Thirty problems from real projects and how to work through them with JMX. Every article answers a question somebody actually asks, names the steps through the application one by one, and says at the end where the tool stops.
The text is in the browser, not in the source. How to separate the ranking problem from the GEO problem with JMX: the GEO tab, the Rendering tab, the counter-check.
Read on JavaScript site: what Google sees and AI crawlers do not
PageSpeed Insights reports 98, Search Console reports “poor”. Both are right: lab and field measure different things. How to separate them in JMX.
Read on PageSpeed 98, Search Console “poor” — which number counts?
The crawl shows what is on the site, the server log what search engines fetch of it. How to evaluate both against each other in JMX — limits included.
Read on What Googlebot actually fetches: log files against the crawl
Where the quickly liftable points hide in 18,000 Search Console rows: Within reach, weak click-through rates and Reach without foundation in JMX.
Read on Search Console: finding the wins that cost little effort
How to use JMX's spell check to find typos across the whole crawl, separate specialist vocabulary from them — and where the method ends.
Tracking down images without an alt attribute and files that are too heavy with JMX: the seven image rules, the right tab for each step, and the limits of the check.
Read on Finding images without alt text and image files that are too large
When visitors see the wrong language version, it is rarely down to one page. How to check hreflang clusters with JMX — and where the check ends.
Read on The wrong language version: checking hreflang with JMX
A nightly jmx-cli run compares the new crawl with the previous one and turns the CI job red as soon as a page answers 404 or carries noindex.
A strong backlink profile is worth nothing if the links point at 404s, redirect chains or noindex pages. How to find those places with JMX.
Read on Finding backlinks that point at error pages and redirects
A list of findings says what is wrong. The client asks what to start with and what it costs. How to turn a crawl into a document that answers that question.
Read on From the findings export to a report the client understands