A founder walked into a pipeline review with a dashboard most SEO teams would celebrate: organic sessions up, rankings improving, and 12 new page-one terms since spring. The problem? Qualified pipeline was down 18% and had been falling for two consecutive quarters.
The dashboard couldn’t explain the gap. A closer look at five buyer prompts across four AI assistants revealed why: the company appeared in only three of 20 AI-generated answers, and all 12 of those page-one wins were informational searches. The visibility was reaching learners, not buyers.
That distinction matters now that AI assistants sit between vendors and purchasing teams. When a buyer asks which contract management tool a mid-market legal team should use and your brand isn’t named, you don’t exist in that evaluation.
Five metrics that expose the real pipeline
Winston Francois, a growth consultancy for venture- and PE-backed companies, uses five measures to see whether AI search actually creates opportunities.
- Brand presence on buyer prompts: Run five vendor-selection questions across four assistants each week. Record whether your company is named, linked, or absent. The share of answers naming you is your brand presence rate.
- Answer accuracy: Read what the models say about your product, buyer, price tier and competitors. A Series A client was named in 11 of 20 answers, but only four descriptions were accurate. After cleaning up inconsistent information, 10 of 11 were accurate six weeks later and demo-to-opportunity conversion recovered the next quarter.
- Conversion from AI referrals: In GA4, group sessions from ChatGPT.com, perplexity.ai, gemini.google.com, copilot.microsoft.com and claude.ai. Track demo or trial conversion rate and session volume. Across the consultancy’s clients, these visits convert three to five times better than organic sessions, but small samples change fast.
- Branded search: Watch branded impressions and clicks in Search Console, plus direct traffic to homepage and pricing pages. The consultancy sees branded search lift four to eight weeks after brand presence improves.
- Self-reported AI attribution: Add “ChatGPT or another AI assistant” to inbound forms and have sales record the answer in HubSpot or Salesforce. One client saw nine of 54 new opportunities select the AI option by July after adding the field in April; those deals closed 30% faster.
What to do before you move more budget
Start with a spreadsheet, not a dashboard. Have one owner run the 20-answer audit every Monday. Budget about 40 minutes for the answer review, 15 minutes for the GA4 segment, and five minutes for the inbound form edit. Expect brand presence and accuracy to move first, usually in two to six weeks. AI referral conversions tend to show up in weeks four to eight, and branded search around weeks eight to twelve. Self-reported attribution can lag by an entire sales cycle.
If brand presence improves and nothing else moves after 90 days, re-read the answers. Are the descriptions accurate? Are the prompts ones your buyers actually type? It’s easy to choose prompts that make the report look better without reaching anyone likely to buy.
The founder from the opening now reports AI presence alongside pipeline. That’s the point: rankings are still useful, but they’re no longer sufficient evidence of buyer demand.
Source: Search Engine Journal
