Tracking Systems and you get ten different reads, and a couple barely open the file at all. The form in front of you leaks which engine you are dealing with, and it starts with the URL in the address bar. Read that fingerprint and you know what to optimise before you fill in a single field.

You have found a job you actually want. You click Apply, and a form loads that looks nothing like the last one. Before you type a single word, that form has already told you which robot will read your CV first.

Most applicants never notice. They paste the same CV into every portal and hope for the best. The sharp ones read the signals hiding in plain sight and adjust before they submit.

This is a detective guide. By the end you will glance at an application form and know, within seconds, whether a Workday, Greenhouse, Lever or iCIMS system is standing between you and the recruiter. Then you will know exactly what to change.

Why the ATS behind the form matters more than the form itself

An Applicant Tracking System is the software that receives, parses, ranks and stores your application. Every large UK employer runs one, and they do not all behave the same way. A CV that sails through one can stall in another.

The difference is not cosmetic. Some systems read your PDF cleanly, others mangle two-column layouts, and a few ignore your uploaded file almost entirely in favour of text you re-key by hand. Knowing which one you face changes what you optimise.

Here is the contrarian bit most guides skip. When a form makes you re-type your entire history after uploading a CV, that re-keyed text is what gets scored, not your beautiful PDF. The upload becomes an attachment a human might glance at later. The fields you fill in are the actual application.

Key Takeaway: The ATS is the first reader of your CV, and each platform reads differently. Identify it before you apply and you stop optimising blind.

Signal one: the application URL domain

This is the fastest tell and the most reliable. The moment you land on the application page, look at the address bar. The ATS almost always hosts the form on its own domain or a recognisable subdomain.

You will often see the employer redirected away from their careers page to a third-party URL. That URL is a fingerprint. Read the domain and you have named the system before reading a single form field.

Keep this map handy. Match the domain to the platform, then apply the adjustment in the third column.

URL domain you see	ATS it belongs to	What to do differently
myworkdayjobs.com	Workday	Expect to re-key everything. Perfect the typed fields, not just the PDF.
greenhouse.io / grnh.se	Greenhouse	Clean parser. Use a simple single-column CV and answer custom questions carefully.
lever.co / jobs.lever.co	Lever	Strong parsing. Keep formatting plain and let the LinkedIn autofill check itself.
jobvite.com	Jobvite	Older parser. Avoid tables and headers or footers that hold contact details.
icims.com	iCIMS	Legacy fields. Keep job titles standard and expect a long multi-step form.
workable.com	Workable	Decent parser. Fill the knockout questions honestly, they gate your application.
teamtailor.com	Teamtailor	Design-led forms. Content over layout, and expect optional video or questions.
pinpointhq.com	Pinpoint	Modern parser. A clean PDF uploads well, so review the auto-filled fields.
ashbyhq.com	Ashby	Excellent parser. Keep it simple and check the structured data it extracts.
smartrecruiters.com	SmartRecruiters	Reliable parsing. Mind the mandatory screening questions before you submit.

One caveat worth remembering. A company can embed the form inside its own branded careers page using an iframe, so the visible URL may not change. When that happens, the other signals below give you the answer.

Key Takeaway: The domain is your quickest identifier. Learn ten domains and you will recognise the ATS on most UK applications at a glance.

Signal two: the look and behaviour of the form

When the URL hides behind a branded wrapper, the form's shape gives it away. Each platform has a house style that is hard to disguise once you know what to look for.

The Workday tell: re-keying your whole CV

Workday is unmistakable. You upload your CV, it parses for a moment, and then it asks you to confirm or re-enter every role, date and responsibility by hand. The Work Experience section stretches on with individual fields for each position.

If you find yourself retyping your career into little boxes, you are almost certainly in Workday. Treat those boxes as the real submission and write them as carefully as you wrote the CV.

Greenhouse and Lever: short, clean, question-led

Greenhouse forms are famously tidy. A single upload field, autofilled name and email, then a short set of custom questions the employer added. The parser is reliable, so a simple CV comes through intact.

Lever looks similar and behaves well too. Both reward substance over styling, which is why a single-column, standard-heading CV performs best in them.

iCIMS and Jobvite: the long, dated multi-step journey

Older systems feel older. iCIMS and Jobvite tend to spread the process across several pages with progress bars, EEO questions and dated form styling. Their parsers are less forgiving of tables and columns, so plain formatting matters most here.

Key Takeaway: If the form makes you re-key everything, it is Workday and the typed text is what counts. If it is short and question-led, it is likely Greenhouse or Lever and your clean CV will carry.

Signal three: the parse step, upload limits and the confirmation email

Three smaller clues confirm what the domain and layout suggested. Together they remove any doubt.

The "parse my CV" moment

Watch what happens the instant you upload. A brief spinner followed by pre-filled fields tells you the parser has read your file. Whether those fields come back accurate tells you how well your CV is structured for machines.

If the parse scrambles your job titles or drops your dates, that is a warning. A human recruiter may never fix it, so correct every mis-parsed field before you submit.

File type and upload limits

The upload box itself leaks information. Modern systems such as Ashby, Pinpoint and Greenhouse happily take a PDF up to several megabytes. Older platforms sometimes cap uploads tightly or nudge you towards .doc, a sign the parser is dated and picky.

When a form prefers Word or limits you to a small file, take the hint. Strip graphics, keep it text-first, and favour a lean document the older engine can digest.

The confirmation email sender

The final proof arrives after you submit. Check the sender domain on the confirmation email. A message from a myworkday, greenhouse or icims address confirms the platform beyond argument, and it is useful intelligence for your next application to the same firm.

Key Takeaway: The parse result, the upload rules and the confirmation sender each confirm the ATS. Read all three and you never guess again.

What to change once you know

Identification is only half the job. The point is to adjust your application to the system in front of you. Here is how the tailoring shifts.

For Workday, pour your energy into the typed fields. Match your job titles to the advert's language, write full bullet points inside those boxes, and do not assume the attached PDF will be read at all. Our Workday guide walks through the re-keying trap step by step.

For Greenhouse, Lever and the other clean parsers, keep the document simple and let it do the work. Single column, standard headings, no text buried in headers or footers. The Greenhouse guide covers the custom-question fields that quietly influence your ranking.

For older engines like iCIMS and Jobvite, formatting discipline wins. Remove tables, columns and graphics entirely, and put your contact details in the body rather than a header. When in doubt, test your file against a scanner first.

That is where a tool earns its place. CVPilot scores your CV the way these parsers do and flags the exact lines that will trip a given system. You can browse every platform in our ATS platform guides and match your CV to the one you have identified.

Key Takeaway: Same CV, different emphasis. Workday needs strong typed fields, clean parsers need a simple document, and legacy systems demand strict formatting discipline.

Put it into practice on your next application

Start treating the application form as evidence, not an obstacle. Read the URL, watch the parse, note the confirmation email, and log which employers use which system. Within a few applications you will recognise them instantly.

The reward is real. You stop sending one generic file into ten different machines and start submitting a version each system can actually read. If you want the full grounding, our complete ATS-friendly CV guide ties the tactics together.

Detective work only pays off if you act on the clues. Identify the ATS, adjust the CV, and let CVPilot confirm the file is clean before you hit submit.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/which-ats-is-this-company-using?utm_source=devto&utm_medium=organic&utm_campaign=which-ats

Which ATS has burned you before, and did you spot it from the form or only after the rejection?

