Verified 2026-05-03. Number of content brief routes (model tiers per content type) drifts as features ship — verify against agent registry before quoting in pitches.

05 — Actions & Content

The fifth stage of the loop. A Booster from Stage 4 becomes a tracked Action, which becomes content, which becomes a published URL, which gets monitored for AI citations in the next collection cycle. This is where Nudg3 stops being a metrics tool and becomes a workflow.

Actions Pipeline

Actions is the project management layer that turns report recommendations into measurable work.

The workflow

1. Report generates Boosters — opportunities, threats, quick wins, strategies (Stage 4)
2. User selects Boosters — clicks “Add to Actions” on the ones worth pursuing
3. Actions become a work queue — with status tracking (pending → in progress → ready → completed / dismissed), priority, team assignment, optional user assignment
4. Content briefs are generated — AI generates content briefs and drafts for content-type actions. Multiple model routes per content type. Users can refine briefs iteratively.
5. Published URLs are tracked — users add the URL of the published content; the system monitors whether AI providers cite it in future responses (back-feeds into Stage 1)
6. Feedback loops into next report — completed/dismissed/pending actions feed back into the next visibility report (Stage 4 + Stage 6)

Why “actions are never auto-created”

Every Action was explicitly promoted from a Booster by a user. The platform doesn’t create work behind the user’s back. This matters because:

• Trust — clients want to see their judgement on what to act on, not the platform’s
• Quality signal — the dismiss rate on Boosters is a direct quality measurement we use to improve future Booster generation
• Auditability — every action has a human author and a reason

Content Generation

For content-type actions (e.g. “write a comparison article”), the platform generates:

• A content brief (audience, intent, structure, target keywords, distinctive angle)
• A first-draft document via the chosen model route
• Iterative refinement on user feedback

Content lives in the action record until published. After publish, the URL is registered for citation monitoring.

URL Monitoring (the ROI loop)

When a user marks an Action as completed and supplies a URL:

1. The URL is registered against the workspace’s monitoring set
2. Future collection cycles (Stage 1) capture every response that cites that URL or domain
3. Citation appearances feed Stage 2 extraction normally
4. Aggregated stats (Stage 3) credit the action with citation events
5. Future reports (Stage 4) show the trend — citations earned, score delta since publish

Over time, the actions pipeline builds a history of what content was published, whether it earned AI citations, and how visibility changed. This is the measurable link between specific actions and score improvement — the moat versus competitors that stop at metrics.

See also

• Stage 4 — Reports & Boosters — where Actions originate
• Stage 6 — Historical Tracking — how URL-monitoring outcomes show up in trend lines
• Stage 1 — Collections & Prompts — the loop closes here when next-cycle queries pick up cited URLs