# The strategist panel

> The "Strategist" button above the Opportunities table opens an AI briefing of your project: the next move it recommends, counts of the work still waiting with a link to each, and when it was generated.

sourcesignal documentation, Opportunities. Updated 2026-10-18. The page for people: https://www.sourcesignal.ai/docs/opportunities/the-strategist-panel. Every article: https://www.sourcesignal.ai/llms.txt

The strategist panel is the briefing above the Opportunities table. You open it when you want one answer to "what should I do next in this project?" rather than a page-by-page read of the queue. The AI reads the project and names the most important focus right now. Beside its read, the panel counts the work still waiting, each count with a link to where you can clear it.

Use it at the start of a working session, after a data update has landed, or whenever the queue has grown faster than you can read it. It is a read, not an action: nothing moves until you act on a row, a selection or a status.

## What you see

The "Strategist" button sits at the top right of the Opportunities masthead. It opens a side panel with the eyebrow "Strategist".

### Before the first briefing

The panel reads "What to do next, grounded in this project's state." with one button, "Generate strategist briefing". Press it and the label counts the seconds while it works, and placeholder lines hold the space. When the briefing lands, a toast reads "Strategist briefing generated" and the panel fills.

### The briefing

The headline comes first: one sentence that names the most important focus right now, set in the panel's display type. Under it sit "Regenerate" and the date and time the briefing was generated. Then four sections, top to bottom:

- **Tactical pulse.** One line per count that is above zero: an icon, the number, what it is, an explainer behind a hint, and a link to where that work is. Overdue follow-ups show in red and stale pitches in amber; every other row is neutral.
- **Strategic read.** One to three short paragraphs on where the project stands and what to focus on next, written from this project's data.
- **Themes.** Up to five themes, each a short label with a short note on what to do and why.
- **Rationale.** Folded by default. Open it to see the evidence from this project's data that the recommendation rests on.

### The pulse rows

| Row | What it counts | Where the link goes |
|---|---|---|
| "N opportunities ungrouped" | Pages with no page type yet | The table, with those pages selected (the link reads "ungrouped") |
| "N opportunities unscored" | Pages with no priority yet | The table, selected ("unscored") |
| "N opportunities unclassified" | Pages with no opportunity type yet | The table, selected ("unclassified") |
| "N opportunities without recommendations" | Pages with no recommended action yet | The table, selected ("without recs") |
| "N actionable opportunities without a draft" | Pages whose opportunity type is Outreach, Content or PR/Authority, with no draft yet | The table, selected ("without drafts") |
| "N follow-ups overdue" | Pages whose follow-up date has passed, apart from those at Live placement, Lost, Parked, Rejected or Live already | Reports, the Outcomes tab, with "Follow-ups due" open ("follow-ups") |
| "N pitches stale with no measured lift" | Pages pitched two weeks ago or more with no measured rise in mentions or citations since | Reports, the Outcomes tab, with "Pitched or later" open ("pitched") |
| "N prompts not yet classified" | Prompts in this project the AI has not classified yet | The Prompts page ("prompts") |

A row that would read zero is left out, and when every count is zero the section is not shown. A short pulse is good news: you have already cleared the work it would have named. The counts cover the whole project, including the Parked, Rejected and Live already pages the default table view hides.

## What you can do

- **Generate or regenerate the briefing.** "Generate strategist briefing" the first time, "Regenerate" after that.
- **Follow a pulse row.** The panel closes and you land where the work is. For the five opportunity rows, the table opens on "All time" with every matching page selected once it has finished loading, so the selection bar is ready. For the two outreach rows, you land in your report's Outcomes tab with that working list unfolded (see [Impact and outcomes](https://www.sourcesignal.ai/docs/reports/impact-and-outcomes.md)). The prompts row opens the Prompts page (see [Prompts](https://www.sourcesignal.ai/docs/tracking/prompts.md)).
- **Read the rationale.** Open the "Rationale" fold before you act on a theme, to see which evidence it rests on.

> **Heads up:** Regenerating replaces the current briefing, and so does the AI review when a run ends. The panel keeps no history. If you want to keep a headline or a theme, copy it first.