Full guide: https://cvpilot.pro/blog/which-ats-is-this-company-using

Answering "what are your salary expectations?" in a UK interview: deflect once, then commit

Hanzala Mehmood — Mon, 27 Jul 2026 13:00:01 +0000

TL;DR

Endless deflection now reads as evasive to UK recruiters, so deflect once to learn the band, then give a number.
Build a tight range, roughly a 10 to 15 percent spread, from three agreeing sources: live adverts, published pay scales, and crowdsourced data.
You never have to disclose your current salary in the UK; redirect to expectations so past pay does not anchor your next offer.

I have watched the salary question break more strong candidates than any coding test ever did. While building CVPilot I kept seeing the same pattern, capable people freezing on a figure they had never actually researched. Here is the approach that holds up in a real UK screening call, deflection and all.

You are 20 minutes into a screening call. The chat is flowing, you like the role, and then it lands: "So, what are your salary expectations?" Your mouth goes dry and you mumble a number you have not thought through.

This moment trips up brilliant candidates every week. The question is not a trap, but it is a test, and how you handle it shapes the whole conversation that follows.

This guide is about that single question, asked during an interview or screening call, before any offer exists. If you already have an offer and want to push it higher, that is a different job, linked once below.

Why They Ask About Salary Expectations

Interviewers are not being nosy. They ask to check three things at once: whether you fit the budget, whether you understand your market value, and whether your expectations suit the level.

Most UK roles have a banded salary range signed off long before you applied. The recruiter is quietly matching your number against that band. Land inside it and you progress. Sit above it and they worry you will reject any offer.

There is a filtering motive too. When 200 people apply, salary is a fast way to shrink the shortlist. A figure far below the band makes you look junior. One far above prices you out.

Key Takeaway: The salary question is a fit check against a pre-set band, not an ambush.

Who Is Asking Changes Your Answer

The same question means different things depending on who says it. A recruiter and a hiring manager are solving different problems, so your response should shift accordingly.

An agency or internal recruiter usually asks early, often on the first call. Their job is to screen for budget fit. With a recruiter, a researched range is your friend because they can tell you on the spot if it works.

A hiring manager asks later and cares more about value than budget. By the time they raise money, they often want you. With a hiring manager, anchor higher and tie the number to your impact.

Who is asking	Best response	Why it works
Agency recruiter (first call)	Give a researched range and ask for the band	They can confirm fit instantly and often share the range back
Internal recruiter / HR	Range, plus a light question about the budget	They own the band and appreciate a realistic candidate
Hiring manager (later stage)	Anchor at the top of your range, linked to impact	They value outcomes and have more room to flex
Application form field	Enter the middle of the advertised or researched band	A blank or a silly figure can auto-reject you

Key Takeaway: Recruiters screen for budget, so give them a range. Hiring managers weigh value, so anchor higher.

The Contrarian Truth About Deflection

You will read a lot of advice telling you to dodge the question forever. Never say a number first, make them show their hand. That playbook is ageing badly in the UK market.

Recruiters have heard every deflection line a thousand times. When you refuse to engage after two polite nudges, a hard dodge now reads as evasive or naive. It can stall the process while they chase someone who will just answer.

Here is the shift worth making. A researched range, given with quiet confidence, beats a clever dodge nearly every time. It signals you have done your homework and you are easy to work with.

Deflection still has a place. Use it once, early, to buy information, then lead with a number. A tool like CVPilot helps you pitch a CV that justifies the top of your range.

Key Takeaway: Endless deflection now signals evasion to UK recruiters. Deflect once, then give a confident, researched range.

Deflection Scripts: The Ones That Work

A single, graceful deflection early on is smart. The goal is to learn the band before you commit to a figure. These lines work because they stay collaborative:

"I'd love to understand the band for the role first, so I can be realistic. Are you able to share the range you're working with?"
"Happy to talk numbers. To make sure we're aligned, what has been budgeted for this position?"
"My expectations depend a little on the full package. Could you tell me the salary range and how the benefits sit alongside it?"

Notice these all end with a question and a reason. You are not refusing, you are sequencing, and that tone keeps things friendly.

The Scripts That Irritate

Some deflections make recruiters wince. Avoid anything that sounds like a negotiation tactic lifted from a blog. These land badly:

"I'd rather not say until I have an offer." Reads as cagey and stalls the screen.
"I'm sure you'll be fair." Passive, and it hands them the power anyway.
"Salary isn't important to me." Nobody believes it, and it can invite a lowball.

Key Takeaway: One warm deflection that asks for the band is professional. Repeated stonewalling reads as difficult before you have started.

When to Give a Number, When to Give a Range

Once you are ready to commit, the format matters. A range is almost always the safer opening on a call. It shows flexibility without pinning you to a single figure.

Build your range with a floor you would genuinely accept and a top that reflects a strong fit. Keep the spread tight, around 10 to 15 percent, so it reads as considered. A £42,000 to £48,000 range beats a £35,000 to £55,000 one.

Give a single number in two situations. When a form demands one field, and when a hiring manager who wants you asks directly late on. In both cases, quote a figure near the top of your range, because they rarely offer above what you name.

Here is a script for the confident range answer:

"Based on my research and the level of the role, I'm looking in the region of £45,000 to £50,000, depending on the wider package. Does that sit comfortably within your band?"

Key Takeaway: Default to a tight range on calls and forms. Reserve a single top-of-range number for the hiring manager who asks outright.

How to Research a UK Salary Range

Your range is only as good as the homework behind it. Never pull a figure from thin air, because a well-briefed recruiter will spot it instantly.

Start with live job adverts. Search the exact job title on Indeed, LinkedIn and Reed and note the posted ranges for your region. London roles run higher than the same job in Leeds, so filter by location.

Next, use published pay scales. The public sector, NHS Agenda for Change, and many large employers publish exact bands you can read off directly. Professional bodies often share salary surveys too.

Then lean on UK salary transparency. More adverts now list ranges by law or convention, and sites like Glassdoor crowdsource real figures for specific employers.

Cross-check three sources before you trust a number. When adverts, pay scales and crowdsourced data agree, you have a defensible range and the confidence to say it out loud.

Key Takeaway: Triangulate live adverts, pay scales and crowdsourced data. Three agreeing sources turn a guess into a figure you can defend.

The Application Form Field That Forces a Number

Some online forms will not let you proceed without a salary figure. Leaving it blank or typing zero can trigger an automatic reject, so you have to enter something sensible.

Enter the middle of your researched range as a clean round number. Avoid "negotiable" where a number is expected, since many systems read it as invalid and bin the application.

If the form offers a free-text box, add one line: "Open to discussion based on the full role and package." That single sentence signals flexibility without weakening the number you gave.

Key Takeaway: Forms that demand a figure will reject a blank. Enter the middle of your range as a round number and note you are open to discussion.

The "Current Salary" Question You Do Not Have to Answer

This one catches people off guard. A recruiter asks what you earn now, and many candidates answer on reflex. You are under no obligation to disclose your current salary in the UK.

Sharing it anchors your next offer to your past pay, which is exactly why they ask. If you are underpaid today, that anchor works against you and caps your rise before you have negotiated.

Redirect politely toward expectations instead. A calm pivot protects your position:

"I tend to keep my current salary private, but I'm happy to share what I'm looking for. For this role I'd expect somewhere around £48,000 to £52,000."

