/
Company
/
3 min read
Why we started Cited
Buyers now ask AI engines for a shortlist before they visit a vendor’s site. We started Cited to make that step measurable, and today we’re announcing our Series A.
Daniel Okafor
Co-founder and CEO
Last year, the head of marketing at a payroll software company showed us a spreadsheet. Every Monday for three months, her team had pasted 60 buyer questions into ChatGPT and logged which vendors came back. It took four hours a week, and it was the report her CEO read most closely. When we asked why she kept doing it by hand, she said:
We didn’t need another rank tracker. We needed to know what the answer said about us on Monday morning.
That spreadsheet is where Cited started. Today we’re announcing an $18M Series A led by Quarry Lane Ventures, and this post explains what we’re building and why.
Buyers changed first
The shift showed up in buyer behavior before it showed up in any marketing tool. In a survey of 1,100 B2B buyers we ran this winter, 62% said they had used an AI assistant to build a vendor shortlist in the past six months. Most never clicked a search result during that step.
That leaves marketing teams with a blind spot. Web analytics can’t see an answer that never sent a visit, and rank trackers measure a results page many buyers now skip.
What Cited measures
With Prompt Tracking, Cited runs the prompts your buyers ask on ChatGPT, Perplexity, Gemini, Claude, Copilot and Google AI Overviews every week, and records what each engine says. From those answers we report:
Share of Answer: the percentage of tracked prompts where you’re cited or recommended
Citation Sources: the pages and domains the engines cite
Answer Diff: what changed in the answers since last week
Crawler Insights: how AI crawlers reach and read your site
What the funding is for
Today, 180 teams use Cited across software, commerce, finance, health and travel. The Series A goes to three things: a larger research team, coverage of new engines as they launch, and tools that help teams act on what they see, starting with content recommendations later this year.
We’ll keep publishing our research methods alongside the numbers, and we’ll describe engine behavior only from what the answers show. We won’t sell guaranteed placement, because nobody can offer it honestly. If that sounds like work you’d want to do, we’re hiring in New York and remotely across the US and EU.