> ## Documentation Index
> Fetch the complete documentation index at: https://docs.neariq.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Action Center

> Use the Action Center to review prioritized next-best actions without sending anything automatically.

## Action Center

The Action Center ranks the next useful work from the signals NearIQ already has: leads, conversations, review requests, appointment reminders, calls, reviews, ranks, citations, competitors, market trends, and AI visibility evidence.

It is designed for review-first operations. NearIQ can route you to the right workflow, but the Action Center does not silently send messages, publish posts, or create paid provider requests.

## How actions are ranked

Actions are grouped by revenue proximity:

| Tier                | What it means                                                                       |
| ------------------- | ----------------------------------------------------------------------------------- |
| Money now           | Missed calls, hot leads, quote recovery, and other near-term revenue opportunities. |
| Reputation risk     | Reviews or public questions that can affect trust if they sit unanswered.           |
| Convert and retain  | Review requests, reminders, and customer follow-up that keep the loop moving.       |
| Competitive opening | Competitor changes, ads, closures, or market signals worth acting on.               |
| Visibility          | Rank, citation, AI visibility, and profile fixes that improve discoverability.      |

Every item includes the evidence that triggered it, a workflow destination, and a send-readiness state. SMS and email-related actions must pass the existing consent and opt-out checks before any outbound workflow is used.

Recent action feedback now nudges ordering inside each tier. Repeated dismissals, snoozes, consent blocks, and send failures lower the confidence and urgency of that action type for the business, while repeated clicks, completions, and auto-executed actions nudge it up. The hard tier order still dominates, so a low-scoring money-now item continues to outrank a high-scoring visibility item.

## Dashboard behavior

Open **Dashboard -> Action Center** to see the full queue. The Overview page also shows a compact **Next actions** metric that links to the same tab.

Use the primary action button to jump to the existing workflow, such as Messaging, Reviews, Contacts, Rank Tracker, Citations, Keywords, or AI Visibility. Schema-backed action rows also support persistent user status changes for done, dismissed, and snoozed states, and every transition is recorded in the append-only action event log.

## Persistence foundation

NearIQ includes a persistent Action Center schema for the follow-up engine: `next_best_actions` stores queued actions, `next_best_action_events` stores append-only shown, clicked, done, dismissed, snoozed, auto-executed, consent-blocked, and send-failed events, and `next_best_action_settings` stores per-business automation defaults, paused types, and quiet-hour preferences.

The dashboard's computed `GET /api/businesses/me/action-center` endpoint remains review-first and read-only. The schema-backed `GET /api/businesses/me/next-best-actions` API lists persisted rows, updates persistent user status changes, stores deterministic reviewed drafts, exposes owner/admin automation settings, and can execute manual or SMS actions only after explicit confirmation and compliance checks. Email, profile, social, and ad actions route to their existing workflows instead of sending or publishing directly from the Action Center API.

Reusable compliance and duplicate-prevention checks are now in place for those writes. They check SMS consent and opt-out state, SMS plan access, quiet hours, known email unsubscribe state, paused action types, snoozes, cooldowns, and overage policy before NearIQ sends or saves an action.

Feedback summaries are aggregate counts keyed by action type. They do not include customer names, message text, phone numbers, emails, provider payloads, or prompt content. If the account has opted out of training-data collection, NearIQ ignores those feedback counts for ordering.

## API

The dashboard uses `GET /api/businesses/me/action-center` for the computed view. Persistent queue clients use:

* `GET /api/businesses/me/next-best-actions`
* `PATCH /api/businesses/me/next-best-actions/{id}`
* `POST /api/businesses/me/next-best-actions/{id}/draft`
* `POST /api/businesses/me/next-best-actions/{id}/execute`
* `GET/PATCH /api/businesses/me/next-best-actions/settings`