The Action Center
Monitor tells you where you're losing. Optimize and Generate tell you how to act. The Action Center keeps workflow, opportunities, and measurement in separate lanes so unlike signals are never disguised as one comparable score.
What the Action Center is
Open a project and go to Act › Action Center and you’ll see three lanes plus a Wins sidebar: work already approved or started, evidence-backed opportunities, and measurement or results. Nothing on this screen is invented for the occasion. Cards come from the Priority measure behind the Optimization Queue, the Opportunity measure behind Content gaps, and the actual workflow, publish, and tracking history in your project.
Action card
One row in the feed. Every card follows the same shape: signal (what the data shows) → recommended action (what to do) → destination (the module that handles it) → evidence (the factual state that put it here). In code, that’s an ActionCard: a kind, a title stated as evidence, a detail, an evidence label, a native rank value used only inside its comparable group, a machine-readable ordering basis, a factual work-scope label when one can be counted, and an href that jumps straight into the right module.
Deterministic, without false precision
Nothing about placement is randomized. Page recommendations compare only with other page Priority values; content and earned opportunities stay in their own groups. Ready work is oldest first and results are newest first. Stable tie-breaks keep cards from reshuffling. There is no invented numeric weight for “publish,” “review,” or “run tracking.”
Anatomy of a card
Here’s a realistic example, close to what you’d actually see — a page that used to answer a comparison question well and has since fallen behind:
Example card
OptimizeTracked page evidencePriority 82
‘Best CRM for Startups’ is losing on 4 matched prompts
Competitors cited on 4 matched prompts · Never cited on its own page
Optimize → jumps into the editor for that page.
Ordering: compared only with other Page Priority cards
Work scope not estimated
Each part of the card maps to a specific, real field:
| Part | What it is | Where it comes from |
|---|---|---|
| Badge | Which action this is | Fixed label per kind, e.g. “Optimize” |
| Title | The signal, stated as evidence — never vague | Built from the page’s stored “why chips” (the same evidence shown in the Optimization Queue) |
| Evidence label | The factual provenance or workflow state | e.g. “Confirmed in 2 batches,” “Approved brief,” or “2 before + 2 after batches” |
| Native measure | Shown only when the source engine produced a real comparable measure | “Priority 82” for page work or “Opportunity 74” for a gap. Workflow and result cards show no generic score. |
| Ordering basis | The exact policy that placed the card inside its group | For example, “compared only with other Page Priority cards” or the timestamp at which approved work began waiting |
| Known work | A factual count or scope already present in the data | For example, “3 accepted fixes to publish” or “2 playbook steps remaining.” It never predicts time or difficulty; unknown scope is shown as not estimated. |
| CTA button | One click into the exact destination screen | A project-relative link, e.g. the editor for that page |
Clicking the CTA fires a small, fire-and-forget beacon so the product can record that the recommendation link was clicked — it does not claim the destination task was started or completed, and it never blocks or delays the click itself.
Evidence tones
The evidence label is color-toned, so you can read a card’s confidence at a glance — from a pure workflow state up to a fully quantified result:
| Tone | What it means |
|---|---|
| Workflow | Queued by workflow state, not a measurement. |
| Observed | Movement seen in answers — not yet confirmed. |
| Confirmed | Repeated in later tracking runs. |
| Measured | Quantified against the pre-publish baseline. |
| Inconclusive | The window closed without a clear read. |
The three lanes
The lane answers a more important question than “which number is biggest?”: is this committed workflow, a new judgment backed by evidence, or measurement of work that already shipped?
| Lane | Card kinds | How it is ordered |
|---|---|---|
| Ready to finish | Generate, Review, Publish, Resume playbook | Oldest approved or started work first; up to 3 cards |
| Evidence-backed opportunities | Optimize, Audit, Content gap, Earned source | Separate Page Priority, Content Opportunity, and Earned Opportunity groups; up to 2 cards in each |
| Measurement & results | Tracking, Impact | One eligible measurement task, then up to 2 newest results; outcome does not change the order |
How each group is ranked
Each group has one defensible ordering rule and its own cap. The caps total 12, but there is no global final slice that lets one kind of card crowd out every other kind.