That answer refuses the anchor and offers useful information instead. You look composed and fair, not obstructive. A sharp CVPilot profile that evidences your level makes this pivot far easier.

Key Takeaway: You never have to reveal your current salary. Decline warmly and redirect to your expectations, so past pay never anchors your next offer.

What to Do If You Name Too Low

It happens to almost everyone once. You blurt a figure, then realise it was under your worth. A low number is recoverable if you act before the offer arrives.

The moment you learn more about the scope, revisit it. Extra responsibilities or a wider remit justify a revised range. Tie the correction to new information, not to regret.

Try this recovery line at the next natural opening:

"Now I understand the full scope, including the team leadership, I'd revise my expectations up to around £55,000. That reflects the level of responsibility we've discussed."

If the offer has already arrived at your low figure, that is where separate negotiation begins. For the full playbook on pushing an existing offer, read our salary negotiation guide. And if competency questions are next, our STAR method guide will sharpen your answers.

Key Takeaway: A low number is fixable while you are still interviewing. Anchor the correction to new scope, not to second thoughts, and do it before any offer arrives.

Bringing It Together

The salary expectations question rewards preparation over reflex. Research a tight range, deflect once to gather intel, then answer with quiet confidence. Read who is asking, guard your current salary, and correct a low number the instant scope changes. Do these well and you walk into offer talks from strength rather than hope.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/salary-expectations-uk-interview-answer?utm_source=devto&utm_medium=organic&utm_campaign=salary-expectations

What is your rule on a first call: name a range early, or hold out until they show the band first?

Full guide: https://cvpilot.pro/blog/salary-expectations-uk-interview-answer

The post-interview email is a one-line commit, not a rewrite

Hanzala Mehmood — Sun, 26 Jul 2026 13:00:02 +0000

TL;DR

A thank-you email almost never wins the job, but a sloppy one can lose it
Its real job is to be brief and add exactly one piece of fresh evidence
Reference one concrete thing from the interview, then tie it to what you would bring

Think of the follow-up email like a hotfix, not a refactor. The interview already shipped the decision; you are only patching in one small, high-value line. When I was building CVPilot I read a lot of candidate follow-ups, and the ones that landed were always the shortest.

You walked out of the interview feeling good. Now you are staring at your inbox wondering whether to send anything, and if you do, what on earth to write. Most people either send nothing or send a gushing paragraph that reads like a hostage note.

The interview follow up email is one of the most misunderstood moves in hiring. Get it right and you nudge a hesitant panel back towards you. Get it wrong and you hand them a reason to pause.

After fifteen years placing candidates across UK firms, I can tell you the difference between a follow-up that helps and one that hurts comes down to timing, restraint, and a single well-chosen detail.

The 24-hour rule: when to send it

Send your thank-you email within 24 hours of the interview, ideally the same working day. This is the window where you are still fresh in the interviewer's mind and the decision has not yet hardened.

Do not send it at 11pm the night of the interview. A timestamp in the small hours reads as anxious or automated. Aim for the next morning, between 8am and 10am, when it lands at the top of a freshly opened inbox.

If you interviewed on a Friday afternoon, Monday morning is perfectly acceptable and better than a weekend email nobody reads. Be timely without looking like you have been refreshing your inbox for eighteen hours.

Key Takeaway: Send within 24 hours, in working hours. Same-day is fine; midnight is not.

Who to send it to

Send it to the person who actually interviewed you, not the recruiter who booked the meeting. If a panel of three grilled you, a short individual note to each is ideal, but a single email to the chair copying the others is fine when addresses are scarce.

You will not always have their email. If you only have the recruiter's address, send it there and ask them to pass it on. A good internal recruiter forwards a sharp note gladly because it makes their shortlist look stronger.

Guessing corporate email formats is risky, so do not agonise over it. LinkedIn is a legitimate fallback if the email route is closed, though keep it to a couple of lines.

Key Takeaway: Reach the actual interviewer where you can. Otherwise route it through the recruiter rather than guessing addresses.

Subject lines that actually get opened

Your subject line has one job: to be recognised instantly as relevant, not deleted as spam. The safest, highest-opening subject line is simply "Thank you, [role] interview". It is clear, it references the meeting, and it never looks like a pitch.

If you had a warm conversation, you can reply within the existing email thread instead of starting a new one. That keeps your name attached to the original booking and lands you in a thread they already recognise.

Avoid anything cute. "Great to meet you!" or worse, "Don't forget me", tells the reader you are treating this as marketing rather than a professional courtesy.

Key Takeaway: Keep the subject boring and clear. "Thank you, [role] interview" beats anything clever.

The contrarian truth about thank-you emails

Here is the part nobody tells you. The thank-you email almost never wins you the job on its own. Hiring decisions are made on the interview, the CV and the fit, not on a polite note sent afterwards.

What a bad thank-you email can do is lose you the job. A typo in the hiring manager's name, a wall of desperate text, or a factual error about the role hands a wavering panel an easy tie-breaker.

So the real job of your thank-you email is to be forgettable in the best way and add exactly one piece of fresh evidence. It should never be the thing they remember; it should quietly reinforce what they already liked.

Key Takeaway: A great thank-you rarely wins the job, but a sloppy one can lose it. Reassure, do not dazzle.

The one-specific-callback technique

This is the single move that separates a memorable follow-up from filler. Reference one concrete thing from the conversation, something only someone who was in the room would know.

Maybe they mentioned a migration project that had stalled, or a tool the team was trialling. Name it, then connect it to something you would bring. That proves you listened and were already thinking like a member of the team.

Generic gratitude is invisible. "Thank you for your time, I am very excited about this opportunity" could have been written before you walked in. The callback is the only sentence that cannot be copied and pasted.

If part of your evidence is a strong CV or portfolio piece, this is a natural place to attach an updated version. A quick pass through CVPilot before you resend makes sure it is the sharpest version they have seen.

Key Takeaway: One concrete callback is worth more than three paragraphs of enthusiasm.

What never to include

Certain things turn a helpful note into a liability. Never negotiate salary, demand a decision timeline, or list every reason you are perfect. The follow-up is a courtesy, not a second interview.

Skip the flattery that curdles. Calling them "the best team I have ever met" after a 45-minute chat reads as insincere. Keep it grounded in what actually happened in the room.

And do not re-answer something you fumbled at length. One clean sentence adding a point you missed is fine. A three-paragraph re-answer only reminds them that you flubbed it.

Key Takeaway: No salary talk, no smothering flattery, no re-litigating the interview. Restraint reads as confidence.

Template 1: the post-interview thank you

Here is the model to copy. Notice how short it is and where the callback sits.

Subject: Thank you, Marketing Manager interview
Hi Sarah,
Thank you for meeting me yesterday. I enjoyed the conversation and left even more interested in the role.
Your point about the email programme underperforming against paid social stuck with me. In my last role I rebuilt a neglected email flow and lifted repeat purchases by 18 percent in a quarter, so it is exactly the problem I would relish here.
I have attached an updated CV in case it is useful. Do let me know if you need anything further from me.
Best wishes,
James Okafor

That email does one thing well: it proves he listened and offers one concrete result. It does not beg or sprawl, and it gives Sarah something to repeat in the debrief.

Key Takeaway: Short greeting, one callback tied to a result, a light offer to help, sign off. Nothing more.

When and how to chase after silence

You were told you would hear back by Wednesday and it is now Friday. Wait until the stated date has passed, then give it two or three working days before you chase. Hiring runs late for reasons that have nothing to do with you.

