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

# Review Analytics

> What your guests are telling you, across four lenses on the same reviews.

**Review Analytics** is where individual reviews become a picture. Four tabs, each a different lens, with a period selector on the ones where a period makes sense.

***

## What is on the screen

### Overview

A written briefing that compares the **last 7 days against the 7 before them**, alongside what is open and at risk right now. It has no period selector because its window is fixed at a week — but it is a rolling week, so it moves with you rather than resetting on a Monday.

**Ask Recep** lives here: a box you type a plain question into — *"what are guests complaining about this month"* — answered from your own reviews.

### In-House

Feedback collected directly in RecepAI rather than from a review site: given **during the stay**, answered in a **post-stay** survey, or added **manually** by your team. This is the half that never reaches a public site.

### Online Reputation

Your published standing: your rating over time, and how it sits against nearby hotels. Your own line is drawn solid; the neighbourhood is context, not a competition table.

### Guest Pulse

A deeper report that reads the actual language of your reviews and groups it into what guests keep raising — the topic, not the star count. It is produced on its own schedule rather than recalculated as you watch.

***

## What you do here

<Steps>
  <Step title="Start on Overview">
    It compares this week with last, and shows what is open now. It is the fastest read of whether something moved.
  </Step>

  <Step title="Use In-House to catch what never went public">
    A complaint made in the room is the one you could still have fixed. If In-House is negative while your online rating holds, you are being given warnings.
  </Step>

  <Step title="Read Online Reputation as position, not score">
    Your own rating on its own says little. Seen against nearby hotels it tells you where you sit — which is a starting point for a conversation, not a conclusion about why.
  </Step>

  <Step title="Take Guest Pulse to a department, not to a meeting">
    Its output is topics. "Breakfast queue" is something housekeeping and F\&B can act on; "4.2 stars" is not.
  </Step>
</Steps>

<Info>
  **The four tabs do not move together.** Overview reads the last week, Online Reputation follows published reviews on their own timeline, and Guest Pulse is produced on its own schedule. A quiet Overview does not mean the others have nothing new.
</Info>

<Tip>
  **Where the two halves disagree is worth a look.** Guests who raise something in the room but not online leave a gap between In-House and your public rating — not a verdict, but a place to go looking.
</Tip>

***

## What happens next

Nothing changes from this screen. It points at work elsewhere — an unanswered review in the [Inbox](/front-desk/reviews/inbox), a recurring fault as a work order on [Active Requests](/front-desk/requests/active), or a survey setting under [Reviews](/front-desk/settings/reviews).