Skip to main content

Live Exercise: Build the First GEO/AEO Baseline

This is the core of the workshop. Everything before this module is setup; everything after it is what to do with what gets found here.

Run this as a genuinely hands-on exercise, not a demo. If facilitating a group, give each attendee (or small group) their own target brand and 20 minutes — the exercise lands harder against a brand someone actually cares about than a shared example on a projector.

Step 1: Select Target Brand + Priority Audience​

Pick one real brand (the attendee's own company, or a client if this is an agency-led session) and one specific target audience segment — not "everyone," a specific buyer persona. This matters because step 2's question list should reflect what that persona asks, not a generic keyword list.

Do this in the portal: confirm Organization Settings has the correct brand name and domain configured — every downstream report (Citation Monitor, AI Footprint, Agent-Readiness) pulls these as defaults, so getting this right first saves re-entry later.

Step 2: Choose 10–20 High-Intent Questions​

This is the step most groups want to rush, and it's the one worth slowing down on — a weak question list produces a weak baseline no matter how good the reports are.

Two ways to build this list, both legitimate:

  • From existing knowledge, if the brand already has a rough sense of common buyer questions — build a Citation List manually.
  • From real data, if a Q&A Analysis report already exists for this brand — use Generate a Citation List from Q&A to pull the actual Query Frequency Histogram and Citation Phrase Suggestions instead of guessing. This is the stronger path whenever it's available, because it reflects what people actually ask instead of what the team assumes they ask.

Facilitator note: if no Q&A Analysis report exists yet for this brand, that's fine — a manually built list of 10–20 questions is a perfectly valid starting baseline, and running Q&A Analysis becomes the natural first item on the 30-day roadmap instead.

Step 3: Check AI Citation Status + Competitors​

Run Citation Monitor against the list from Step 2. Add 3–8 real, named competitors to the Competitors table in the generate-report modal — either typed manually or via Generate with AI if the room doesn't have a competitor list ready.

What to point out live: the Citation Health Score and verdict tier (Emerging / Present / Strong / Dominant) is the single number to write down as the baseline — but the Query Scorecard and Diagnostic Quadrant per question are what actually drive the action list two steps from now. Don't let the room stop at the headline score.

Real example to reference: in a real run, the headline score was 32/100 ("Emerging"), but the useful finding wasn't the number — it was that the one branded query landed in the Amplify quadrant (ranked and cited) while all three unbranded category queries landed in Full Gap or Content Gap, with a named competitor (Ada) cited instead on every one of them. That's a specific, buildable content gap, not just a low score.

Step 4: Review Landing Pages and Chatbot Answers​

For the pages Citation Monitor's Google Organic Rankings section identifies as ranking (even if not cited), pull them into URL Analysis. For the brand's own chatbot, spot-check 3–5 of the Step 2 questions directly in the live chat widget.

What to point out live: a page can rank in Google and still fail nearly every GEO signal — URL Analysis's real trust-center example scored 52.4/100 overall despite ranking reasonably, because it had zero FAQ schema, no visible dates, and vague claims like "industry-standard encryption" instead of named standards. That's the exact kind of gap this step should surface.

Step 5: Create First Prioritized Action List​

Take every gap surfaced in Steps 3 and 4 and sort it into Critical / High / Medium / Low, the same structure the Consolidated Action Plan uses. Don't wait for every GEO Suite report to exist for this brand — a hand-built list from just Citation Monitor and URL Analysis findings is a legitimate first version.

Output by the End of the Exercise​

  • Baseline citation rate — the Citation Health Score and verdict tier, written down and dated
  • First Keyword List or Citation List — saved, named, and reusable next month
  • Most obvious content / schema / trust gaps — a short list, not exhaustive
  • A clear next test date — put an actual date on the calendar before the room disperses; a baseline nobody re-tests is not a loop, it's just a report

What Makes This Exercise Credible​

  • Uses real questions and live answer behavior — nothing in this exercise is simulated; every score reflects an actual model call against actual content
  • Creates re-testable lists — the Citation List and/or Keyword List built here persist after the session and can be re-run with zero setup next time
  • Links findings to measurable traffic and engagement — if AI Referral Analytics is already live for this brand, close the loop by checking whether any of the queries just tested are already driving real sessions
Last updated on