| Group | What enters it | Order and cap |
|---|---|---|
| Ready or in progress | Approved briefs, generated drafts, accepted fixes ready to publish, and playbooks already in progress | Oldest lifecycle timestamp first; 3 cards |
| Page Priority | Losing pages with real tracking evidence and high-priority pages that have never been audited | Stored Priority, highest first; 2 cards. When the same unaudited page qualifies for both, Audit is shown instead. Publish also replaces Optimize once accepted fixes are ready to go live. |
| Content Opportunity | Open gaps at Opportunity 40+ whose validation status is actionable, with the tracking evidence shown on the card | Stored content Opportunity, highest first; 2 cards |
| Earned Opportunity | Third-party sources cited on at least 2 prompts, excluding work already in progress or verification | Earned Opportunity, highest first; 2 cards |
| Measurement needed | Eligible live publishes whose frozen cohort can still advance; local, draft, legacy, and baseline-deficient publishes are excluded | One project-level tracking card |
| Results ready | Fresh on-site Impact Cards, honest inconclusive notices, and completed earned-play measurements | Newest result first; 2 cards. Completed on-site and earned results use their measured-at timestamp, while an immediate inconclusive on-site notice with no computed result uses its publish timestamp. |
A native measure keeps its meaning
An Optimize card’s Priority is the same Priority shown in the Optimization Queue. It is useful for comparing page work. It is not treated as comparable to a content gap’s Opportunity, the age of an approved brief, or the outcome of an Impact Card. Priority, Content Opportunity, and Earned Opportunity are deterministic in-product ordering indices, not calibrated probabilities of success or predicted lift.
The Wins sidebar
Next to the feed sits a narrower panel called Wins — “receipts from the loop,” as the app puts it. Where the main feed is forward-looking (what to do next), Wins is backward-looking: recent workflow milestones and qualified measurement receipts. A milestone says work advanced; it does not by itself prove that work caused a visibility gain. Wins pulls from four source families:
- Product milestones — a gap that closed without a tracking-scope change, a confirmed live page update, or a draft that finished generating. Local snapshots, WordPress drafts, and non-live updates are omitted.
- On-site citation gains — a cached Impact Card enters Wins only when the exact page moved from 0/2 cited before batches to 2/2 cited after batches. A continuous observed change does not qualify on its own.
- Earned-play receipts — completed off-site playbooks whose own verification window qualifies as a win.
- Win alerts from tracking runs — a prompt you newly won, a citation you newly earned, or a competitor you overtook.
The eight most recent wins show up here, newest first. Every win links back to the page or gap it came from. Wins are covered in full — including how an Impact Card is computed — in Impact Cards & wins.
Note
An Inconclusive Impact Card can appear in Results ready so you can see why no comparison was made. It is ordered by recency like every other result, never downranked as a bad outcome, never added to Wins, and never treated as “No change.”
From measurement to to-do list
This is the piece that makes the whole loop feel like one product instead of three separate tools. Monitor’s tracking runs feed the Priority Score engine in Optimize and the gap-detection engine in Generate. Those engines already do the hard work of measuring pages and gaps. The Action Center collects the results, applies each group’s eligibility threshold, and orders only comparable cards. A Priority under 30, for instance, never generates an Audit card. The result is that you never have to open the dashboard, notice a drop, then separately dig through the queue and the gap list to figure out what it means — the Action Center has already done that connecting for you.
Measure
The Visibility dashboard
Where the signals that feed every Action Center card start — mentions, citations, and score trends.
Fix
The Optimization Queue
The Priority Score and why-chips behind every Optimize and Audit card.
Create
Content gaps
The Opportunity Score behind every Content gap card.
Prove
Impact Cards & wins
What happens after you act: the verification window, the receipts, and the Wins feed.
Key takeaways
- The Action Center separates ready workflow, evidence-backed opportunities, and measurement or results.
- Priority and Opportunity appear only inside the groups where those deterministic native indices are comparable; they are not predicted lift, and operational cards have no invented scores.
- Independent group caps reserve space for each kind of work, with oldest workflow first and newest results first.
- Every card explains its ordering basis and shows a factual known-work count when available; unknown scope is never turned into a duration or difficulty guess.
Next step
Once you act on a card and publish a change, the page doesn’t just disappear from the feed — it enters a verification window. Read Impact Cards & wins to see how OptimizeCamp compares fixed before-and-after cohorts and reports what tracking observed without claiming the publish caused it.