Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.topify.ai/llms.txt

Use this file to discover all available pages before exploring further.

This page walks through what happens after you create a project — from the daily prompt run to the metrics on your dashboard.

​
Daily prompt run

Topify sends every active prompt in your project to all monitored AI providers, once per day. Each response is collected within hours of the run, then handed off to analysis. For the full provider list and which platforms count toward your plan, see Integrations.

​
What gets analyzed

Every response is parsed for four things:
  • Brand mentions — which brands appear, used to compute visibility and position
  • Sentiment — how positively each mentioned brand is described (0–100)
  • Source citations — which websites and URLs the AI references
  • Competitor signal — brands that consistently appear alongside yours
For metric definitions and formulas, see Brand metrics and Source metrics.

​
Data retention

Every response is preserved indefinitely. The dashboard exposes 7-, 14-, 30-, and 90-day query windows for trend analysis, but no data is dropped after those windows.

​
Bootstrap: prompts and competitors

When you create a project, Topify runs a one-time setup pass:
  • Prompt generation — your website is crawled to understand your industry and offerings, and a starter set of prompts is drafted across the four intent types (informational, comparison, transactional, navigational).
  • Competitor detection — your starter prompts are sent to AI providers, and brands that consistently appear in the responses are flagged as competitors. This captures your AI search competitors, which are often different from your traditional market competitors.
Both are reviewable: you decide which suggested prompts to activate and which competitors to track.

​
FAQ

​
Can I track competitors that AI doesn’t surface yet?

Yes. From Preferences > Competitors, manually add any brand by name and Topify starts watching for it in every response we collect going forward. This is useful for emerging players or adjacent-category brands you want to monitor before they become AI-visible.

​
How does Topify get the AI responses — through official APIs?

We collect responses by driving the actual AI assistant interfaces (ChatGPT, Perplexity, Google AI Overview, and Gemini) — automated browser sessions against the same web UIs your buyers use. This means the data reflects what real users actually see, not a stripped-down API view that may differ from the consumer experience. Claude is the exception — we use Anthropic’s official API, since the consumer Claude product doesn’t differ meaningfully from the API output.