The problem this solves
Most agent-written reports describe numbers in prose: organic sessions rose eleven percent, paid efficiency held, the branded share dipped. Prose is how you report numbers when you cannot draw, and it quietly caps how much a reader absorbs. A trend that would be self-evident as a line becomes a sentence someone must parse; a channel comparison that a column chart settles in a glance becomes a paragraph the reader may skim past; the one anomaly worth discussing hides in the middle of text formatted exactly like everything around it.
So someone rebuilds the figures by hand. The marketer pastes the agent's numbers into a spreadsheet, fights the chart defaults, exports images, and assembles the deck, which means the monthly report costs an afternoon even though the analysis was free. Worse, hand-built charts drift: each month's figures use slightly different colors, scales, and framings, so the deck never develops the visual consistency that makes month-over-month comparison automatic.
This mission renders the figures itself. Every number that deserves a shape gets one, drawn by the platform's own chart engine with a consistent designed theme, assembled into a deck, and delivered to your inbox. The afternoon of chart assembly drops to zero, and the figures look identical in structure every month, which is exactly what makes them comparable.
How the mission runs
- Gather the month's evidence. Search Console supplies the organic record, the ad accounts supply spend and return by channel, and the AI visibility checks supply the citation picture. Every figure in the final deck traces to a query the mission actually ran this month.
- Choose the right form for each number. The mission matches each finding to the chart form its shape demands: time series as lines, channel comparisons as columns, spend against return as paired columns, share-of-citation as a heatmap of engines by questions. A number with no shape worth drawing stays a stat, deliberately.
- Render the figures. Each chart is rendered by the platform's native engine with the designed theme: consistent palette, thin marks, one axis, direct labels where they help. The same figure types use the same colors every month, so a reader who learned January's deck reads December's at a glance.
- Assemble the deck. The figures flow into a Gamma deck with a one-line takeaway above each chart stating what the figure shows and why it matters. The narrative connects the figures; the figures carry the numbers.
- Deliver to the inbox. The finished deck goes out by email to the list you name, on the schedule you set. The first anyone hears of the monthly report is the finished report arriving.
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 finished monthly deck in which every number that deserves a figure has one: the organic trend line, channel performance columns, spend-versus-return comparison, and the AI citation heatmap, each rendered in a consistent designed theme with a one-line takeaway, emailed on schedule with zero assembly work on your side.
Make it yours
- Render the figures in the dark theme when the deck is presented on screen in meetings rather than read in documents.
- Add a competitor slide using the share-of-voice data, so the month's internal numbers sit next to the external picture.
- Tighten the cadence to weekly during launches, keeping the same figure set so week-over-week movement stays visually comparable.
Frequently asked questions
What guarantees the charts are readable and honest?
The chart engine enforces a designed rulebook: a validated colorblind-safe palette, one axis always, sequential color for magnitude, thin marks, and labeled legends. Those rules are applied by construction on every render, which is precisely the consistency hand-built monthly charts never achieve.
Can it use our numbers from systems you do not integrate?
Yes. Any table you include in the mission prompt, or deliver to the workspace as a CSV, can be rendered alongside the connected-source figures. The chart engine draws whatever well-shaped data reaches it.
What does the citation heatmap actually show?
Engines on one axis, your tracked buyer questions on the other, and color encoding how strongly your brand is cited in each cell. One figure answers the question executives actually ask about AI visibility: where are we present, where are we absent, and where did it change.