How it works
BrandPresence keeps the path from question to answer to action visible, so a team can move from “we should check this” to a repeatable operating habit.
Choose the commercial prompts your buyers actually ask.
Use OpenAI, Perplexity, and Gemini with the same prompt settings.
Inspect answers, mentions, observed entities, citations, and source domains.
Choose a supported next step, verify it, and see what changed.
First value
Honest measurement
Queued
What is happening
Your measurement is accepted and waiting for an available worker.
What to expect
The prompt set and selected providers are saved; sampling has not started yet.
Next
Wait for the run to move to Running. You do not need to submit it again.
Running
What is happening
Provider requests are in progress and completed observations are being recorded.
What to expect
Samples can finish at different times, so the count may grow before the run closes.
Next
Keep working or return to run history; wait for a terminal state before comparing totals.
Partially succeeded
What is happening
Some prompts or providers finished while others returned an error or timed out.
What to expect
Successful evidence remains usable, but the run is not complete coverage.
Next
Review the completed answers, then rerun the missing coverage when it matters.
Failed
What is happening
The run finished without enough usable observations to report a result.
What to expect
The failure is retained with its context instead of being presented as zero visibility.
Next
Check the configuration or provider error, fix the cause, and start a fresh measurement.
AI assistant integration
Connect an MCP-compatible AI assistant to investigate findings, compare provider runs, retrieve retained evidence, and prepare briefs through a read-only organisation-scoped connection.