Estimatein 1 hourEstimate a projectreply within 1 hourA sales analytics system for Wildberries sellers: official API access, product-card parsing, and clear purchase recommendations inside one cabinet.
A Wildberries seller lacked a clear picture: official reports covered only part of the numbers, while restock and assortment decisions were still made by gut feel. They needed a system that unifies sales, orders and product cards in one cabinet and suggests what to reorder from sell-through speed — not from scattered marketplace exports.
Some data comes from the official Wildberries API; prices, reviews, visibility and attributes often live only on product pages. Both channels had to be joined reliably, SKUs normalized, and history kept when WB’s feed or API methods change — otherwise analytics quickly lost the purchasing team’s trust.
We built an analytics service with background collectors, dashboards and automatic recommendations based on sell-through speed. The client’s team got one contour instead of spreadsheets and manual exports, and could decide on assortment from a single agreed metric model.
The core is Laravel with PHP queues; large cabinet tables use AG-Grid for client-side filters and grouping. Workers pull Wildberries API reports and crawl product cards with rate limits. Raw payloads normalize into one model of SKUs, sales, stock and logistics so recommendations match the manager screen.
The analytics layer computes sell-through, turnover and recommendations; the cabinet exposes slices as tables and charts. Critical API/parse mismatches are logged for manual review so purchasing sees where data temporarily diverges instead of deciding blind.
Active development took about four months, then support and metric tuning to the client’s purchasing process. The contour stays extensible for new Wildberries reports and extra slices without a full rewrite of the analytics core.
The client got a clear analytics system and a way to manage new-product orders from automatic sell-through recommendations — instead of scattered marketplace reports and hand-built Excel. Sales, stock and cards live in one cabinet, and restock decisions rest on shared team metrics.
Restock decisions now rest on agreed indicators: stock, order dynamics and turnover without jumping between the WB cabinet and local spreadsheets. The team spends less time assembling data and reacts faster to SKU dips and demand spikes.
Splitting API and parsing layers made post-launch support practical and lets the product absorb Wildberries changes without a full analytics rewrite. The contour is ready for new reports and slices while the business scales assortment on the marketplace.
Describe the task — we will name the timeline and price within an hour.