The chase email should be shorter than the thank-you and free of reproach. You are reaffirming interest, not demanding an update. One friendly nudge is professional; a second within days is not.

The timing below is what I coach candidates through, with the outcome to expect at each stage.

Stage	When to send	What to include	Expected outcome
Thank-you	Within 24 hours	One specific callback	Reassures the panel, no reply needed
First chase	2-3 days after the promised date	Brief interest, gentle check-in	A status update or honest "still deciding"
Second chase	5-7 days after the first	Final polite nudge, soft deadline	A decision, or a signal to move on
Withdrawal	As soon as you decide	Thanks, clear exit, door left open	Goodwill, future roles stay possible
Competing offer	Immediately on receipt	Honest deadline, continued interest	An accelerated decision

Key Takeaway: Chase only after the promised date has passed, keep it warm, and never send a third.

Template 2: chasing after silence

Subject: Re: Marketing Manager interview
Hi Sarah,
I hope the week has treated you well. I wanted to check in on the Marketing Manager role, as I know you had hoped to decide around the end of last week.
I remain very keen and happy to answer anything else that would help. No rush at all, I simply did not want my interest to be in doubt.
Best wishes,
James Okafor

The magic word there is "keen", not "waiting". He frames the silence as their busy schedule, not rudeness, and leaves them feeling helped rather than hounded.

Key Takeaway: Frame the chase around your continued interest, not their delay. Warmth gets replies; pressure gets ignored.

The polite withdrawal and the competing offer

Sometimes you accept elsewhere or decide the role is not right. Withdrawing gracefully is one of the most underrated career investments you can make. Recruiters have long memories and small worlds.

When a competing offer lands and you would still prefer this role, say so and give a real deadline. A polite "I have an offer I must answer by Thursday, and you are my first choice" often unsticks a slow process overnight.

Practise your answers to the tougher questions before the next round too. Our STAR method guide is the quickest way to make sure the interview gives you a strong callback to reference later.

Key Takeaway: Leave every process on good terms, and use a real competing deadline to prompt a decision without bluffing.

Template 3: responding to a rejection

A rejection is not the end of the relationship unless you make it one. A gracious reply keeps you in the pipeline for the next role, and second-choice candidates get called more often than you think.

Subject: Re: Marketing Manager role
Hi Sarah,
Thank you for letting me know, and for the time your team gave me. I am disappointed, but I understand it was a competitive field.
I thought highly of the team and the direction you described, so if a similar role opens up I would welcome the chance to be considered. I would also value any brief feedback, if you have a moment.
Wishing you and the new hire every success.
Best wishes,
James Okafor

That reply does the one thing most rejected candidates never do: it keeps the door open without a hint of bitterness. A quick polish through CVPilot means you are ready the moment they come back to you.

Key Takeaway: Answer a rejection with grace and a soft ask to stay in mind. The runner-up is often the eventual hire.

Pulling it together

Few interviews are won by the follow-up email. A handful are quietly lost by it. Send promptly, write briefly, land one specific callback, and stay warm through silence and rejection alike.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/interview-follow-up-email-template?utm_source=devto&utm_medium=organic&utm_campaign=follow-up-email

What is the one callback line you would add to your last interview?

Full guide: https://cvpilot.pro/blog/interview-follow-up-email-template

UK competency interviews are a caching problem: store six stories, not twenty answers

Hanzala Mehmood — Sat, 25 Jul 2026 13:00:02 +0000

TL;DR

UK competency panels score evidence against a checklist, not your personality
Twenty question phrasings map to seven or eight underlying competencies, so the answers overlap heavily
Build six flexible STAR stories and compose them on the fly instead of memorising twenty scripts

Preparing for a competency interview by writing a separate answer for every question is like caching one entry per request when the keys overlap. It does not scale, and it thrashes the moment the wording shifts. While building CVPilot I kept seeing the same pattern in how people prep, so this guide lays out the smaller, reusable set that actually holds up under pressure.

Here is the uncomfortable truth about competency interviews. The panel is not really listening for whether you are good at teamwork or resilience. They want evidence, delivered in a structure their scoring sheet can tick.

UK employers love this format because it is fair on paper. The NHS, the Civil Service, most councils, the big graduate schemes and a huge slice of private sector hiring all run on it.

The single biggest mistake candidates make is trying to prepare a fresh answer for every possible question. That way lies twenty half-memorised stories and a blank mind when the wording shifts. The smarter play is to build six flexible stories and learn to bend them to whatever gets asked.

Why six stories beat twenty rehearsed answers

Competency questions recycle. Behind the twenty phrasings sit maybe seven or eight underlying competencies, and a strong story usually covers three or four at once.

A single project where you led a nervous team through a tight deadline can answer teamwork, leadership, resilience and time management, depending on which thread you pull. Prepare the raw material, not the script.

The structure that makes this work is the STAR method guide: Situation, Task, Action, Result. Learn it once and it becomes the frame for every answer, which is why building a reusable story bank pays off far more than cramming.

Key Takeaway: Prepare six versatile STAR stories, not twenty scripts. The questions overlap, so one good example can be steered to answer several competencies.

The 20 questions, grouped by competency

These are the real phrasings that recur across UK interviews. For each, I have flagged what the interviewer is actually testing and which story to reach for.

Teamwork and collaboration

Question	What they are really testing	Story to pick
Tell me about a time you worked in a team to achieve a goal.	Whether you share credit or hog it	A project where your specific contribution mattered but the win was collective
Describe a situation where a team member was not pulling their weight.	Tact and initiative, not tattling	A time you addressed it directly before escalating
Give an example of working with someone whose style clashed with yours.	Adaptability and self-awareness	A colleague you learned to flex around

The trap here is the word "we". Panels want to hear what you personally did, so a team answer that never says "I" scores badly. Name your action inside the collective effort.

Conflict and difficult relationships

Question	What they are really testing	Story to pick
Tell me about a disagreement with a colleague and how you resolved it.	Emotional control under friction	A dispute you defused by listening first
Describe a time you had to give someone difficult feedback.	Directness balanced with respect	A moment you were honest but kind
How did you handle a disagreement with your manager?	Whether you can challenge upwards professionally	A time you pushed back with evidence, then committed to the decision

Never pick a story where you were simply right and they were wrong. The competency being scored is resolution, so the ending must show a working relationship intact, not a scorecard.

Resilience and pressure

Question	What they are really testing	Story to pick
Tell me about a time you failed and what you learned.	Ownership and growth, not blame	A genuine setback with a concrete lesson applied later
Describe how you coped with a heavy workload or tight deadline.	Prioritisation, not heroics	A crunch you handled by triaging, not by staying up all night
Give an example of a time you dealt with a significant setback.	Recovery and steadiness	A knock you absorbed and kept moving through

Choose a real failure, not a humblebrag. "I work too hard" fools nobody, whereas a project that missed its target and taught you something shows the self-awareness they are scoring.

Influencing and persuasion

Question	What they are really testing	Story to pick
Tell me about a time you persuaded someone to see things your way.	Whether you use evidence or pressure	A case you won by reframing around their priorities
Describe how you gained buy-in for an idea from a sceptical audience.	Stakeholder awareness	A proposal you tailored to what the room cared about

Persuasion answers live or die on why the other person said yes. Spell out their objection and how you met it, or the story reads as luck rather than skill.

