Public Lab

Public Lab · Real Public Evidence

We audited Growthract’s own AI-search readiness. Here’s what the evidence showed.

A public technical audit of growthract.com covering 49 production HTML pages, crawler access, robots.txt, sitemap integrity and legacy routes — with the limits of the evidence stated explicitly.

Client

Growthract

Focus

Public Lab

Published

Aug 22, 2026

Reading time

9 min read

Public Lab 01 · Baseline

We ran the technical audit on ourselves.

Growthract talks about evidence, crawler accessibility, machine-readable information and technical verification. We wanted the same standard applied to our own production website.

On August 21, 2026 at 20:03 UTC, version 2 of our public site-health audit was run against www.growthract.com. The snapshot covered production pages, robots.txt, the XML sitemap and a defined set of legacy URLs.

This Lab publishes that baseline as it existed at the time of the audit. The website can change after the snapshot — including through the publication of this Lab itself.

49

HTML pages audited

Production HTML pages included in the August 21 baseline.

49/49

Direct HTTP 200

Every audited HTML page returned a direct successful response.

49

Sitemap URLs

The XML sitemap returned HTTP 200 and contained 49 URLs.

0

Issues reported

Across the checks defined in audit version 2.

What was tested

Deterministic checks, not a black-box readiness score.

The audit fetched public production resources and evaluated observable technical responses. It did not ask a language model to decide whether Growthract had “passed” AI search.

Production HTML

HTTP responses and supported page-level technical signals.

robots.txt

Availability and published user-agent access rules.

XML sitemap

HTTP accessibility, content type and published URL set.

Legacy URLs

Whether known old routes redirected intentionally or returned an explicit retired-page response.

Evidence boundary

This audit did not determine whether ChatGPT, Claude, Perplexity, Gemini or another external system would rank, cite or recommend Growthract. Those outcomes require different evidence.

Production responses

All 49 audited HTML pages returned direct HTTP 200 responses.

The run crawled 50 resources in total: 49 HTML pages and one XML resource. Every audited HTML page returned a direct successful response. The audit recorded zero error resources and zero noindex HTML pages.

Total resources crawled50
HTML pages49
XML resources1
Direct 200 HTML pages49
Error resources0
Noindex HTML pages0

Sitemap

The sitemap returned HTTP 200 with 49 published URLs.

At snapshot time, sitemap.xml was accessible as XML and contained 49 URLs. That number represents the site before this Public Lab was published, so the current sitemap may contain more URLs.

GET /sitemap.xml
HTTP 200

Content-Type: application/xml
URLs found: 49

robots.txt

AI-related user-agent rules were explicit and publicly accessible.

robots.txt returned HTTP 200. The production configuration contained explicit groups for the following agents.

OAI-SearchBotChatGPT-UserGPTBotClaude-SearchBotClaude-UserClaudeBotPerplexityBotPerplexity-UserGoogle-Extended
Allow: /
Disallow: /api/

Sitemap: https://www.growthract.com/sitemap.xml

What this proves

It proves what the published access configuration said at audit time. It does not prove that any provider had already crawled, indexed, retrieved or cited the site.

Legacy route hygiene

Five known old URLs redirected to intentional replacements.

The audit separately tested eight known legacy routes. Five returned permanent 308 redirects and resolved to their intended current pages.

/how-it-works/sprint
/seo-aeo/what-we-fix
/ai-search-visibility/what-we-fix
/llm-entity-citation-science/entity-science-guide
/insights/resources/technical-guides

Three intentionally retired URLs returned 404 rather than being redirected to unrelated pages.

404/services/websites-digital-experiences
404/insights/blogs/where-ai-reduces-manual-work-marketing-sales
404/insights/blogs/cs-clarifying-b2b-positioning

Interpreting the result

“0 issues” does not mean “perfect AI SEO.”

Audit version 2 reported zero critical, high, medium and low issues across the technical checks defined by that audit.

That is not a universal website quality score. It should not be converted into a claim that Growthract has perfect AI visibility or guaranteed citations.

Supported by this audit

  • • Production page accessibility
  • • Sitemap accessibility and baseline URL count
  • • Published robots.txt configuration
  • • Defined page-level technical checks
  • • Known legacy-route behavior

Not supported by this audit

  • • Guaranteed AI citations
  • • Continuous AI visibility
  • • Search ranking guarantees
  • • Traffic or revenue outcomes
  • • Proof that every permitted agent actually visited

Why Public Labs

Keep different kinds of evidence separate.

AI-search marketing can easily combine configuration checks, AI opinions, crawler activity and citation outcomes into a single vague visibility claim.

We prefer to keep those layers distinct. A deterministic technical audit should publish deterministic observations. Crawler logs should show observed requests. An AI-response test should show the actual tested response.

This baseline is the first Growthract Public Lab built around that principle.

Next

Establish the technical baseline, then test the next question.

Future Public Labs can examine other evidence independently, including crawler-log observations, point-in-time AI responses and technical changes across repeated audits.

Continue exploring

Explore Growthract’s SEO + AEO approach

Start with the problem

Find out what your search system should improve next.

Start with a focused analysis of your website and identify the SEO, AEO and search visibility opportunities that deserve attention.

Book a call