uhren4you/Watches and jewelleryCase Study
From eight products a day to 6,985 in ten weeks.
uhren4you could carry more products than anyone could create. One full-time person managed eight a day, several hundred a month were needed.
By the numbers
37.8 %of weekly revenue from newly created products
+103 %catalogue growth in ten weeks
873working days the same volume would have taken by hand
−59 %return rate versus the shop average
The bottleneck sat in the product data. The range did not grow with supply. It grew with whoever had time to get product data into the shop. At around 6,500 products across 40 to 50 third-party brands, that was the lid on the entire business.
Eight products per day and full-time person, against a need of several hundred a month
Eight out of ten suppliers with no usable data export, delivery as PDF or by fax
Product images and product data collected by hand, brand by brand
One person for the entire Shopify operation, permanently at capacity
No figure on whether a newly created product ever sells
One tool became a whole chain. What started as a brief for product creation now runs from finding the data all the way to measuring the effect. Here are the six parts that carry the most.
Product creationBrand and model number become a complete product. Five input fields per row, everything else is sourced.
EnrichmentA median of 38 structured fields per product: material, diameter, movement, water resistance. They feed filters and feeds.
Multiple languagesEvery new product is created in five languages automatically. Translation used to be a separate step.
Range gap detectionThe catalogue is continuously compared against the relevant market. 30,710 range gaps identified.
Creation autopilotCreates without manual triggering, but only what sits on the approved list of 52 brands.
ReportsDaily report on range gaps, weekly report on revenue per product. 50 reports without manual triggering.
What changed.
Before
Eight products a day, against a need of several hundred a month
Product data and images collected by hand, brand by brand
Translation as a separate, downstream step
Range gaps against the market unknown
No figure on whether a newly created product ever sells
After ten weeks
6,985 products created, 1,400 on the peak day
Five input fields per row, the rest is sourced
Five languages automatically, with no extra step
30,710 range gaps identified and ready to prioritise
Revenue share, activation and brand ranking measurable every week
The revenue was added. The existing range lost nothing. Every attributed share hangs on a product ID. That ID would not exist without the automation. Attribution runs on it, not on an estimate.
Connected to
Sound familiar?
If your catalogue only grows when someone has spare time, the bottleneck is product creation, not the market.