Decision-making and judgement

Question	What they are really testing	Story to pick
Tell me about a difficult decision you had to make with limited information.	Judgement under uncertainty	A call you made with a clear rationale, even if imperfect
Describe a time you had to make an unpopular decision.	Backbone and fairness	A choice you stood behind and explained honestly
Give an example of a time you had to weigh competing priorities.	Structured thinking	A trade-off you reasoned through out loud

Show your reasoning, not just the outcome. The panel is scoring how you decided, so walk them through the options you weighed and the criteria you used.

Customer focus and service

Question	What they are really testing	Story to pick
Tell me about a time you went above and beyond for a customer or client.	Genuine service instinct	An extra step you took without being asked
Describe how you handled an angry or dissatisfied customer.	Composure and problem-solving	A complaint you turned around by owning it

The best service stories end with a measurable turnaround. A saved account, a repeat booking or a written thank-you gives the panel something concrete to mark.

Leadership and initiative

Question	What they are really testing	Story to pick
Tell me about a time you took the lead without being asked.	Initiative and ownership	A gap you spotted and filled voluntarily
Describe how you motivated a team through a challenging period.	People awareness	A slump you lifted with a specific action
Give an example of when you delegated a task effectively.	Trust and judgement	A moment you matched the work to the right person

Leadership is not the same as job title. A panel will happily score a graduate who organised a rota or rallied a stalling group project, so long as the impact is clear.

Key Takeaway: Every question maps to a competency and a hidden test. Read the phrasing, name the competency, then reach for the story that shows the specific behaviour, not just a good outcome.

Two full worked answers

Reading twenty model answers is a fast route to sounding rehearsed. Here are two done properly so you can see the STAR structure carrying the weight, and how much of the score sits in the Action.

Worked answer 1: "Tell me about a disagreement with a colleague."

Situation. On a product launch last year, our designer and I disagreed sharply over the checkout flow. She wanted a single long page, I was sure a three-step version would convert better.

Task. We had four days before the build froze, and needed to agree without pulling the project manager in to referee.

Action. I asked to see the reasoning behind her single-page view first, and it turned out she was worried about mobile drop-off, which was fair. Rather than argue from opinion, I suggested we pull the analytics from our last two launches and run a quick five-user test on both mockups. I booked the sessions, she prepped the prototypes, and we agreed in advance to go with whatever the data showed.

Result. The three-step flow tested better on desktop but her mobile worry was real, so we shipped a hybrid that adapted by device. Conversion rose 14% on the previous launch, and we have used that "test it rather than debate it" habit ever since.

Why it scores: it opens with listening, resolves through evidence rather than seniority, and ends with the relationship and a result both intact. That is exactly what a conflict competency marks.

Worked answer 2: "Tell me about a time you failed and what you learned."

Situation. In my first year managing a small team, I owned the rollout of a new booking system to forty branch staff.

Task. My job was to get everyone trained and confident before the go-live date at the end of the quarter.

Action. I built what I thought was a thorough email guide and slide deck, sent them round, and assumed people would work through them. When I spot-checked two weeks in, barely a third had opened the material and confidence was low. I stopped, admitted the approach was not working, and switched to short hands-on sessions in each branch, plus a named contact per site for questions.

Result. We slipped the go-live by ten days, which I owned to my manager directly. Adoption hit 95% within a month, and I have never again confused sending information with making sure it landed.

Why it scores: it is a real failure with a real cost, the recovery is active rather than lucky, and the lesson is specific enough to believe. A rehearsed "my weakness is perfectionism" answer dies in the first ten seconds.

Key Takeaway: The Action section is where most of the marks live. Compress the setup, expand what you personally did, and always land a concrete, measurable result.

How to prepare in a week

Start by listing the six moments in your career you are proudest of, or most changed by. Write each as raw notes under Situation, Task, Action, Result, with real numbers where you can.

Then map each story against the seven competency groups above and check for gaps. If nothing covers conflict or customer focus, that is where you dig for a fresh example now, not the night before.

Practise saying them aloud, aiming for roughly two minutes per answer, long enough for detail and short enough to hold attention.

It also helps to mine your CV for evidence you have forgotten. Running it through CVPilot surfaces the quantified achievements that make the strongest STAR results.

Key Takeaway: Six stories, mapped to the competencies, rehearsed aloud to two minutes. That beats twenty scripts every time, because real interviews reward flexibility over memory.

The night before

Do not try to learn new material. Re-read your six stories, remind yourself of the numbers, and get a proper night's sleep, because a fresh mind adapts far better than a tired one that has crammed.

Walk in expecting to bend your stories, not recite them. When a question you did not prepare for lands, pause, pick the closest story and steer it. That composure is itself a competency, and panels notice it.

Sharpen your evidence beforehand and the answers get easier. CVPilot is a quick way to pressure-test your CV before the day.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/uk-competency-interview-questions?utm_source=devto&utm_medium=organic&utm_campaign=competency-questions

Which competency do you find hardest to cover with a single reusable story?

Full guide: https://cvpilot.pro/blog/uk-competency-interview-questions

Structured answers under pressure: what the STAR method gets right about evidence

Hanzala Mehmood — Fri, 24 Jul 2026 13:00:01 +0000

TL;DR

An ATS gets you into the room. What happens next is a verbal audit of whether your claims have mechanisms behind them.
Half a STAR answer belongs in Action, the only section that evidences your own capability. Most candidates underweight it and overweight Situation.
An answer that stops before the Result is an anecdote. A number, a timeframe or an adopted decision converts it into evidence.

I build CVPilot, so most of my time goes on the machine-readable end of hiring: parsers, keyword extraction, scoring. The interview is where that pipeline hands off to a human, and the failure mode changes completely. Nobody is scanning for keywords any more. They are checking whether the thing you claimed on paper has a causal story behind it.

Ask a hiring manager why they rejected a candidate who looked perfect on paper, and you will usually get the same answer. They could not give a straight example.

Competency interviews exist to test whether you have actually done the thing, or merely watched someone else do it. The STAR method is the structure that separates the two, and it is the most useful interview technique nobody uses properly.

Most people know the acronym. Very few notice that the R is where interviews are won, and it is the letter candidates skip.

What STAR actually asks of you

STAR stands for Situation, Task, Action, Result. It is a container for a story, not a script. The purpose is to prove that a specific outcome exists in the world because of something you personally did.

Letter	What it means	Time to spend	Common failure
Situation	The context, briefly	15%	Rambling backstory nobody needs
Task	What you specifically had to achieve	15%	Describing the team's job, not yours
Action	What you did, step by step	50%	Saying "we" instead of "I"
Result	The measurable outcome	20%	Omitted entirely, or vague

Notice the weighting. Half your answer belongs in Action, because that is the only part that evidences your capability rather than your circumstances.

Key Takeaway: Situation and Task are scene-setting. Action proves you can do the job. Result proves it mattered.

The two failures that sink most answers

The "we" problem

Under pressure, candidates retreat into the collective. "We identified the bottleneck, we redesigned the process, we cut the delay." The interviewer learns that a team did something, and nothing at all about you.

Say "I" when it was you. Say "I persuaded the team to" when persuasion was your contribution. Credit colleagues once, then return to first person and stay there.

The missing Result

People run out of runway and finish on Action, as though the doing was the point. An answer without a Result is an anecdote. A number, a timeframe, or an adopted decision converts it into evidence.

