What is prompt tracking?

Prompt tracking is the measurement side of LLM SEO. You pick the questions your buyers ask an AI assistant, such as "best help desk software for small teams". A person or a tool sends each question to each assistant, again and again, and stores the answers. The record shows how often your brand appears, in what words, and next to which competitors.

The term has a second meaning. Developers say "LLM tracking" or "prompt tracking" for logging the prompts and outputs of their own AI app. This entry is about the marketing meaning.

Why does prompt tracking matter for B2B SaaS?

Buyers now ask assistants for shortlists. G2's 2026 survey of 1,076 B2B decision-makers found that 51% of B2B software buyers now start their research in an AI chatbot more often than in Google.

Your usual reports show little of this. Google says traffic from AI Overviews and AI Mode is included in Search Console, in the Performance report under the "Web" search type (Google Search Central: AI features and your website, last updated 2025-12-10, read 1 November 2026). That counts clicks and impressions. It does not show what an answer said about you, and it does not cover ChatGPT, Claude or Perplexity.

Bing goes a step further. Its AI Performance report in Bing Webmaster Tools shows where your content is cited in AI-generated answers. In June 2026 Bing added Citation Share: "the percentage of citations attributed to your site out of all citations shown across all sites" for one grounding query (Bing blog, 16 June 2026, read 1 November 2026). For the other assistants, you track prompts yourself.

How does prompt tracking work?

  1. Write 20 to 50 prompts your buyers would type. Mix category questions ("best X for Y"), comparisons ("X vs Y") and questions about your brand.
  2. Pick the assistants: for example ChatGPT, Perplexity, Gemini, Claude, Copilot and Google AI Overviews.
  3. Run each prompt several times on each assistant, because answers change between runs.
  4. Record for each answer: are you named, are you linked, how are you described, which competitors appear and which sources are cited.
  5. Keep the prompt set fixed and compare month by month.

Step 3 matters more than it looks. Prefer ran 80 questions three times on four engines (960 answers, measured 13 September 2026). Two runs of the same question shared on average 37% of the cited sites on ChatGPT and 35% on Gemini, against 96% on Perplexity and 68% on Claude. One run is a sample, not a result.

What does prompt tracking look like? (hypothetical example)

A hypothetical invoicing tool tracks 30 prompts on five assistants, three runs each: 450 answers a month. In October it is named in 81 of them. Most of those answers cite two comparison lists, and the tool is in only one of them. The next task is clear: get into the second list, or give its writer a reason to add the tool.

  • AI visibility tracking is the wider practice. Prompt tracking is its core method.
  • AI share of voice is a number you work out from tracked prompts: your mentions against your competitors' mentions on the same questions.
  • An AI visibility score is a tool's own index built from the same kind of data. Each vendor calculates it differently, so scores from two tools are not comparable.
  • Rank tracking checks your position in a list of links. Prompt tracking checks whether you are in the answer at all.

How does prompt tracking connect to third-party lists?

Tracking often shows that answers cite pages you do not own. Foundation and AirOps looked at 5.1 million AI responses for 50 B2B brands: on unbranded questions, only 2.24% of citations went to the brand's own pages. When Prefer checked 188 answers in July 2026, 40% of the sources cited for buyer questions were independent lists.

So the sources column is often the most useful part of a tracking report. It tells you which lists, reviews and posts the assistants read for your category, and which of them leave you out.