Search
Search work has a bad name, and it’s earned it. There are a lot of monthly reports out there full of things that were done and nothing that happened. So I’ll say how I do it, and then I’ll show you what it did.
The first thing is to work out which searches are actually worth having, and I mean the ones that lead to work rather than the ones with the biggest numbers. On the Claims Bible site there are three medical negligence searches that showed the site 47,280 times over sixteen months and produced six clicks, and there are 48 pages written for that section that I’ve parked, because there’s no buyer for it and the traffic would be worth nothing. Knowing what not to chase is most of the job.
The second thing is to fix the structure before the content. Titles and descriptions that are the right length and say the right thing, pages that aren’t competing with each other for the same search, old addresses that still work, and a site that’s fast enough that search engines bother with all of it. On Claims Bible that was 125 titles and 163 descriptions trimmed in one pass, 480 keywords flagged where more than one page was competing for the same search, and seven holes in the redirects fixed that had been sending roughly 470,000 impressions’ worth of personal injury searches to dead pages.
The third thing is to measure whether it worked, and to say so either way. I track a set of keywords and I write it up weekly. When I got one of those write-ups wrong, and blamed a 32% traffic drop on a crawl problem the evidence didn’t really support, I wrote the correction into the same record. You’d want to know that about the person reading your numbers.
What it did
For the Claims Bible relaunch in August 2026, search clicks went from 484 a week to 611 in the week after cutover, up 26%. The week after that they came back to 443, so it was a one-week lift rather than a step change, and over the 18 days after cutover against the 18 before they were up 7%. Visibility on the tracked keyword set went from 28% on the day before cutover to 40% by 8 September, and passed MoneySavingExpert on 1 September, which is the result I’d point to because it kept going after the click count settled. 332 keywords were improving against 168 declining in that week, and 78 went into the top ten for the first time.
For Express Encapsulation, a Bristol trade laminator, the site went to number one on Google for “print encapsulation” after the 2018 rebuild and is still there on 8 September 2026.
There’s also one that’s about conversion rather than traffic. Claims Bible had 40 credit card pages that were being shown 53,625 times and clicked 45 times, a rate of 0.08%, with nothing on them for a visitor to do. I built a set of refund calculators into the site and put them on the pages people were arriving at. Coverage went from 11 pages to 51, and in the same period the site’s key event rate went from 2.28% of sessions to 5.07%, though I wouldn’t put all of that down to the calculators.
The local search version
For businesses with locations, the search that matters is the one with a map on it. In August and September 2026 I audited 16 UK franchise networks and kept finding the same thing. The brand ranked nationally, but its individual locations were missing from the map results in their own towns, with the free spots going to local independents, and in several cases the brand was paying for ads on searches its locations should have owned for nothing. That’s a fixable problem and it’s cheaper to fix than to keep paying for.
How I work
Search Console, SEMrush, Screaming Frog and GA4, with the analytics set up so consent’s handled properly and the key events are the ones that mean something to the business. The technical fixes are built into the site rather than bolted on, because I build the sites too. And the write-up is honest. If a month was flat I’ll tell you it was flat and why I think so.
The proof
The case study is the Claims Bible rebuild, and the longer-run example is Express Encapsulation.
“Since completing the work, plus while working on other long-term projects with us, our search visibility has improved, the website performs better, and our IT systems are far more reliable.”
Chloë McGuire at Express Encapsulation and Laminationin 2026
Who this is for
Businesses that are paying for search work and can’t say what it’s produced, businesses that get traffic and no enquiries, and multi-location businesses and franchise networks whose brand is visible and whose branches aren’t.
Send me the address of the site and tell me what you’re currently paying for, and I’ll tell you what I’d do differently.
Tell me what’s not working.
I’ll tell you whether I can help and roughly what it would involve.