Recruitment Website Audit: A Practical Framework for Careers Sites That Convert

Most recruitment websites are leaking candidates, and nobody is measuring the leak.

The careers page gets a redesign every three years. The job ad template hasn’t changed in five. The application form is six pages long because that’s what the ATS spat out at go-live. And every quarter, someone in TA gets asked why cost-per-hire keeps creeping up.

A recruitment website audit fixes that — but only if you audit the right things in the right order. After 20 years auditing recruitment systems for in-house TA teams, agencies and HR functions, I can tell you that most “audits” stop at the bits that are easy to score (page speed, broken links) and skip the bits that actually move the application rate.

This is the framework I run. Use it on your own site this week. The bottom of this post explains where the £499 SetpointHQ audit fits if you’d rather have it done properly.

What a Recruitment Website Audit Actually Covers

A recruitment website audit isn’t a page-speed report with a logo on it.

It’s a systematic look at every point where a candidate touches your hiring system — from the first time they meet your employer brand on Google or LinkedIn, through to the moment they hit submit on the application form. Anywhere they drop off, the audit names it, costs it, and gives you a fix.

A useful audit answers three questions:

If your audit doesn’t answer all three, it’s documentation — not diagnosis.

The framework below covers seven areas. The first six map to the deeper SetpointHQ audit. The seventh is the one almost no one is checking yet, and is also the cheapest organic visibility win available to recruitment teams in 2026.

The 7-Area Framework

1. Job Ad Performance and Channel Mix

Start where the candidate starts.

Pull the last 90 days of application data and break it down by source. For each role, you want: applications, qualified shortlist, hires, and cost per applicant by channel. If you can’t produce that table in 30 minutes, that’s the first finding.

Then look at the ads themselves. Are they written for the candidate or for the hiring manager? Are they readable on a phone in 30 seconds? Are they ranking for the role title in Google, or are aggregator sites outranking you for your own jobs?

Common findings: 60–70% of paid spend going to one channel that converts at half the rate of an underused alternative. Job ads written like internal job descriptions. Zero on-site SEO for “[role] jobs in [city]” terms.

2. Employer Brand and Candidate Experience

Walk the candidate journey end to end. Phone first. Desktop second. Don’t log in.

Time how long it takes to find a job that matches a real person you’d want to hire. Time the application. Count the clicks. Note every form field that’s required but adds no value. Note every dead end — the careers page that links to a careers blog that links nowhere.

Cross-check what your site says about your culture against what Glassdoor, Indeed reviews and LinkedIn comments say. Mismatches between the two are the single biggest reason qualified candidates ghost between offer and start date.

Common findings: a 12-field application form when a 4-field one converts twice as well. A careers page that ranks number one for the company name but never gets sent to candidates because the recruiters use a Greenhouse link instead. A “values” section that nobody internally believes.

3. Hiring Infrastructure and Process

This is the unglamorous part. The wiring.

Audit your ATS configuration: are stages set up to measure conversion or just to satisfy reporting? Are screening questions doing the screening, or are recruiters re-doing the work in spreadsheets? Are rejection emails sent within 48 hours, or is your time-to-reject longer than your time-to-hire used to be?

Map the path from application to offer. Count the handoffs. Every handoff is a place where a good candidate can disappear because someone forgot to email them back.

Common findings: candidates sitting in “review” status for 11 days because nobody owns the queue. Five interview stages for roles that should have three. Hiring managers using personal calendars to schedule, recruiters using a different one to confirm.

4. Talent Pipeline and Candidate Ownership

Who owns the candidate after they apply, and what happens to the ones you don’t hire?

If the answer is “nothing”, you’re buying every candidate twice. The good silver-medal candidate from six months ago is still looking — but you’ll pay an agency to find them again because nobody nurtured the relationship.

Audit your CRM. Audit your talent communities. Audit your alumni database. If any of those phrases made you laugh, that’s the finding.

Common findings: 8,000 historic applicants in the ATS, zero communications sent to them in the last 12 months. No segmentation by role, level or location. No automated re-engagement when a similar role opens.

5. Spend Efficiency and ROI

The number that matters: fully-loaded cost per hire.

Not just agency fees. Recruiter time at fully-loaded cost (we use £28 per hour as the practical UK benchmark for an in-house recruiter — salary, NI, pension, software, overhead). Job board spend. Internal referral payouts. Hiring manager time. Onboarding. ATS licence cost prorated to hires.

Compare it to the cost of getting it wrong: an open role at typical UK salary costs roughly £400–£700 per day in lost productivity. Slow hiring isn’t free. Fast bad hiring isn’t free either.

Common findings: a true cost per hire that’s two to three times the figure being reported in the board pack, because the board pack only counts external spend.

6. Competitive Positioning

Your candidates are comparing you to your competitors before they apply. Your audit should too.

Pick three direct competitors for talent — not your business competitors, your talent competitors (often different). Run the same candidate journey on their sites. What do they pay? What do they say about flexibility? How fast does their application form load? Whose careers page ranks higher in Google for “[role] jobs”?

