How do I set up a GEO baseline to measure progress?

Q: How do I set up a GEO baseline to measure progress?

A GEO baseline is a snapshot of where you stand in AI search before you start optimizing. Here's how to establish one in under an hour so you can measure real progress.

Why This Matters in 2026 Without a baseline, you can't know whether your GEO work is producing results. Many businesses implement GEO changes and then have no way to tell if anything improved — they just hope it did. A baseline established before you start optimizing gives you a before-and-after comparison that shows real progress, identifies what's working, and keeps your efforts focused on what matters.

A GEO baseline is a documented snapshot of your current AI citation status across the platforms and queries that matter most to your business. You establish it before starting any GEO optimization work, and you compare against it at regular intervals. Here's exactly how to set one up in under an hour.

Step 1: Define your query set

Choose 8–12 queries that represent the searches your ideal customers would use to find a business like yours. Include:

3–4 high-volume local queries ("best [service] in [city]", "[service] near [city]")
2–3 specific service queries ("[specific service] in [city]", "who does [specific service] in [city]")
1–2 problem queries ("my [thing] is broken — who should I call in [city]")
1–2 brand queries (your business name, your business name + city)

These are your tracked queries. Run these same queries every time you measure — consistency is what makes trend data meaningful.

Step 2: Run your baseline audit

In a private/incognito browser, run each query on:

ChatGPT (with web browsing enabled)
Perplexity
Google (with AI Overviews enabled)

For each query × platform combination, record: Is your business cited? (Y/N) What's the name of the first business cited? What source does the AI reference when it cites any business?

Step 3: Record your baseline metrics

In your baseline spreadsheet, record:

Date of baseline audit
Citation score: how many of your 10 queries resulted in your business being cited (e.g., 2/10)
Google review count and rating as of today
Current AI referral traffic sessions (from GA4, last 30 days)
Competitors cited instead of you — which businesses appear consistently?

Step 4: Note your current state for each GEO factor

Before optimizing, record the current state of each major GEO input:

llms.txt: exists or not?
LocalBusiness schema on homepage: yes or no?
FAQPage schema on service pages: yes or no?
Number of service pages with answer-first content: count
AI crawlers explicitly allowed in robots.txt: yes or no?

This documents your starting point so you can see exactly which changes correlate with citation improvements later.

When to run your next measurement

After implementing changes, wait 6–8 weeks before measuring again. AI crawlers re-index content on a schedule, and citation changes take time to materialize. Don't check weekly — you'll see noise, not signal. Monthly checks after the initial 8-week wait are the right cadence.

Pro Tip Date-stamp your baseline clearly and save your actual AI response screenshots or copy-pastes alongside your spreadsheet. AI response phrasing changes frequently — having the raw response text from your baseline lets you compare exactly what the AI said then vs. now, not just whether your name appeared.

Key Takeaways

A GEO baseline is a documented snapshot of citation status, review counts, and GEO factor completion before you start optimizing
Define 8–12 consistent test queries covering local, service, problem, and brand searches
Run each query in ChatGPT, Perplexity, and Google AI Overviews in a private browser and record results
Note current state of GEO factors (llms.txt, schema, FAQ blocks) so you can correlate changes with results
Wait 6–8 weeks after implementing changes before measuring again — citation changes take time to appear