If you genuinely have no number, use a consequence: "the process is still in use two years later", or "my manager asked me to run the same review for the other two regions".

Key Takeaway: Say "I", and never end on the Action. The Result is the whole point of telling the story.

A worked example, before and after

Before

"We had a lot of complaints about delivery times, so we looked at the process and made some changes, and things got better after that."

Three sentences, no ownership, no method, no measurement. It answers the question grammatically and not at all substantively.

After

Situation: "Complaints about late deliveries had roughly doubled over a quarter, and we had no idea where in the chain the delay sat."

Task: "I was asked to find the cause and propose a fix within a month, without extra headcount."

Action: "I pulled six months of dispatch logs and mapped every order against its promised date. That showed 70% of the delay came from one warehouse re-checking stock it had already picked. I sat with two pickers for a day to understand why, and it turned out a system update had removed a confirmation screen they relied on. I wrote the change request, got it prioritised by showing the complaint cost, and retrained the shift on the new flow."

Result: "Late deliveries fell 62% within six weeks. The complaint queue dropped from around 40 a week to under 15, and the fix was rolled out to two other sites."

Same story. The second version is hireable because every claim has a mechanism behind it.

Key Takeaway: Specificity is not decoration. It is the only thing that distinguishes you from a candidate who read the same advice.

Ten questions, and the story each one wants

Competency questions recycle endlessly. Prepare six flexible stories rather than thirty rehearsed answers, because most questions are asking for the same handful of proofs from different angles.

"Tell me about a time you handled conflict." Wants: you de-escalated and preserved the working relationship.
"Describe a time you missed a deadline." Wants: you owned it, communicated early, and changed something afterwards.
"Give an example of influencing without authority." Wants: you used evidence, not seniority.
"Tell me about a difficult decision." Wants: you weighed a real trade-off and can name what you gave up.
"Describe a time you improved a process." Wants: a before number and an after number.
"Tell me about a mistake you made." Wants: genuine fault, not a humblebrag about caring too much.
"When have you handled competing priorities?" Wants: a stated principle you used to choose.
"Describe working with a difficult stakeholder." Wants: you diagnosed their incentive rather than their personality.
"Tell me about a time you learned something quickly." Wants: the method, not the fact that you are a fast learner.
"Give an example of going beyond your role." Wants: initiative with a result, not unpaid overtime.

Six well-chosen stories, each carrying two or three of these angles, will cover almost any competency interview. If you have already built a set of these, reusing them across applications is far more efficient than starting fresh each time.

Key Takeaway: Do not memorise answers. Build six stories and learn to angle them.

Preparing without sounding rehearsed

Write each story as four bullets, one per letter, and never as prose. Prose gets memorised and memorised answers sound dead. Bullets force you to reconstruct the story live, which is what makes it sound like a memory rather than a recitation.

Rehearse the Result out loud until the number arrives without hesitation. That single moment of fluency does more for your credibility than any amount of polish elsewhere.

Your stories should already exist on your CV. If an interviewer asks for an example of process improvement and nothing on your CV hints at one, the mismatch is doing quiet damage long before the interview. Scoring your CV against the advert with CVPilot tends to surface exactly which competencies the employer will probe.

Key Takeaway: Bullets, not scripts. Rehearse the number, not the wording.

The follow-up probe

Strong interviewers rarely accept the first answer. They probe: "what would you have done differently?", "what was the hardest part?", "how did you know it worked?" Treat each probe as an invitation to show judgement.

Answer honestly and specifically. Naming a genuine trade-off you got wrong, and what you learned from it, reads as far more senior than defending a flawless account that nobody quite believes.

Prepare one probe answer per story: the thing you would change. It costs two minutes and it is the question that most often separates two otherwise identical candidates.

Key Takeaway: Expect two probes per story. The probe is where the interviewer decides whether the story is genuinely yours.

A short checklist before the interview

Six stories written as four bullets each.
Every story ends on a number, a timeframe or an adopted decision.
Each story uses "I" for your contribution.
Action is the longest section of every answer.
At least one story is an honest failure with a real lesson.
Each story maps to a requirement in the job advert.
You can tell any of them in ninety seconds.

Competency interviews reward the candidate who has genuinely reflected on their own work. CVPilot can help you get to the interview. What you do once you are in the room is the part that stays yours.

An interview is not a test of who you are. It is a test of what you can evidence.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/star-method-competency-questions-examples?utm_source=devto&utm_medium=organic&utm_campaign=star-method

When you interview engineers, what is the question that most reliably separates people who did the work from people who watched it happen?

Full guide: https://cvpilot.pro/blog/star-method-competency-questions-examples

Your CV Is a Parser Problem: Evidence Density for Skills-Based Hiring

Hanzala Mehmood — Thu, 23 Jul 2026 13:00:01 +0000

TL;DR

Skills-based hiring still runs through the same ATS, so the real change is evidence per line, not a new format.
Climb the proof ladder: a number beats a named artefact, which beats a named tool, which beats an adjective.
With most applications now AI-drafted, specific verifiable detail is the signal that a real person did the work.

Every recruitment blog is telling you to "highlight your skills" for skills-based hiring, and almost none of them show you what that means on the page. When we were building CVPilot, we kept hitting the same failure: strong candidates whose bullets described duties a parser cannot tell apart from anyone else's. This is the fix, treated like the text-extraction problem it actually is.

A hiring manager reads your CV for roughly seven seconds before deciding whether you make the shortlist. In 2026, what they are hunting for has shifted: they want proof you can do the work, not a list of places you once did it.

Skills-based hiring is the headline of every recruitment conference and half the LinkedIn posts in your feed. The trouble is that most advice about it is vague. It tells you to "highlight your skills" without explaining what that actually looks like on the page.

So let me be specific. Skills-based hiring rewards evidence density, the amount of verifiable proof packed into each line of your CV. This guide shows you how to build that in.

What skills-based hiring actually changes

The promise sounds radical. Employers screen candidates on demonstrable ability rather than degrees, job titles or the prestige of past employers. A self-taught developer competes with a computer science graduate on equal footing.

Parts of that are real. More UK employers have dropped degree requirements from job adverts, and structured skills assessments now sit earlier in the process. If you can do the job, a missing credential hurts you less than it did five years ago.

Here is the contrarian bit that saves you from bad advice. Skills-based hiring mostly still runs through the same applicant tracking system. Your CV is parsed by software before a human sees it, exactly as before.

So the practical change is not dropping your employment history or reformatting everything into a skills matrix. The change is how much evidence you cram into each bullet point. Same structure, far more proof.

Key Takeaway: Skills-based hiring lowers the weight of credentials but keeps the ATS pipeline. Compete on evidence per line, not on a radical new CV format.

Turning a duty into evidence

Most CVs describe responsibilities. "Responsible for managing the social media accounts." That sentence tells a recruiter what you were meant to do, not what you achieved. A duty is a job description; evidence is a result.

The fix is a simple mental swap. For every line, ask what changed because you were there. What moved, by how much, and how do you know. If a claim carries no number and no named outcome, it is decoration, not proof.

Recruiters in 2026 are ruthless about this. With a large share of applicants using AI to draft their applications, generic phrasing has become a warning sign. Specific, verifiable detail is now the single clearest signal that a real person did real work.

