How should you use this toolkit?

Every tool here implements a piece of the open methodology documented in this encyclopedia — nothing is gated behind a pitch. The sequence that works: establish a baseline, identify the gap type, apply the playbook, re-measure. Tools without the loop are trivia; the loop is the product.

The free GEO Checker

The GEO Checker runs a fast visibility baseline: enter your domain and it assesses the machine-readability layer — indexation signals, structured data presence, robots and crawler access (GPTBot, OAI-SearchBot), answer-block formatting on key pages — and returns a prioritized fix list. It takes minutes and it is the right first step for any site. What it deliberately does not do: it does not invent an AI visibility score. Visibility is measured by asking the engines real questions, which brings us to the measurement stack.

The four-dimension measurement stack

The core toolkit is procedural rather than software:

  1. Question-set template. The structure for your fixed 30-50 question set: category questions, brand questions, situation questions, per market and language. Frozen with stakeholder sign-off; refreshed quarterly at most.
  2. Run protocol. Same day monthly, fresh sessions, verbatim prompts, mode noted (browsing, DeepSearch, AI Mode), full screenshots archived with date and engine visible.
  3. Scorecard. Visibility, Citation Rate, Semantic Coverage and Brand Perception, each 0-100, per engine. Rubrics: Visibility, Citation Rate, Semantic Coverage, Brand Perception.
  4. Delta log. Every gap mapped to its fix class: indexation, content, entity, perception — with an owner and a date.

Supporting checklists

  • Entity consistency checklist — the audit walkthrough from our entity audit post as a stepwise list: inventory, canonical record, reconcile, schema, verify.
  • Structured data checklist — Organization/LocalBusiness, FAQPage, Service and Product patterns with validation pointers: structured data for AI.
  • Crawler access checklist — robots.txt review for GPTBot, OAI-SearchBot, ChatGPT-User and Google-Extended (which controls training only, not retrieval): AI crawler management.
  • IndexNow setup — the Bing-side visibility that feeds ChatGPT’s retrieval.

Can agencies use these tools with clients?

Yes — the templates and rubrics are deliberately reusable, and the standards pages are public so agencies can cite an independent methodology rather than a proprietary black box. The two requirements we ask everyone to honor: keep the question set frozen within a measurement period, and keep screenshot evidence for any score you report. Agencies that report unfalsifiable composite numbers are exactly what the open standard exists to displace.

What we do not build, and why

No automated promise-you-a-ranking dashboard. AI answers are stochastic, personalization varies, and a single automated score invites the cherry-picking the methodology exists to prevent. Where automation helps (scheduled question runs, evidence archiving), we use it inside engagements; where it substitutes theater for measurement, we refuse it. If you want the measurement run for you with audit-grade evidence, that is the monitoring service.

Start here

  1. Run the GEO Checker and save the output.
  2. Ask ChatGPT, Perplexity and Google five questions your customers ask; screenshot the answers.
  3. Compare: are you named, cited, described accurately? That comparison is your baseline — and your to-do list.

Frequently asked questions

Is the GEO Checker really free?
Yes — it is the entry point of the methodology, not a lead trap with the useful parts removed. Run it, keep the results, fix what it flags. If you want the full answer-layer measurement done for you afterward, that is a service; the checker itself stands alone.
Which third-party GEO tools do you recommend?
Any tool that runs real prompts against real engines and archives evidence can support the loop. What to avoid: tools reporting a single composite AI score without a frozen question set or screenshots — the number is unfalsifiable, which is the tell.