Find margin anomalies over the last weeks
Revenue holds steady, yet less money is left at the end. Margin — what a sale leaves after costs — gets counted per item, week by week.
Medium · 25 min · once a week
What you get
Margin by item, weeks 22–25 Item Usual Week 25 Deviation Main factor ART-4471 24% 11% −13 pp fee up 6 pp ART-2210 31% 19% −12 pp promo discount ART-9008 18% 9% −9 pp logistics x1.8 ART-1102 22% 27% +5 pp cheaper supply Cost of goods taken from 1C at shipment date. 6 items have none — they stayed out of the math.
A sample on made-up data — your numbers will be your own.
Who it fits
- The marketplace changed its fees and you need to see which items lost all their profit.
- Purchase prices rose at some suppliers only — you want to see where that already shows.
When it won't work
Cost of goods in 1C isn't current: margin will compute, but the swings describe your reference data, not the market.
How the agent does it
Set depth and threshold
Three or four weeks is usually enough for the normal level to show. Set the threshold in percentage points, or the whole tail of tiny items comes along.
Run the calculation
Margin is computed per item, week by week, and each swing is split into money: price, marketplace fee, logistics, purchase cost.
Check before repricing
Ask the agent to review its own work: where a swing rests on a handful of sales, and where a one-off marketplace penalty is to blame rather than price.
What you set
Connect Ozon, Wildberries and 1C. From you: how many weeks to look back, what size of swing counts as notable, and the date to price cost of goods at.
What you'll need
Starter prompt
Copy the prompt or open it straight in a chat with the agent.
Calculate margin by SKU by week for [period] using Ozon and Wildberries sales and cost of goods from 1C. For each week break the selling price down into marketplace commission, logistics, discounts and promos, and cost of goods. Find SKUs whose margin deviated from their own average for the period by more than [threshold]. For every deviation name the factor with a figure: how many rubles came from the increased commission, how many from the discount, how many from purchase cost. Don't name a cause if the breakdown doesn't show it — write "factor not determined". Take cost of goods as of the shipment date. Where it's missing, don't substitute the category average, move the SKU into a "no cost of goods" list. Output a table by SKU and week and the top deviations in rubles.
Open in chatReview the margin calculation. List: — SKUs where the deviation rests on fewer than [20] sales in a week; — items where cost of goods wasn't taken as of the shipment date or is missing; — cases where margin dropped because of a one-off marketplace compensation or penalty rather than price; — costs you didn't allocate to SKUs and left in a common pile; — deviations where the factor was named on your judgement rather than derived from the breakdown.
A second prompt — the agent uses it to review its own work and show what's left for you.