2 min read
In brief
- Launch size should be set by how many products you can test and maintain per month, not by how many you can import.
- A small catalogue concentrates spend, which is how you learn anything from it at a realistic budget.
- Early catalogues should be built to be cut: the products that survive validation fund the next batch.
The real constraint
At a small daily budget spread across a hundred products, no product receives enough impressions to be judged. The store appears to be running and produces no information, which is the most expensive state a new store can be in.
The constraint is attention as much as money. Every product needs photography that is not obviously reused, a description that is accurate, a delivery expectation you can meet, and a weekly check on its cost per purchase. That is a real workload per SKU.
- If you can test ten products a month, launch ten — not a hundred and test none.
- If your budget is enough for three products to exit learning, launch three.
- If the supplier's catalogue is your product list, you do not have a product list.
The shape of an early catalogue
- 1
One hero product, chosen for margin headroom
It takes most of the budget and most of the creative. Everything else exists to support it.
- 2
Two or three supporting products in the same category
They raise average order value and give the store a reason to look like a store rather than a landing page.
- 3
One deliberate test
A product you expect to fail, run cheaply, so failure costs little and teaches something specific.
- 4
Nothing else
Importing the rest of the supplier's catalogue is not optionality; it is the illusion of it.
Built to be cut
The products that do not clear their break-even CPA in the first fortnight should leave the catalogue. Kept out of fondness, they keep consuming budget and dilute the signal from the ones that work.
A store that launched with ten and cut to four is healthier than one that launched with a hundred and learned nothing about any of them — and its second batch will be better chosen, because the first one produced evidence.
Set launch size by your monitoring capacity, concentrate spend on one product, and treat the first catalogue as a test that produces a better second one.
