Guides
There is no shortage of general SEO advice on the internet and we have no intention of adding to it. What goes here are problems we have genuinely diagnosed on live sites, written up properly, including the parts where the first two theories were wrong.
Currently being written
11 guides are published so far, and we would rather show you an honest short list than pad it out with filler. The two below are actually being written. If one of them would be useful to you sooner, say so and we will bump it up the list.
Published
Published
The hub for the Page Indexing report. What each not-indexed status actually means, grouped by the only question that matters: is it Google working correctly, or a real problem to fix. Links to the full write-up on every status.
The Search Console statuses series
Read the guidePublished
Multi layer invalidation across the CDN edge, OPcache, Redis and application transients. Including the failure mode almost nobody checks: curated pages that query content rather than being taxonomy archives, so no term based purge can reach them.
From the cache invalidation work
Read the guidePublished
A login page absorbing 46% of origin requests at 21,626 hits a day. How to check whether the attack is actually slowing the site, stop it at the edge in minutes with an admin allowlist, and avoid paying for a bigger server you do not need.
From the 21,626 hits a day case
Read the guidePublished
A DR 72 page ranking for zero keywords while DR 16 to 32 competitors held page one. Why authority was never the constraint, and the four on-page causes that actually held it down: a canonical pointing at a redirect, split intent, spun copy and a 25 link block.
From the DR 72 diagnosis
Read the guidePublished
Google knows the URL exists and has chosen not to request it yet. Nothing on the page caused it, because nothing on the page has been read. Why Google rations crawl, what genuinely moves it (led by internal links from pages already crawled), and the uncomfortable case where it is not a fault to fix.
From the Search Console diagnoses
Read the guidePublished
Google fetched the page, read it, and declined to index it. That is a judgement about the content, not a technical fault, which is why resubmitting never fixes it. The five causes in the order that finds it fastest, including when the honest answer is to delete the page.
From the Search Console diagnoses
Read the guidePublished
Google found duplicate URLs, looked for a canonical telling it which one counts, found none, and picked one itself. The version it picks is frequently not the one you would have picked. Where the duplicates come from and how to take the decision back.
From the indexation work
Read the guidePublished
The server says 200, the page says nothing is here, and Google believes the page. The five patterns that cause it, how to test for it in one command, and which status code each case should return instead.
From the status code audits
Read the guidePublished
The status people panic about most and need to act on least. Usually it is Google honouring the canonical you set. The four cases where it genuinely hides a problem, starting with the canonical pointing at a redirect that we found on a live site.
From the canonical diagnosis
Read the guidePublished
A repeatable way to find how ChatGPT, Gemini, Perplexity and AI Overviews describe you, without a tool or a subscription. Write the questions customers actually ask, run each several times logged out, and record presence, the exact wording, who appears alongside you, and what was cited.
A method you can run in an hour
Read the guidePublished
The redirect map built from every URL, canonical and internal link continuity, the content parity check, the staging controls that tank a launch, and the post-launch monitoring everyone skips. In the order that actually protects the rankings.
From real migrations
Read the guidePlanned
In progress
A short, practical method. Hit the origin directly with a cache buster and compare. It takes two minutes and it decides which half of the stack you spend the next two days in.
A method we use on nearly every diagnosis
In progress
Getting @id relationships right, when to consolidate duplicate organisation entities, and which markup to remove rather than fix. Includes why we no longer add FAQPage markup at all.
From the Search Console entity work
Approach
Every guide here comes from a problem that appeared on a real site, was diagnosed properly and was fixed. That constraint rules out a lot of topics, which is the point.
They include the wrong turns. When a problem has four plausible explanations and the first two are wrong, that process is the genuinely useful part, and it is the part that usually gets edited out of tidy write ups. Knowing which check ruled out which theory is worth more than the final answer.
Where a fix carries a risk or a trade off, it says so. Some of these changes are ones we would not make on every site, and a guide that does not tell you when not to do something is not much of a guide.
No affiliate links, no tool sponsorships, and no padding to hit a word count.
Keep reading
Book me
If you are sitting on the exact issue one of these covers, do not wait for us to write it up. Send us the URL and we will just tell you.
End to end
This is drawn from real client work: building sites from scratch, ranking them, earning the links and keeping them fast and safe. If you would rather have a part of that done than read about it, it is one team from end to end.
Start with a free 10-minute diagnostic