## How to clear a pulse row

1. Open Opportunities and press "Strategist".
2. In "Tactical pulse", find the row you want to clear and follow its link.
3. In the table, the matching pages are already selected. When the selection matches one of the quick bar's "Select" shortcuts exactly, the menu button reads "Selected:" with that shortcut's name, for example "Selected: Unscored".
4. Review the selection. Uncheck anything you do not want to run, or add rows: the selection is yours once it lands. The AI steps take at most 1,000 selected pages at a time. Selected pages the current view hides (Parked, Rejected or Live already) are left alone unless you press "Include all N" under the bar.
5. From the selection bar, press the step that fills the gap: "Group pages" for ungrouped pages, "Score priority" for unscored, "Re-classify" for unclassified, "Recommend outreach" for pages without recommendations, "Generate drafts" for pages without a draft. On a narrow screen some of these sit under "More actions". Progress shows in the engine bar. The selection bar is described in [Bulk actions and promoting to outreach](https://www.sourcesignal.ai/docs/opportunities/bulk-actions-and-promoting-to-outreach.md).
6. Come back to the panel and press "Regenerate" when you want the counts and the read to reflect what you did.

> **Tip:** Work the pulse from the top. The rows lower down depend on the ones above: a page only counts under "without a draft" once its opportunity type is Outreach, Content or PR/Authority, so classifying first can change the rows below it.

## When it refreshes

You generate it on demand. The AI review also rewrites the briefing on its own at the end of a run: the run that follows a data update that brought new answers, and any run you start with "Run the AI review" above the table (see [The answer feed and data updates](https://www.sourcesignal.ai/docs/tracking/the-answer-feed-and-data-updates.md)). The date beside "Regenerate" tells you how fresh the briefing is.

While a run is in progress for this project, the panel shows a "Pipeline running" chip and the generate button is off. Its hint says why: "The pipeline refreshes this briefing automatically when it finishes." The button comes back the moment the run ends.

## Limits

- A project needs at least one cited page before a briefing can be written. On a new project the "Strategist" button appears once the first pages land, and a request on a project with no pages answers "No opportunities yet. Strategist needs imported data to analyze."
- Generating a briefing is an AI action, so it uses the account's AI allowance. No plan limits that allowance today: the work is counted for the account, not capped. See [Usage](https://www.sourcesignal.ai/docs/account-and-billing/usage.md).
- An account whose trial has ended, or that has an open invoice, cannot start new AI work. The message points you to Billing and confirms that your data is safe.
- The pulse counts are taken when the briefing is generated. A link's selection is taken from the table when you arrive, so the two can differ when work has happened since, or in a project with more pages than the table loads.

## Good to know

- The briefing reads the project as a whole, not your filtered table. Changing the analysis window or your filters does not change what it says.
- The pulse counts are counted, not written by the AI, so the same data gives the same numbers every time. The headline, the strategic read, the themes and the rationale are the AI's words and can be phrased differently from one run to the next.
- The briefing never moves a page. Every change still goes through you: a status you set, an action you press on the selection bar, or a sequence you approve.
- Spot, the in-app agent, can read the current briefing when you ask about it. See [Spot, the in-app agent](https://www.sourcesignal.ai/docs/spot-and-agents/spot-the-in-app-agent.md).
- Movement in the strategic read is an association, not proof. "Mentions rose after the pitch" means the two happened in that order, nothing more.

## Related

- [The queue, views and filters](https://www.sourcesignal.ai/docs/opportunities/the-queue-views-and-filters.md): The Opportunities table lists every page AI engines cite for your prompts, ordered by priority. Narrow it with filters, quick chips and saved views, sort it, export it, and follow a prompt straight to the pages it cites.
- [Bulk actions and promoting to outreach](https://www.sourcesignal.ai/docs/opportunities/bulk-actions-and-promoting-to-outreach.md): Check rows in the Opportunities table and the selection bar acts on all of them at once: promote them to outreach, find contacts, approve, park, reject, or rerun an AI step. Promoting creates drafts that wait for your approval; nothing sends.
- [Impact and outcomes](https://www.sourcesignal.ai/docs/reports/impact-and-outcomes.md): See what followed your work: prompts newly won and lost, the movement after each pitch and each win, and the weekly what-worked read. Every read reports association, never cause.
- [The answer feed and data updates](https://www.sourcesignal.ai/docs/tracking/the-answer-feed-and-data-updates.md): Connect one Ahrefs Brand Radar report, managed for you or your own, and the daily data update brings in AI answers, cited pages and visibility every morning. The Data tab keeps every run, the full schedule and the way to fix a feed that stops.
- [The engine room and pipeline](https://www.sourcesignal.ai/docs/autopilot/the-engine-room-and-pipeline.md): Where you watch the autopilot work: the engine room shows the run in progress and your levers, and the Pipeline page shows the steps your data moves through, each with a status and when it moves next.