Step 1: List your high-intent queries

Start from buyer intent, not vanity. Pull the 10–30 queries a potential customer asks an AI right before they buy: "best tool for X," "X vs Y," "how to do Y." These are the queries where a citation — or a competitor citation — changes the pipeline.

If you're not sure, mine support tickets, sales calls, and search-console queries. The closer to real intent, the more the gap matters.

Step 2: Run each query across the engines

Ask ChatGPT, Perplexity, and Claude the same query and record every brand the answer named. Don't stop at "did we show up?" — write down the full set. The competitor who led is the one you're losing to.

Do this on a schedule, not once, or the result is just noise.

Step 3: Mark the gaps

For each query, flag any case where a competitor was named and you were not. That flag is your citation gap. Sort gaps by intent: a gap on "best tool for X" beats a gap on a niche how-to.

A tool that records the gap per query and engine (like BrandPulse) turns this from a spreadsheet chore into a standing list.

Step 4: Decide what to fix

Not every gap is worth it. Close gaps that (a) sit on high-intent queries, (b) map to a page you can realistically improve, and (c) where the competitor's cited claim is something you can match or beat. Skip gaps that are low-intent or off-strategy.

Then make the target page clearly answer the query with citable sources, and re-check the gap next cycle.

Authoritative references

  • Google Search Central: https://developers.google.com/search/docs
  • OpenAI: https://openai.com/blog