Look at the difference when you rewrite a tired duty into evidence:

Duty (weak)	Evidence (strong)
Responsible for managing social media	Grew Instagram following from 4,000 to 21,000 in 11 months, lifting referral traffic 34%
Handled customer complaints	Resolved 40+ escalations weekly, cutting repeat contacts 18% via a rewritten returns script
Improved team processes	Redesigned the onboarding checklist in Notion, reducing new-hire ramp time from 6 weeks to 4
Used data to inform decisions	Built a weekly Looker Studio dashboard that flagged a £12k monthly billing leak

Notice that none of the strong versions are longer than a sentence. Evidence is not about writing more, it is about swapping vague verbs for measured outcomes. You will find this easier once you have a system, and tools like CVPilot can flag which of your bullets still read as duties.

Key Takeaway: Rewrite every bullet so it answers "what changed, by how much." A responsibility with no result attached is filler a skills-based screen will ignore.

The proof hierarchy

Not all evidence is equal. When you strengthen a bullet, you are climbing a ladder of credibility. The higher you climb, the harder your claim is to fake and the more a recruiter trusts it.

Here is the hierarchy, strongest first:

A number. A figure, percentage, timeframe or volume. "Cut load time by 2.3 seconds" beats anything softer because it is measurable.
A named artefact. A specific deliverable you produced: a dashboard, a campaign, a policy document, a shipped feature. It exists and can be pointed to.
A named tool. The specific software or method you used. "Salesforce" or "Kubernetes" is more concrete than "CRM software" or "container tools".
An adjective. "Excellent communicator", "detail-oriented". This is the bottom rung, and on its own it proves nothing.

The strongest bullets stack the top three. A number, tied to a named artefact, built with a named tool, is almost impossible to fabricate convincingly. That is exactly why it reads as authentic.

This matters more than ever because of the authenticity gap. Naming a tool alone is no longer enough, because everyone lists tools; you have to name the outcome the tool produced. "Used ChatGPT" says little. "Used ChatGPT to draft 30 support macros, cutting first-response time by 25%" says you can wield it.

Key Takeaway: Climb the ladder: number over artefact over tool over adjective. Stack the top three in a single line and your claim becomes both specific and unfakeable.

Evidencing skills without the job title

The common worry: what if you never held the title? Maybe you did the analysis but were called a coordinator. Maybe the skill lives in a side project, a volunteer role or a course. Skills-based hiring is designed precisely for people whose ability outran their job title.

The trick is to attach the skill to any context where you genuinely used it. The setting matters less than the result. A measured outcome from a volunteer project counts as real evidence, because the skill was exercised and the number is true.

Consider these sources most people leave off the page:

Volunteering: "Rebuilt a charity's booking system in Airtable, saving trustees around 6 hours weekly."
Side projects: "Shipped a budgeting app to 400 users, maintaining a 4.6-star rating across 90 reviews."
Coursework with output: "Completed a Google Data Analytics capstone analysing 2 years of transit data, presented to a panel of 12."
Internal work outside your remit: "Volunteered to run the team's weekly reporting, a task later folded into the analyst role."

Each of these evidences a skill without needing the formal title attached. What a screen rewards is the demonstrated ability, wherever it happened to be demonstrated.

Key Takeaway: Skills do not require permission from a job title. Pull proof from volunteering, side projects and coursework, as long as each carries a real, checkable result.

Portfolios and links: proof you can inspect

A number on a CV is a claim. A link that lets a recruiter verify that number is proof of a higher order. In skills-based hiring, anything a reviewer can click and inspect carries disproportionate weight.

This is where you widen the gap between yourself and an AI-generated application. A generic bot cannot produce a real GitHub repository, a live dashboard or a case study with your name on it. A working link is the one thing a fabricated CV cannot manufacture.

Add the ones that fit your field. A developer links a repository or a deployed app. A designer links a portfolio. A marketer links a published campaign or a Medium write-up. Place one clean, working URL near the top of the CV where the parser and the human both see it early.

One caution: every link must resolve and must be relevant. A dead URL or an abandoned profile does more damage than no link at all, because it signals carelessness on the exact document meant to prove you are careful.

Key Takeaway: A verifiable link outranks a stated number, because the reader can confirm it themselves. Include one relevant, working URL and test it before you send.

How this interacts with ATS parsing

Here is where the evidence push meets reality. Your beautifully specific bullets still have to survive the parser. An ATS reads text, keys and structure, so evidence that a machine cannot extract is evidence wasted.

Keep the mechanics boring and safe. Standard section headings, a single column, no text buried in graphics or tables the parser might scramble. Skills-based screening often weights how closely your language matches the job advert, so mirror the exact skill terms the advert uses rather than clever synonyms.

There is a neat overlap here. The numbers and named tools that convince a human also feed the keyword matching a machine performs. Evidence-dense writing is naturally ATS-friendly, because concrete nouns and figures are exactly what the parser indexes. If you want the full mechanics, our complete ATS-friendly CV guide walks through formatting in detail.

Before you apply, run your CV through a checker to see the score a real system would assign. CVPilot parses your document the way an ATS does and shows you which bullets landed as evidence and which slipped through as noise.

Key Takeaway: Evidence only counts if the parser can read it. Use plain formatting and mirror the advert's skill terms, and your proof works on both the machine and the human.

Your next hour

You do not need to rebuild your CV from scratch. Open it, and go bullet by bullet with one question: what changed because I was there, and how do I know. Rewrite the three weakest lines first, giving each a number, a named artefact or a named tool.

Then add one working link that proves a skill you claim. That single hour moves you from describing your past to demonstrating your ability, which is the whole point of skills-based hiring.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/skills-based-hiring-rebuild-cv-proof?utm_source=devto&utm_medium=organic&utm_campaign=skills-based-hiring

Which of your CV bullets would a parser file as evidence, and which would slip through as noise?

Full guide: https://cvpilot.pro/blog/skills-based-hiring-rebuild-cv-proof

Why Your CV's Skill Bars Are Invisible to the Parser (and What Actually Gets Indexed)

Hanzala Mehmood — Wed, 22 Jul 2026 13:00:01 +0000

TL;DR

An ATS strips your CV to plain text, so skill bars and star ratings are read as shapes with no value and never get indexed
Hard, named skills such as Python, SQL and Xero get tokenised and matched, while soft-skill strings match with near-zero ranking weight
Mirror the advert's exact wording once, cap the list at ten to fifteen skills, and evidence every one in your experience bullets

When you build a CV parser, the first thing you learn is how little survives the extraction step. Fonts, colours, progress bars and rating dots all get stripped away, leaving a flat token stream that gets matched against the job description. I spent a long time on exactly this pipeline while building CVPilot, and the skills section is where the gap between what candidates write and what the machine actually reads is widest.

Recruiters love to say the CV skills section is where they look first. The truth is more brutal: before a human sees it, an applicant tracking system has already scanned, sorted and scored your list of skills against a job description you never got to read in full.

Most candidates treat this section as a decorative afterthought. They pile in adjectives, add a few coloured bars and hope for the best. That approach quietly costs interviews every single week.

Let me walk you through what a parser actually extracts from your skills section, what it silently discards, and how to phrase things so both the software and the hiring manager give you credit.

How a parser finds and reads your skills section

An ATS does not read your CV the way you do. It ingests the file, strips it back to plain text, then looks for structural signposts such as headings, line breaks and recognisable keywords. Your section heading is the first thing that either helps or hinders the machine.

