Recruitment Agency SEO
I ran SEO for one of the world's largest recruitment agencies.
I was contracted as an SEO Lead for Hays, managing search across a 35-plus website estate spanning 33 countries. It's an engagement that shaped how I think about SEO at scale - and recruitment websites are among the most technically complex SEO projects on the commercial web.
A major agency isn't one site. It's often dozens: country market sites, specialist sector brands, editorial properties and employer-facing service pages. All built independently, over years. All competing with each other for the same searches.
Websites across 33 countries in the Hays estate
Hays via CTI DigitalMarkets taken through Keep/Kill/Consolidate content reviews, in multiple languages
Hays migration programmePages on the flagship market site carrying a broken canonical tag at discovery
Discovery auditAverage broken internal links across one specialist property at discovery
Discovery auditThe estate is usually fighting itself
At Hays, four separate properties - the flagship market site, two specialist solution brands and an editorial brand - were competing for overlapping audiences and keywords. The editorial brand's traffic barely crossed over to the market sites it was supposed to support. Candidates and clients were being served from the same pages. That isn't a content problem you can write your way out of. It's an architecture problem that needs structural decisions.
Candidates and clients are different audiences
A recruitment site serves two completely different users - candidates looking for jobs and clients looking for recruitment services. They need different content, different landing pages and different conversion paths. Most recruitment sites blur them. A clear SEO architecture separates them.
Job pages have their own technical rules
JobPosting schema drives Google for Jobs rich results - a significant traffic source for listings. Pagination on job hubs needs careful handling to avoid cannibalising the hub. And stale job URLs need a policy: redirect, remove or keep live? Thousands of dead job pages left indexed is a crawl budget problem and a user experience problem at once.
The Keep/Kill/Consolidate methodology
The most replicable output from the Hays engagement is a content review method that's now core to how I run any large migration. Every URL in every market gets one of three classifications: Keep (strong enough to migrate as-is), Kill (remove, with a redirect to the most relevant live page) or Consolidate (merge underperforming pages on the same topic into one stronger URL).
I applied it across Belgium, France, Canada, Australia, Hong Kong, Czech Republic, Spain, Chile and the UK, in multiple languages. The output per market is a structured workbook mapping every URL to its classification, redirect destination and rationale. The full sequence runs Scope, Kill, Consolidate, Map, Benchmark, Test.
What the discovery audit found
The structural issues at discovery were significant: broken canonical tags on roughly one in four pages of the flagship market site. Thousands of broken internal links - averaging three per page - on one specialist property. Hreflang implemented incorrectly across multiple properties. Hundreds of pages rated poor or needs-improvement on Core Web Vitals, with only a handful rated good. None of this was a surprise - it's what accumulates on a complex estate when SEO isn't embedded in the product process. The migration was the chance to fix it at the root.
If your market sites are competing with each other for the same candidates, that's not a content problem - it's an architecture problem. I've fixed it at the largest scale this sector has to offer, and the methodology transfers to any multi-site recruitment estate, whatever its size.