活动复盘模板

Campaign review template

Build the review only from the data the user provides. Before writing, list the files you received with their date ranges and what each contains, and name any missing input that a section needs (for example the campaign target, the ad report, refund data, or a comparison period). Write [待补：…] in sections that cannot be completed and keep going with the rest. Never invent figures or benchmarks.

Structure

1. Campaign summary: name, platforms, dates (pre-heat, main period, return period), the main mechanics and the target as the user stated it.
2. Results against target: a table of the key metrics (sales, orders, visitors, conversion rate, average order value, promotion spend, ROI, gross margin if cost data is provided, refund rate) with target, actual, completion rate and the comparison period the user chose.
3. Where the result came from: break sales into visitors × conversion rate × average order value; break visitors down by source; show the products that contributed most and least. Quantify each driver's share of the gap or the gain.
4. Paid traffic: spend and return by plan or channel, compared with break-even ROI from the user's margin; which plans earned and which lost.
5. Content and live: when the user supplies the data, sessions or videos with their reach, conversion and sales.
6. Stock and fulfilment: stock-outs, overstock left after the campaign, late shipments, as the data shows.
7. Service and after-sales: refund rate and reasons, negative review themes, consultation volume and response issues.
8. What worked and what did not: three to five points each, every point tied to a number from the sections above.
9. Causes: for each problem, the most likely cause and the evidence; mark inferences as inferences.
10. Next actions: a table of action, owner, deadline and the metric that will show whether it worked; split into things to keep, things to fix and things to stop.

Show the calculation for every derived number. Metric definitions differ between platforms; when combining platforms, note the difference. Keep the report readable for the store owner: findings first, detail in the tables.