Show the trend of negative reviews by week
You shipped a fix and cannot tell whether reviews improved or it just feels that way. The agent charts negative reviews by week around your fix dates.
Medium · 20 min · once a week
What you get
Share of 1-2 star reviews, Ozon and Wildberries wk 10 11% ##### wk 11 13% ###### wk 12 9% #### <- Mar 18: packaging fix shipped wk 13 6% ### Topic "damaged packaging": 34 reviews before, 9 after. Topic "wrong size" unchanged — the fix did not touch it. Week 14: 12 reviews, too few to conclude anything.
A sample on made-up data — your numbers will be your own.
Who it fits
- You shipped a fix and want to know whether it showed up in the reviews or only feels like it did.
- You watch quality weekly, and what matters is the direction, not the fact of a bad review.
When it won't work
A product with three or four reviews a week: one person swings the share and there is no trend.
How the agent does it
Mark the fix dates
Say which ratings count as negative and when each fix went out. Without the dates you get a curve with no link to anything you actually did.
Run the analysis
The agent computes the weekly share, breaks it down by topic and compares the weeks before and after each date you marked.
Check what backs it up
Have the agent name the weeks with too few reviews, and the cases where negativity grew alongside total volume while the share stayed flat.
What you set
Ozon and Wildberries seller accounts with review access. By hand: the period, which ratings count as negative, and the dates your fixes shipped.
What you'll need
Starter prompt
Copy the prompt or open it straight in a chat with the agent.
Build the week-by-week dynamics of negative reviews on Ozon and Wildberries for [period]. Count [1–2 stars] as negative. For each week: the share of negatives among all reviews that week, the number of reviews, a breakdown by complaint topic. Count the share specifically: growing sales lift the number of reviews too, and that in itself is not a rise in negativity. Mark the dates [list of shipped fixes] and for each topic show the share before and after. If less than [3] weeks have passed since the date or the week has fewer than [20] reviews — write "too early to judge" and do not call it an improvement. Output a week-by-week table, a breakdown by topic and a list of topics where the share of negatives has grown for the third week in a row.
Open in chatCheck the dynamics. List: — weeks with fewer than [20] reviews where the share is computed on too small a sample; — weeks with incomplete data: promotions, sales, gaps in reviews; — topics where the share changed after a fix but the sample does not allow talking about an effect; — cases where negativity grew along with the total number of reviews and did not change as a share; — topics you merged or renamed compared with the previous run; — which of the "fix → reviews" links are visible in the data and which are an assumption.
A second prompt — the agent uses it to review its own work and show what's left for you.