Skip to content
DevMarDe
Marketing · AI visibility · WordPress

AI visibility for WordPress

Most SMB marketing sites and a large share of clinic, school and professional-services sites. Get named when someone asks ChatGPT, Perplexity, Claude or Gemini about your business. On WordPress specifically, this is what changes.

Run the audit first About ai visibility
crawlaudit.devmarde.com/r/8f2aCrawl Audit · Findings
FindingImpact
GPTBot blocked by firewall403 from bot protection on every pageHigh
llms.txt missingNothing tells an AI crawler what the site is forHigh
Entity records unlinkedNo sameAs on Organization schemaMedium
31 pages without dateModifiedModels discount content they cannot dateMedium
200
signals scored
4
crawlers tested
6
assistants tracked

What WordPress gets right, and what it does not

  • Security plugins such as Wordfence and hosts such as WP Engine often block unknown user agents by default; GPTBot returns 403 and nobody notices.
  • Yoast and Rank Math emit Organization schema but rarely include sameAs or a stable @id.
  • dateModified is emitted, but only updates when the post is saved, so bulk edits inflate freshness.

AI visibility on WordPress

  • llms.txt. Upload llms.txt to the site root over SFTP, or serve it with a static-file plugin. Confirm it returns text/plain.
  • Schema. Add the JSON-LD through the theme's header hook or a code-snippets plugin rather than a second SEO plugin, so there is one Organization block.
  • Crawlers. Settings → Reading controls only the noindex flag. Edit robots.txt directly or through Yoast's file editor to allow GPTBot, PerplexityBot, ClaudeBot and Google-Extended.

What we would ship first

  • Technical crawlFetched as GPTBot, PerplexityBot and ClaudeBot, not as a browser, so we see what they see.
  • Entity recordsOrganization schema with sameAs to Wikidata, Crunchbase and your profiles, so assistants know which company you are.
  • Answer-shaped contentPages that answer the question in the first paragraph and carry an author and a date.
  • Schema and llms.txtValidated JSON-LD and an llms.txt that tells crawlers what the site is for.

How the engagement runs

  • Audit. Crawl Audit scores 200 signals. Findings are ranked by impact, each with the code to paste.
  • Fix. Crawler access, schema and entity work first. Most sites clear the high-impact findings in two weeks.
  • Write. Answer pages for the questions you should own, with an author and a date.
  • Track. Weekly share-of-answer report across six assistants.
100
Lighthouse performance on this site. Same for SEO, accessibility and best practices
5
products of our own in production
3
practices, each with a named lead
1 day
reply time on every enquiry
4.9
from 9 reviews

Questions people ask

Do you work inside our existing WordPress setup?
Yes. The scoping sprint starts with what is there. We extend WordPress where that is honest and propose a move only when the platform is the constraint.
Can you fix AI crawler access on WordPress?
Settings → Reading controls only the noindex flag. Edit robots.txt directly or through Yoast's file editor to allow GPTBot, PerplexityBot, ClaudeBot and Google-Extended.
Is the audit manual?
No. Crawl Audit is a tool. You enter a domain and it runs. Nobody reviews it by hand.
Do I need the service if I run the audit?
Not necessarily. Many teams fix the findings themselves with the free generators. The service is for teams who would rather we did it.