All use cases

Surface negative reviews that threaten your rating

Two one-star reviews can pull a listing down in a day. The alert lands the day they appear — the full quote, the rating before and after.

Easy · 15 min · set up once

What you get

Alert 14:20 · SKU 771-04 "Kettle 1.7 L"
Review ★1  "Leaks at the seam after the second use, refunded"
Rating     4.7 → 4.4 (third negative in 24 hours)
Threshold  fired: rating of 2 or less on a priority item
Action     needs a person: looks like a batch, not one unit

A sample on made-up data — your numbers will be your own.

Who it fits

  • A listing's rating slips on two or three reviews, and you need to know the day they are published.
  • "New review" notifications fire on everything, so nobody opens them any more.

When it won't work

Reviews get read once a week in a batch — then you want an analysis, not an alert.

How the agent does it

1

Set the threshold

Decide what a dangerous review is for you: a low rating, a priority product, trigger words, or several negatives in a row on one item.

2

Shape the alert

The notification should settle "do I open the seller account or not" without opening it: the full quote, not a summary, and the rating before and after.

3

Test it on last week

Replay the rule over last week's reviews and count how many alerts it would have sent. Otherwise you learn its quality after a month of silence.

What you set

Connect your Ozon and Wildberries accounts. By hand: the rating that counts as dangerous, your priority products, and where alerts should go.

What you'll need

Ozon
Wildberries

Starter prompt

Copy the prompt or open it straight in a chat with the agent.

Prompt for the agent

Watch new reviews on Ozon and Wildberries and flag only what threatens the rating. Threshold: rating [1–2]; or [3] and below for an SKU from the list [priority SKUs]; or [2] negative reviews on one SKU within [24 hours]; or the words [defect, scam, swap, court, consumer protection agency]. One review — one alert, do not write about the same one twice. Do not raise an alarm for a rating without text or for a review about the marketplace's delivery — collect those in a separate summary once per [period]. In the notification: SKU and name, rating, date, the full quote from the review, the listing's current rating and how much it moved, how many reviews this SKU got in [7 days].

Open in chat
Check the result

Check the trigger rule against the reviews from [the past week]. List: — how many alerts would have gone out per day and at which hours they cluster; — reviews that crossed the threshold because of a single word rather than their substance; — dangerous reviews the rule missed — and which condition was lacking; — repeat alerts on the same review or SKU; — SKUs where two negatives a day is ordinary background rather than a spike; — which threshold you propose to change and to what, leave the decision to me.

A second prompt — the agent uses it to review its own work and show what's left for you.

Related use cases