Call it "Skills" or "Technical Skills" or "Core Competencies" and the parser knows exactly what follows. Call it "What I Bring to the Table" and you have just hidden the whole section behind a heading the software does not recognise. Clever headings are where good candidates lose easy points.

Once it finds the section, the parser tokenises each entry and matches it against the terms pulled from the job advert. A clean, comma or bullet separated list reads perfectly. A skills grid built as a graphic often reads as nothing at all.

For a fuller picture of how the whole document is parsed, our complete ATS-friendly CV guide covers formatting from top to bottom.

Key Takeaway: Use a plain, conventional heading and a simple text list. The parser rewards structure it recognises, not creativity it cannot categorise.

Hard skills score. Soft skills mostly do not.

Here is the distinction that changes everything. Hard skills are specific, nameable and verifiable: Python, Xero, SQL, HGV Class 1, Six Sigma, French. These are the terms recruiters put into their search filters, and they are the terms the ATS scores you against.

Soft skills are traits: communication, leadership, problem solving, teamwork. A parser can detect the word "leadership" sitting in your skills list, but it cannot measure whether you actually have it. It can match the string, yet the match carries almost no weight in a ranking.

This is why "excellent communicator" scores nothing useful. Every applicant claims it, no recruiter searches for it, and no software can validate it. You have spent a line of your CV on a phrase that moves no needle.

Keep your named, technical, tool based skills in the skills section where they get scored. Then prove your soft skills through outcomes in your experience bullets, which is the only place they become believable.

Key Takeaway: Load the skills section with hard, searchable skills. Demonstrate soft skills through results elsewhere, because the parser cannot score a personality trait.

Skill bars, star ratings and graphics are invisible

This is the contrarian point that saves people the most space. Those five-star rating dots and the little progress bars that show "JavaScript 90%" convey literally nothing to a parser and nothing verifiable to a human.

To the software, a bar filled to 70 percent is an image or a shape with no text value. It is not read, not scored, not indexed. You have decorated a data point the machine simply skips.

To a person, the rating is meaningless because you assigned it yourself. Your 90 percent Excel and my 90 percent Excel could be worlds apart, and any experienced recruiter knows it. Self-scored ratings signal padding, not proficiency.

Delete the graphics. Replace them with plain words and, where it matters, a short qualifier such as "advanced" or a real anchor like "built department reporting in Power BI". Specificity beats decoration every time.

Key Takeaway: Skill bars and star ratings are wasted space. Swap them for plain-text skills with concrete, evidenced qualifiers.

Skill listed versus how the ATS reads it

The gap between what you write and what the software registers is often wider than people expect. The table below shows common entries, how a parser handles them, and a stronger rewrite.

Skill as written	How the ATS reads it	Stronger phrasing
Excellent communicator	Matches a soft-skill string with near-zero ranking weight	Move to experience: "Presented monthly figures to a board of six"
JavaScript (90% bar)	Bar ignored; only the word "JavaScript" is indexed if it is real text	JavaScript, React, Node.js
Microsoft Office	Vague match; misses the specific tools recruiters search for	Excel (pivot tables, VLOOKUP), PowerPoint, Outlook
Team player	Unsearchable trait, no scoring value	Move to experience: "Led a cross-functional team of eight"
Good with data	Too generic to match a named skill filter	SQL, Tableau, data cleansing
Fluent English (as a graphic flag)	Flag image skipped entirely	Languages: English (native), Spanish (B2)

Key Takeaway: Trade vague adjectives and graphics for named tools and specific competencies. That is the language a parser and a recruiter both understand.

Where to place the section and how many to list

Placement depends on your field. For technical, engineering and IT roles, put the skills section near the top, just under your summary, because the tools are the qualification. For most other roles, a position after your experience works better, since your track record carries the argument.

On length, resist the urge to list everything you have ever touched. A tight block of ten to fifteen genuine, relevant skills reads as focus, while a wall of forty reads as noise. Recruiters skim, and parsers weight relevance, so padding dilutes both.

Group related skills if the list is long. A short "Technical", "Analytics" and "Languages" split is easier to scan than one undifferentiated jumble, and it keeps each heading keyword rich.

Key Takeaway: Place skills high for technical roles, lower for others, and cap the list at ten to fifteen relevant entries. Focus beats volume.

Mirror the advert, once, in its exact words

ATS matching is literal. If the advert asks for "stakeholder management" and you wrote "managing stakeholders", a strict parser may not connect the two. Copy the advert's exact phrasing for the skills you genuinely have.

Read the job description and pull out the named tools, methods and qualifications. Mirror each relevant term once in your skills section, spelled and cased as they wrote it. Once is enough: repeating a keyword five times is the old spam trick that modern systems and humans both penalise.

If you want a shortcut on which terms carry weight, our guide to the ATS keywords that actually matter shows how to spot the ones worth mirroring. A tool like CVPilot will also flag the exact terms your CV is missing against a specific advert.

Key Takeaway: Match the advert's precise wording once per skill. Literal mirroring wins the match; keyword stuffing loses the reader.

Never list a skill you cannot evidence

This is the trap that catches confident candidates. You add "project management" to your skills because it sounds strong, but nothing in your experience bullets mentions running a project. An alert recruiter reads that gap as a small lie, and it colours the whole CV.

Every skill in your list should have a fingerprint somewhere in your work history. If you claim SQL, a bullet should show you querying a database. If you claim negotiation, a bullet should show a deal you closed or a contract you improved.

Treat your skills section as a table of contents and your experience as the chapters. A skill with no supporting evidence is the first thing an interviewer will probe, and the fastest way to lose credibility in the room.

Key Takeaway: Only list skills your experience bullets can back up. An unevidenced claim is an interview landmine, not an asset.

AI tools on your CV in 2026: name the tool and the outcome

AI fluency is now a genuine differentiator, and recruiters have started filtering for it. But listing "AI" or "ChatGPT" on its own reads as a buzzword, and it tells a hiring manager nothing about what you can do. The winning formula pairs the specific tool with the measurable outcome.

Do not write "used AI tools". Write "used GitHub Copilot to cut code review time by roughly a fifth" or "built a Claude prompt workflow that halved first-draft turnaround on client reports". Name the tool so the parser matches it, and state the result so the human believes it.

The same rule applies across functions. A marketer might list "Jasper and Midjourney for campaign asset production", an analyst "Python with an LLM API for automated report summaries". Specific tool plus specific gain is what separates a real practitioner from someone chasing a trend.

Running your draft through CVPilot will show you which of these phrasings actually register against a live job description, before you send anything.

Key Takeaway: List AI skills as tool plus outcome, never as a lone buzzword. Name the software for the parser and quantify the result for the human.

Putting it together

Your skills section is not a personality showcase. It is a structured, machine-read index of the hard, named, evidenced capabilities that qualify you for one specific job. Strip the graphics, drop the empty adjectives, mirror the advert and prove every claim.

Get that right and you clear the software filter and earn a recruiter's trust in the same few lines. That is the whole point of the section, and most candidates never use it that way.

Ready to optimise your CV? Try CVPilot free and see your ATS score in under 60 seconds.

Read the full guide: https://cvpilot.pro/blog/cv-skills-section-what-ats-reads?utm_source=devto&utm_medium=organic&utm_campaign=cv-skills-section

What is the worst false positive you have hit when pulling structured fields out of a messy user-supplied document?

Full guide: https://cvpilot.pro/blog/cv-skills-section-what-ats-reads