Anomalies

Anomalies help your team catch unusual review patterns before they turn into bigger problems. Instead of one notification per review, AppReviewBot looks at reviews in aggregate and alerts you when something looks off.

What is an anomaly?

An anomaly is a statistical signal that recent reviews for a tracked app differ sharply from that app’s normal baseline.

Examples:

• A sudden burst of 1–2★ reviews after a release
• Many more reviews than usual in a short window
• Average rating falling quickly
• Most new reviews being low-rated
• A flood of 1★ reviews in a few hours

These alerts complement your normal review notifications. Per-review alerts keep you close to every new rating; anomalies highlight patterns that are easy to miss in a busy channel.

How tracking works

AppReviewBot already stores reviews for the apps you track. Anomaly detection periodically compares:

1. A recent window (for example, the last day — or a shorter window for 1★ floods)
2. A baseline from the days before that window

If recent activity crosses clear thresholds versus the baseline, AppReviewBot opens an anomaly report and can notify your team.

Detection uses review statistics from your stored reviews — not AI guesswork. Checks run on a regular schedule (about every 12 hours), so you get a focused summary when a pattern looks real, without flooding your chat with noise.

What AppReviewBot can detect

Each alert has a severity (warning or critical) so your team can prioritize.

Where alerts are sent

Anomaly alerts can go to destinations you use with AppReviewBot, such as:

• Slack
• Discord
• Microsoft Teams
• Email

Anomaly destinations are chosen separately from day-to-day per-review routing, so you can send incident-style summaries to an ops or leadership channel without changing every subscription’s normal alert destinations.

A typical alert includes:

• App name and store
• Which pattern was detected
• Severity and a short summary of the numbers
• Sample low-rated reviews when available
• A View anomaly report link into the dashboard

View anomalies in the dashboard

In the AppReviewBot dashboard, open Anomalies to see reports for your active workspace.

From the list you can:

• Filter by app, status, or severity
• Open a report for metrics, the detection window, and sample reviews
• Jump from a sample review into the full review when it is still available

Reports stay in history with statuses such as open, resolved, or suppressed so you can revisit past incidents. On a report, you can Mark resolved, Suppress, or Reopen so the list filter matches how your team is tracking the incident.

If you open a report link that belongs to another workspace you have access to, AppReviewBot will ask you to switch workspaces first.

How to investigate a report

When you open an anomaly report, check:

1. What changed — recent counts, low-rating ratio, and average vs baseline
2. When — the detection window on the report
3. Sample reviews — common themes, versions, or countries called out by users
4. Your recent releases — correlate with a ship, pricing change, or outage

From there, reply to users, file product tickets, or Share the report link with the team that owns the app. The link opens a read-only report page with metrics and sample reviews. If the anomaly is removed later, the shared page will show that the report is unavailable.

Tips for getting value

• Keep apps you care about tracked so reviews (and therefore anomalies) stay up to date
• Route anomaly alerts to a channel people actually watch during incidents
• Treat anomalies as a signal to investigate — not every spike is a crisis, but most deserve a look
• Remember that app stores moderate and delay reviews; see Why reviews are not real-time

Related guides

• Add your first app
• Configure destinations
• Ratings vs reviews
• Ratings & Reviews FAQ