Verifex is Vohkus' internal website audit tool. Enter any domain or page URL and it produces a full diagnostic report in about ninety seconds: an overall score out of 100, eleven individual scores covering design, copy, conversion, user experience, performance, SEO, AEO, GEO, trust, psychology and clarity, and most importantly the reasoning behind every number and a complete, ranked fix plan for each category.
The intent is practical: our web developer should be able to read a report, work through the fix lists, re-audit, and watch the scores move. Access is by sign-in code, restricted to Vohkus and Injekt Media email addresses, and every audit is logged.
Verifex combines three engines. Google's PageSpeed Insights (the Lighthouse engine, run on Google Cloud) loads the actual page under mobile conditions and returns real laboratory measurements like load times, Core Web Vitals, and category scores for performance, SEO, accessibility and best practices, along with a rendered screenshot. Verifex's own technical inspector then parses the page's source code directly and runs around twenty deterministic checks: title and meta description lengths, canonical tags, Open Graph and Twitter cards, H1 structure, image alt text, JSON-LD structured data validity, robots.txt, sitemap presence, and whether AI crawlers such as GPTBot and ClaudeBot are permitted. Finally, Claude Sonnet 4.6 (Anthropic's AI model) receives the screenshot, the live page content and all of the measured facts, and performs the qualitative analysis. Accounts and audit logging run on Supabase; hosting is on Vercel.
Because the measured data contains no opinion. Performance and SEO scores come straight from Google Lighthouse, the same tool Google itself provides to developers. They are labelled "measured" on the dashboard. The technical checks are facts: when Verifex says a title tag is 64 characters or a page has two H1s, that is read directly from the code and can be verified by anyone in seconds by viewing the page source. These numbers are repeatable: run the audit again and the facts stay the facts (speed metrics vary slightly run to run, as they do in any lab test, because real network conditions vary).
The AI aspect of this app has been trained to judge the craft of the page the way a consultant would: is the design coherent? the copy persuasive? the path to action clear? the content answerable and citable by search and AI engines? It works under strict evidence rules built and tested during development of this audit tool. It must look at a real screenshot from top to bottom and the actual fetched page before judging. It may never claim something is absent without verifying, including checking obvious sub-pages, and every score must cite the specific evidence behind it. The fix lists must be complete and concrete, with generic advice banned. Even if you put in world famous brands like amazon.com, they get no credit and no penalty, because the page is being judged, not the business.
These scores are evidence-grounded expert judgment, not physical measurements. Like two consultants reviewing the same site, a re-run may differ by a few points, and occasionally a judgment can be argued with, which is why every claim is written to be checkable against the page itself, and why the report separates what is measured from what is analysed. The right way to use the analysed scores is the way you'd use a trusted reviewer: read the evidence, fix the gaps, re-audit, and let the results show the improvement.
The Compare tab, available after any audit, lets you audit a second site and see both results side by side across the same eleven categories, scored by the same engine under the same conditions. Type a competitor's domain and Verifex runs a full audit against it: Google Lighthouse measures their speed, the same AI reviews their page under the same evidence rules, and the results appear as a head-to-head table with a per-category delta and an overall verdict chip ("you lead by 7" or the uncomfortable alternative).
Because both audits use identical methodology, the comparison is fair in a way that informal benchmarking is not: the same eleven criteria, the same mobile-emulated speed test, the same evidence requirements applied to both pages. The competitor's AI verdict also appears at the foot of the comparison, giving the team an independent read on what that site does well and where it is weak. Both audits are logged, so running the same comparison over weeks builds a trend over time.
AI search engines including ChatGPT, Claude, Perplexity and Google's AI Overviews increasingly answer buyer questions directly rather than listing links. The AI citation test checks whether this site gets cited when a real AI assistant answers questions about your category.
The test works in three steps. First, Verifex's AI studies the audited site and derives five realistic questions a potential customer would ask an AI assistant in that product or service category. The brand name is deliberately excluded from the questions: the aim is to test whether the site earns a citation on its own merits, not just when someone searches for the company by name. Second, the AI runs a live web search for each question and inspects the results it finds. Third, it reports whether the audited domain appears in those results, with a brief note on the context ("appears mid-results" or "absent, competitors dominate").
The result is a count out of five ("2 of 5 answers cited this site") alongside a one-line summary of the site's overall AI-search visibility. This feeds directly into the GEO score (Generative Engine Optimisation) shown in the main report. One honest note on scope: five questions sampled via one search engine is a directional indicator, not a census. Treat it as a signal of AI visibility rather than a definitive measurement, and run it again after making changes to schema markup, content depth or structured answers on the site.