§01introduction
The brand lead at a 200-person PLG company has been running brand there for three years. Twelve months ago, their team made a deliberate shift: they started tracking LLM citation rate as a primary brand health metric, ahead of share-of-voice and organic traffic. We talked about what drove that decision, what they changed, and what the data looks like now.
This conversation has been edited for length and clarity.
§02how it started
Find me Cited: Walk me back to the moment you decided citation rate mattered.
Brand Lead: Honestly it started with a support ticket. A new customer wrote in saying they found us because Claude recommended us when they were evaluating tools in our category. We were curious — had that happened before? We started asking in onboarding surveys. Over three months, 11% of new signups said an AI tool had mentioned us in their research. That number surprised us. We weren’t tracking it at all, which felt like a problem.
FMC: What did you do with that signal?
Brand Lead: We started running our own queries. Manually at first — just asking Claude and ChatGPT the questions we thought our buyers were asking. What we found was that we were getting cited on maybe 20-25% of the queries we ran. Our main competitor was at 60-70%. That gap was uncomfortable.
FMC: What did you think was causing it?
Brand Lead: Our content was all optimized for search. We had a lot of listicle-style posts, keyword-heavy landing pages, that kind of thing. It ranked well on Google but it wasn’t citeable. There was nothing specific to quote. We looked at what the competitor was doing differently and it came down to two things: they had better comparison pages and their docs were genuinely excellent. Ours were functional but not the kind of thing an LLM would pull from.
§03what they measure
FMC: How do you measure citation rate at scale? You can’t manually run hundreds of queries every week.
Brand Lead: We built a query set — about 80 queries that we think represent our buyer’s research journey. Everything from broad (“best [category] software”) to specific (“[our category] for enterprise teams with SAML”). We run that set weekly across four models and track citation rate per query and per model. It takes about an hour of tool time, 15 minutes of human review.
FMC: And you said citation rate comes before MRR in your dashboard?
Brand Lead: It’s a leading indicator of brand health. MRR tells you what happened 90 days ago. Citation rate tells you what’s happening in the research phase right now. When citation rate drops, we investigate immediately — usually it means a competitor published something new, or our content aged out on a topic. When it goes up, we track what we changed and double down. It’s become the most actionable brand metric we have.
§04what they changed
FMC: Talk me through the content changes specifically.
Brand Lead: Three main things. First, we rewrote our docs. Every page now opens with a one-sentence definition of the concept it covers — specific, no jargon. We added a “compared to” section to our main feature pages. We added real numbers: how many API calls per second, what latency looks like under load, what the typical setup time is. Stuff that’s actually quotable.
FMC: Second thing?
Brand Lead: Comparison pages. We had zero. We built five in Q3 last year — one for each of our main competitors, written honestly. Not “why we’re better,” but “when to choose us vs them.” We acknowledged where they were stronger. Those five pages now account for 31% of our LLM citation mentions.
FMC: And the third?
Brand Lead: Reddit. We started showing up in the subreddits where our buyers hang out. Not pitching — answering questions. Disclosing our affiliation when relevant, but leading with actual help. It takes time but the citations follow.
§05results
FMC: What did citation rate look like before vs after?
Brand Lead: We went from around 22% to 61% over about nine months. That’s across all models, all queries. The gap with our main competitor is now basically closed — we’re at 61%, they’re at 68%. Twelve months ago it was 22% vs 68%.
FMC: Did you see that show up in revenue?
Brand Lead: It’s hard to isolate but the “found via AI” response in our onboarding survey went from 11% to 29%. Qualified pipeline is up. I can’t draw a straight line but the correlation is there.
§06advice for teams starting out
FMC: One piece of advice for a brand team that hasn’t started tracking citation rate yet?
Brand Lead: Run your 10 most important buyer queries through Claude and ChatGPT right now. Don’t read the answers for content — just look at who gets named. That’s your competitive landscape in the AI channel. If you’re not in those answers, you have a gap. Start there.
You can do that check in 60 seconds at findmecited.com/check.