RankMerlin blog
Get found. By both.
Plain-English guides on classic SEO and AI visibility: what the checks mean, why sites fail them, and exactly how to fix each one.
What is AI visibility, and why your site probably scores low
People increasingly ask AI assistants instead of searching. AI visibility is whether those assistants can find, read, and cite your site. Most sites fail it without knowing.
Read the post
An SEO plugin is installed. Nothing is configured.
Half the WordPress sites we scan have RankMath or Yoast active and almost nothing filled in. The plugin is an editor, not an author. Here is what actually needs doing.
Read the post
llms.txt: what it is and how to add one in 10 minutes
One markdown file at your domain root tells AI systems what your site is, what you sell, and where to look. The format, a template, and where to put it.
Read the post
Are you blocking GPTBot without knowing it?
We audited our own site and found Cloudflare quietly serving 403s to every AI crawler. The two settings responsible, how we found them, and the score before and after.
Read the post
JSON-LD for small business: the four schema types that matter
Structured data sounds enterprise. It is actually a paragraph of JSON that tells machines what your business is. The four types that cover a typical site, with examples.
Read the post
Answer-shaped content: write so AI assistants can quote you
Assistants do not summarize your brand story. They lift direct statements. The question-heading plus direct-answer pattern, and how to retrofit it onto existing pages.
Read the post
Titles and meta descriptions: the 60-character sales pitch
The title tag is the most-read sentence you will ever write about each page, and the description is its ad copy. Formulas, length rules, and the duplicate trap.
Read the post
FAQ blocks: one section, two search engines
A good FAQ section feeds classic search and AI assistants with the same few paragraphs. What to ask, how to answer, and the schema that makes it machine-readable.
Read the post
Entity clarity: does anyone know who you actually are?
Machines cannot infer from your hero image that you are a Warsaw tattoo studio. Name, place, and offer need to be stated in text. The NAP test and where sites fail it.
Read the post
Thin content: how 200-word pages drag down your whole site
Search engines judge sites, not just pages. A tail of empty service stubs and auto-generated archives tells them most of your site is not worth indexing. Find and fix the tail.
Read the post
robots.txt and sitemap.xml in plain English
Two small text files decide how machines walk your site. What each one does, what a healthy version looks like, and the copy-pasted rules that quietly hide sites.
Read the post
SEO vs GEO: what carries over and what changes
Generative engine optimization is not a replacement for SEO. It is a second scoreboard on the same foundation. What transfers directly, what is genuinely new, and what to ignore.
Read the post