GEO Lab #03 · S3-GROK-M1-R2 · Original response

This page presents one selected response from GEO Lab #03. It adds a smaller reading and citation entry point; it does not replace the full archive or collect a new answer.

The OpenAI arm uses the Responses API with web search and does not represent consumer ChatGPT Search. Grok records identify their actual consumer-interface mode and available model label.

Recorded conditions

Record ID
S3-GROK-M1-R2
Platform
Grok
Question / condition
M1
Repeat
2
Model label
Exact backend version undisclosed
Visible mode
Expert
Reasoning
Not independently selectable
Submitted / first recorded time
2026-11-02T06:34:46.219Z
Completeness
Recorded body complete
Search
Actual search activity observed; exact controls and evidence remain in the raw record.

Original prompt

I'm a solo SaaS founder with no GEO budget. How can I reliably measure whether ChatGPT and Grok recommend my product to potential buyers?

The agreed prompt is reproduced exactly. Any recorded difference in the actual submitted text or search UI is retained in the raw record and full appendix.

Original answer

There is no official “how often we recommend you” dashboard from OpenAI or xAI. Measurement is sampling-based. For a solo founder with zero GEO budget, the most reliable approach is a disciplined manual process plus free snapshot tools and outcome signals.

1. Systematic Manual Sampling (Your Core Reliable Method)

This is the gold standard at zero cost. Answers vary between runs, so treat it like A/B testing.

Write 10–15 real buyer prompts (never put your brand name in them). Examples:
“Best [category] for a solo founder / small team on a budget”
“[Competitor] alternatives”
“How do I solve [specific pain your product fixes]”
“Top [category] tools in 2026 for [use case]”
Comparison or feature-specific versions
Test cleanly:
Use temporary/incognito chats or log out.
Fresh chat for every prompt.
Enable search/browse where available (especially ChatGPT).
Run each prompt 3–5 times on both ChatGPT and Grok.
Log everything in a Google Sheet:
Date, prompt, engine
Mentioned? (Y/N)
Position (1st, listed, etc.)
Sources cited
Competitors named
Reason given
Calculate a simple visibility rate: percentage of runs where you appear. Track average position and which sources keep appearing.

Re-run the same prompt set monthly (or weekly for your top 5). Consistency matters more than volume.

2. Free Snapshot Tools for Quick Baselines

Use these for a fast first look or quarterly check, not as your only ongoing method:

HubSpot AI Search / AEO Grader — Free, no (or minimal) account required. Runs prompts across ChatGPT, Perplexity, and Gemini and scores visibility, sentiment, and share of voice.
GetMentioned free AI visibility report — No signup. Runs real buyer-style prompts and shows visibility score, competitors, and cited sources.
Mangools AI Search Watcher (free plan) — 5 prompts, weekly updates, covers ChatGPT, Grok, Gemini, Claude, AI Overviews, and others. Useful for light ongoing tracking.
AthenaHQ Essential — Free credits (around 300) across several major engines for prompt analysis and source insights.
Limited free checks from Semrush or Ahrefs AI visibility tools when available.

These give you a structured starting point without payment.

3. Outcome Signals (Often More Actionable Than Mentions)

Mentions only matter if they drive awareness or signups.

GA4 referrals: Create a custom channel group for AI assistants. Look for traffic from chatgpt.com, openai.com, grok.x.ai, x.ai, perplexity.ai, etc. Note that not all AI sessions pass a clean referrer.
Google Search Console: Check generative AI / AI Overview impressions if available for your property.
Bing Webmaster Tools: AI performance and citation reports (more relevant to Copilot, but useful context).
Customer survey: On signup or in onboarding, ask “How did you first hear about us?” and include ChatGPT, Grok, “AI recommendation,” or “AI search.” This is the cleanest free signal of real impact.
Practical Tips for Reliability
Focus on non-branded discovery prompts. Branded queries only tell you the model knows you exist.
Prioritize cited sources. For SaaS, third-party pages (comparison listicles, review sites, Reddit threads, competitor pages) often drive recommendations more than your own site. Knowing the sources is more useful than the mention rate alone.
Expect high variance. One run means almost nothing; patterns across many runs matter.
Grok leans more on social, X, YouTube, and community content than ChatGPT, so track those surfaces too if Grok is important.
Re-test after you publish new comparison content, get mentioned on key third-party pages, or improve structured data — then measure the delta.
Minimal Weekly/Monthly Workflow
Maintain your fixed prompt list.
Run the key prompts manually (or use a free tool’s 5-prompt limit).
Update the sheet.
Glance at GA4 AI referrals and any “AI” mentions in customer surveys.
Note which sources keep appearing and whether you can influence them.

This gives you a consistent, zero-budget measurement system that is more reliable than one-off checks or expensive tools. Start with the manual baseline this week — it takes 1–2 hours and tells you immediately where you stand.

Recorded sources

No source URL was captured for this selected response. This does not establish that no search occurred.

Evidence and qualifications

This is the existing redacted public record. It has not been rewritten for this page. Earlier privacy edits and collection qualifications continue to apply. Editorial text is English; a few original captures retain non-English UI badges or source-panel date labels. An answer’s empirical claims are platform output, not verified findings of this archive.

Capture method

Complete rendered Grok article innerText; UI heading/duration and source badge removed. Exact inline hrefs extracted from scoped rendered article.

Record notes

[
"Complete answer has no observable inline citation anchors despite an 85-source badge. Badge entries are not assigned as answer citations."
]

Existing public-text privacy policy

{
"body_changed": false,
"policy": "Only actual URL tracking, signed/session URL parameters and necessary identifying information are removed; original wording, including analytics parameter examples outside URLs, remains.",
"annotation_positions": "Original position indexes omitted because URL/privacy edits can change character locations."
}

Prompt SHA-256: 709fd6fc82fe3f87b623fb3dd8b2adecdd6655c3508115f8b40c573d2397776d
Answer SHA-256: 199884a481a5890c47526818d66db8c86a93565cbdadc5e5adf2b91ad8a70139