The problem this solves
AI visibility reporting has a resolution problem. Most teams track a single number, some overall presence score, and a single number cannot answer the questions that actually drive action: which buyer questions cite us, which engines know us, and where exactly is the competitor eating our ground. The score moves and nobody can say why; the score holds steady while a rival quietly takes over the three questions closest to purchase intent.
The underlying data is a matrix: engines on one side, buyer questions on the other, and a citation strength in every cell. Reported as prose or a table, the matrix is unreadable at any useful size; fifteen questions across four engines is sixty cells of text that no one will study weekly. Reported as a single average, it destroys precisely the information that matters, because visibility is won and lost cell by cell.
A heatmap is the native visual form for this exact data shape, and this mission produces one on schedule: every tracked question tested against every engine, the citation picture rendered as a single designed figure, and the figure posted where the team already talks. The whole matrix becomes readable in five seconds, and the cells that changed since last run are the story.
How the mission runs
- Run the tracked questions across the engines. Each of your tracked buyer questions is put to the AI assistants fresh, and each answer is scored for whether and how strongly your brand is cited: named as the answer, listed among options, or absent. The run is dated, so successive runs form a comparable series.
- Score the competitive cells. The same answers are read for competitor citations, so each cell knows who is present alongside you or instead of you. A cell where you are absent has a second fact worth knowing: whether the space is empty or taken.
- Render the heatmap. The matrix is rendered as a designed heatmap by the platform's chart engine: engines by questions, a single-hue scale from light to dark encoding citation strength, clean cell spacing, readable at Slack-preview size. One figure carries all sixty cells.
- Write the delta note. The mission compares this run against the previous one and writes a short note naming the cells that moved: questions gained, questions lost, engines that shifted. The heatmap shows the state; the note shows the motion.
- Post to the channel. Figure and note land together in your named Slack channel on schedule. The team absorbs the state of AI visibility in one glance in the channel they already read, with no dashboard visit required.
The prompt
This is the exact objective the agent receives. Swap the obvious placeholders for your own domain, segment or channel and run it as-is from the console, Slack, or the API.
What comes back
A scheduled Slack post carrying one designed heatmap of your entire AI citation picture, engines by buyer questions with citation strength as color, plus a delta note naming exactly which cells moved since the last run. Sixty cells of visibility data, readable in five seconds, on a cadence.
Make it yours
- Render a second heatmap for your primary competitor and post the pair, turning the weekly glance into a side-by-side territory map.
- Expand the question set with the buyer questions your content targets next quarter, so the heatmap shows the ground you intend to take while you take it.
- Route a monthly version into the reporting deck in dark theme, so leadership sees the same figure the team works from.
Frequently asked questions
Why a heatmap instead of a score or a table?
Because the data is a matrix and the decisions are per-cell. A score averages away the information; a table demands study. The heatmap keeps every cell distinct while staying readable in seconds, and the dark cells you lack literally look like territory to take.
How stable are the results run to run?
Assistant answers vary, which is why the mission tests a consistent question set on a consistent cadence and reports deltas rather than treating any single run as gospel. Patterns that persist across runs are the signal; the delta note is written with that discipline.
What do we do with a weak column or row?
A weak row is a question your content does not yet answer citably, which is a brief for the content team. A weak column is an engine that has not absorbed your brand, which points at the sources that engine trusts. Either way the heatmap converts a vague worry into a named target.