What good career page analytics actually tell you

Almost every company measures visits to its careers site. Very few measure whether those visitors actually do anything useful.

Bounce rate and page views tell you something about reach. They tell you nothing about why qualified candidates abandon your application form halfway through, why visitors click a job listing but never read the full description, or why your careers site converts well for one role type but fails completely for another.

At Livewall, we build working-at websites and employer brand campaigns for organisations that take their talent position seriously. What we see repeatedly: teams making changes to their recruitment site based on traffic data, when it is behavioural data that actually reveals what is going wrong.

The metrics that actually matter

Start with the question: what does a visitor need to do to become a qualified applicant? That path has steps. Each step has drop-off. You want to know where drop-off happens and for which type of visitor.

Job view to application start rate. How many people click a vacancy but never start the application? If this is high, the problem is in the job description or the barrier to the form. Not in your recruitment campaign.

Form abandonment rate per step. Most ATS systems do not track this by default. Build custom events in via Google Tag Manager or a similar tool. If people drop off at question 4 of 7, you know exactly where the friction is.

Time on page per job category. Candidates for operational roles behave differently from candidates for specialist positions. If someone spends five seconds on a job page, that page did not communicate what it needed to communicate.

Returning visitors. Strong candidates research multiple times before applying. A high proportion of returning visitors who still do not convert signals hesitation, not lack of interest. That is a very different problem to solve.

Connecting behavioural data to application quality

The real insight comes when you link front-end behaviour to what happens inside your ATS. That sounds technical. It is honestly not as complicated as it seems.

Set up UTM parameters for every channel and every campaign. Make sure those parameters carry through to your application form. That way you can measure not just volume per source, but conversion quality: what percentage of applicants from each source make it through initial screening?

If LinkedIn sends ten times more traffic than your own job boards, but your direct traffic produces twice as many suitable candidates, you know exactly where to invest.

This is what we mean by analytics that actually tell you something. Not more visitors. Better decisions.

Heatmaps and scroll depth: the underused tools

Quantitative data tells you where people stop. Qualitative tools tell you why.

A heatmap on your vacancy pages shows which elements attract attention and which get skipped entirely. If candidates scroll to the benefits section but ignore the apply block, the visual hierarchy of your career page design is the problem.

Scroll depth analysis shows how far people get on your job description pages. If most visitors see less than 40% of the page, your content is probably too long, too generic, or too slow getting to the point.

Combine this with session recordings for pages with high exit rates. You do not need to watch hundreds of recordings. Fifteen recordings on the page where people leave will tell you enough to start making changes.

A measurement setup you can build in two weeks

You do not need to wait for a new platform or a big analytics project. Here is what you can do now:

1. Set up goals for application start and completion in Google Analytics 4. Use events, not pageviews.
2. Add UTM parameters to all campaign links and make sure they carry through to your ATS.
3. Install a heatmap tool on your three most visited vacancy pages.
4. Build a weekly report with five numbers: visits, job views, application starts, completions, and channel breakdown per application start.
5. Compare by job category, not just overall. Operations roles, customer-facing jobs, and management positions have different candidate journeys.

After four weeks you will have enough data to know where your biggest leak is. That is the starting point for a working-at website that actually performs.

Where analytics fall short

No analytics dashboard tells you how a candidate felt during the application process. Quantitative data has a blind spot for experience.

That is why at Livewall we always pair behavioural data with candidate research. Short exit surveys at the drop-off point. Interviews with recently hired employees about their orientation process. A/B tests on the page elements you suspect are causing problems.

Data tells you what is happening. Research tells you why. Together they form the basis for improvements that actually lead to more applications and better ones.