Then audit your messaging. If you removed the logo from your careers page and your competitor’s careers page, could a candidate tell which company they were on?

Common findings: identical messaging to three competitors (“we’re a great place to work, our people are our greatest asset”). No defensible employer value proposition. No data on what your competitors actually pay for the same role.

7. AI Search Visibility — The Area Nobody Is Auditing Yet

Candidates don’t only use Google any more. Increasingly they ask ChatGPT, Claude or Perplexity “is [your company] a good place to work?” before they ever land on your site. The answer they get comes from somewhere — usually Glassdoor, Reddit and old press coverage. Rarely from you.

This is the cheapest organic visibility win available to recruitment teams right now, because almost nobody is competing for it.

What to check:

• AI crawler access. Can GPTBot, ClaudeBot and PerplexityBot actually crawl your careers site? Many robots.txt files block them by accident.
• llms.txt file. Do you have one at the root of your site, telling AI models what your careers content is for?
• Citation-ready content. Is your content structured in 134–167 word self-contained passages — the format AI models cite from?
• Answerable questions. Does your careers content surface clear, citable answers to common candidate questions — pay ranges, benefits, hiring process, time-to-hire?
• Organization schema. Is it published with up-to-date employee count, locations, and sameAs links to your social profiles?

If you fix nothing else this quarter, fix this. The competitive window won’t stay open for long.

How to Prioritise What You Find

A long list of findings is not a plan.

Score every finding against three questions:

Plot them on a 2×2 of impact versus effort. Fix everything in the high-impact / low-effort quadrant in the first 30 days. Schedule the high-impact / high-effort items into the next quarter with a named owner. Park the low-impact items unless they’re a brand or compliance risk.

The single most common audit failure isn’t finding the problems — it’s producing a 40-page report that nobody actions because nothing was prioritised.

DIY or Done-for-You: Which Makes Sense for You

You can run this framework yourself. Most TA teams that do, get partway through area three before the day job swallows the project.

The reasons to bring in an outside audit are practical, not strategic:

• You don’t have the time. A proper audit is 4–6 days of skilled work — not something you can squeeze between requisitions.
• You’re too close to the system. The obvious problems become invisible when you’ve worked around them for two years.
• You need an external recommendation. Sometimes the only way to unlock budget for fixes is to have someone outside the function name the problem.
• You want a market benchmark. Your own historical performance is a low bar. An external view brings comparison data you can’t get from inside.

If any of those apply, the SetpointHQ Recruitment Marketing Audit is the version of this framework I run for clients. £499 (rising to £999 after the first ten places). Six areas audited, one prioritised action plan, one 60-minute debrief call. Delivered in seven working days.

If you want the seventh area — AI search visibility — included as a self-serve check before you commit, the Careers Site Auditor inside SetpointHQ Studio runs that audit on any URL in under three minutes.

Frequently Asked Questions

What is a recruitment website audit?

A recruitment website audit is a systematic review of every point where a candidate interacts with your hiring system — from the careers page they land on, through to the moment they submit an application. A proper audit names where you’re losing candidates, quantifies the cost in pounds and hires, and prioritises what to fix first. It’s different from a generic SEO audit because it’s scored against recruitment outcomes, not just technical metrics.

How long does a recruitment website audit take?

A DIY audit using a structured framework takes a competent recruitment marketer 4–6 days of focused work to complete properly. Most in-house teams that start one don’t finish, because the day job interrupts the analysis. The SetpointHQ Recruitment Marketing Audit is delivered in seven working days from kick-off, including the prioritised action plan and a 60-minute debrief call.

What does a recruitment website audit cost?

Outside agency audits typically run from £2,500 to £15,000 depending on scope and depth. The SetpointHQ Recruitment Marketing Audit is £499 for the first ten beta places, rising to £999 thereafter. It covers six areas — job ad performance, employer brand, hiring infrastructure, talent pipeline, spend efficiency and competitive positioning — with a written report and debrief call included. AI search visibility checks are available separately through the Careers Site Auditor in SetpointHQ Studio from £399 per month.

What’s the difference between a careers site audit and a recruitment website audit?

A careers site audit usually focuses on the technical and content health of the careers page itself — page speed, indexability, schema, on-page SEO, mobile usability. A recruitment website audit is broader: it includes everything in a careers site audit, plus the candidate journey from search to application, the ATS and hiring process behind the form, the channel mix that drives traffic, and the cost per hire that comes out the other end. If you only audit the careers page, you only fix the careers page.

Can I run a recruitment website audit myself?

Yes — the seven-area framework above is the same one we run with clients. Most TA teams can complete the first two or three areas internally without specialist help. The areas where in-house teams typically need outside support are spend efficiency (because the data is fragmented across finance, ATS and job boards) and competitive positioning (because it requires sustained external research). The Careers Site Auditor in SetpointHQ Studio automates the seventh area — AI search visibility — and runs in under three minutes.