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

# Analytics Alerts

> Set rules that watch revenue, orders, returns, stock-out losses and Search Console clicks every night, and read the feed of what changed while you were away.

Alerts watch your numbers while you are not looking. You write a rule once, WISEPIM checks it every night against the last complete day, and anything that trips lands in a feed the whole project shares. Google Merchant Center drift and other automatic findings land in the same feed.

**What this page enables:** you stop opening five reports every morning to see whether something broke. The feed tells you, with the measured value next to the threshold, and one click takes you to the report to investigate.

**Where to find it:** open **Analytics** and click **Alerts** in the sidebar. The page has two tabs, **Feed** and **Rules**.

<Info>
  Rules belong to the project and everyone on it shares them. Alerts show up in WISEPIM and as an **Analytics Alert** [notification](/en/settings/notifications). Email delivery is not available yet.
</Info>

## Create a rule

<Steps>
  <Step title="Open the Rules tab">
    On the Alerts page, switch to **Rules** and click **New rule**. A project can hold up to 50 rules; the counter next to the button shows how many you have used.
  </Step>

  <Step title="Name it and pick what to watch">
    Give it a **Rule name** you will still recognise in a month, such as "Revenue floor". Under **Watch**, choose one of:

    * **Daily revenue**
    * **Daily orders**
    * **Return rate**
    * **Lost revenue from stock-outs**
    * **Daily Search Console clicks** (needs a connected [Search Console property](/en/integrations/google-search-console))
  </Step>

  <Step title="Choose the condition">
    Under **Alert me when it**, pick **Falls below** or **Rises above** a threshold, or **Deviates from its own baseline**. Threshold rules also work as a percentage change against the previous period or the same period last year.

    The baseline option compares the metric with its own recent history and fires when it moves more than a number of **Standard deviations** away. 3 is calm, 2 is jumpy. Stock-out loss has no day-by-day history, so it only supports thresholds.
  </Step>

  <Step title="Set the window, severity and cooldown">
    * **Window (days)**: how many days back the rule looks each night, 1 to 365, default 7.
    * **Severity**: **Critical**, **Warning**, **Opportunity**, **Positive** or **Info**. These are the same levels [Smart Insights](/en/analytics/smart-insights) uses.
    * **Cooldown (hours)**: how long the rule stays quiet after firing, 0 to 720, default 24. At 0 it can tell you every day the condition holds.
  </Step>

  <Step title="Create the rule">
    Leave **Start watching right away** on and click **Create rule**. The rule is checked for the first time in the next nightly run.
  </Step>
</Steps>

<Warning>
  For revenue and orders, a threshold is compared against the **daily average** over the window, not the total. With a 7-day window, "below 500" fires when the average day came in under 500. Return rate is compared against the rate across the whole window, in percentage points.
</Warning>

### Manage rules

The **Rules** table lists every rule with its summary, severity, cooldown and when it last fired. Use the **On** switch to pause a rule without deleting it. A rule in cooldown shows **Quiet right now**. Deleting a rule stops it immediately; alerts it already raised stay in the feed.

## Read the feed

Rules run every night at 06:45 against yesterday, the last complete day. The **Feed** tab shows what they found, newest first, with a freshness line such as **Rules last ran today**. Filter by **Status** and **Severity**; the date range control is disabled here because alerts are not filtered by date.

Each alert card shows:

* **What was measured** against the **Threshold**, or the **Deviation** from the **Baseline** for baseline rules, plus the products involved when the rule scoped them.
* **Rule: name** for alerts raised by your rules, or **Raised automatically** for system findings such as **Channel drift**.
* An impact chip when WISEPIM can estimate it: an amount **recoverable**, **at risk** or **upside**.
* **Seen 3x** when the same condition held on several nights. A re-run never raises a duplicate; it bumps the count.

### Statuses

| Status        | Meaning                                                                        |
| ------------- | ------------------------------------------------------------------------------ |
| **New**       | Nobody on the project has looked at it yet                                     |
| **Read**      | Someone clicked **Mark read**                                                  |
| **Dismissed** | Someone clicked **Dismiss**. This clears the alert for everyone on the project |
| **Resolved**  | WISEPIM set this itself when the condition recovered                           |

**Mark all read** in the header clears the unread badge in one go.

### Channel drift alerts

Alerts titled "Google is serving different data for {sku}" come from the nightly [Merchant Center drift scan](/en/analytics/integration-health), not from a rule. Their **Intended vs live** panel shows the **Title**, **Price** and **Availability** as they are **In WISEPIM** and **Live on Google Merchant Center**, plus the probable cause: the last sync failed, your change is waiting for the next sync, or the snapshot is too old to tell.

## Investigate an alert

Click **Investigate** on any alert to jump into the report behind it with the same date window, comparison and filters already applied. From there you can save the view as an [investigation](/en/analytics/overview#saved-investigations) so teammates see exactly what you saw, and attach the rule to that investigation to keep monitoring it.

## Act on what you find

<AccordionGroup>
  <Accordion title="A rule fires every single day">
    Either the threshold is too tight or the condition is real and nobody fixed it. Raise the **Cooldown** so it stops repeating, or switch the rule to **Deviates from its own baseline** so it only fires on unusual days rather than on every day below a fixed number.
  </Accordion>

  <Accordion title="A rule never fires and you doubt it works">
    Check the **Last fired** column and the freshness line on the feed. If rules ran today, the condition simply did not hold. Remember that revenue and order thresholds compare against the daily average of the window, so a single bad day inside a 30-day window rarely trips it. Shorten the window to react faster.
  </Accordion>

  <Accordion title="A Search Console rule raised an error instead of an alert">
    The rule watches a metric the project cannot read. Connect [Google Search Console](/en/integrations/google-search-console) first, or delete the rule.
  </Accordion>

  <Accordion title="The feed is full of drift alerts after a failed export">
    Fix the export first in [Integration Health](/en/analytics/integration-health), then re-export. Drift alerts resolve themselves on the next nightly scan once Google serves the right values again.
  </Accordion>
</AccordionGroup>

## Related

<CardGroup cols={2}>
  <Card title="Smart Insights" icon="lightbulb" href="/en/analytics/smart-insights">
    Findings WISEPIM raises on its own, using the same severity levels.
  </Card>

  <Card title="Integration Health" icon="heart-pulse" href="/en/analytics/integration-health">
    Sync status per channel and the Merchant Center drift scan.
  </Card>

  <Card title="Notification Settings" icon="bell" href="/en/settings/notifications">
    Turn the Analytics Alert notification on or off.
  </Card>

  <Card title="Revenue & Sales" icon="chart-line" href="/en/analytics/revenue-sales">
    The report behind daily revenue and order alerts.
  </Card>
</CardGroup>