How to Group ASINs

A Lottery campaign is only as useful as the ASIN group behind it.

The campaign setup can be technically correct and still produce confusing data if the seller groups the wrong products together. Low bids, down-only bidding, automatic targeting, and budget caps help control cost, but they do not automatically make the results easy to interpret. The products inside the campaign determine whether the future data will lead to clean decisions or messy guesses.

This is why ASIN grouping matters. A Lottery campaign is not simply a place to dump every product that needs traffic. It is a testing environment. The ASINs inside that environment should be similar enough that the seller can interpret what happens to them.

It helps to be precise about what an ASIN is on Amazon Merch on Demand, because it shapes how grouping works. When a seller uploads a design, they choose which marketplaces and which product types carry it. Each unique combination of design, product type, and marketplace is one ASIN. Within a single ASIN, the colors, fit types, sizes, and price are variations of that same ASIN, not separate ASINs. So the same design on a US standard t-shirt is one ASIN; the same design on a US hoodie is a different ASIN; and the same design on a UK standard t-shirt is another ASIN again. This is why the grouping factors below are marketplace and product type rather than color or fit: those are the axes that actually create separate ASINs, and therefore separate units of evidence.

Group ASINs by the factors that change interpretation.

That is the core rule. Products should be grouped together when they can be judged through similar economics, timing, buyer context, product role, and campaign purpose. Products should be separated when mixing them would make the next decision unclear.

The first grouping factor is marketplace. A US campaign, UK campaign, and DE campaign should not be treated as the same test. Marketplaces can differ in buyer language, competition, price expectations, royalty structure, seasonality, and search behavior. A phrase that works in the US may not translate cleanly into another marketplace. A product type that converts well in one store may behave differently in another.

The second factor is product type. A standard t-shirt, hoodie, sweatshirt, mug, and phone grip do not carry the same buyer expectation or royalty economics.

Mixing them inside the same Lottery campaign may make the campaign average difficult to interpret:

• If a hoodie receives fewer clicks, is that because the design is weak, the price is higher, the season is wrong, or the product type requires stronger buyer commitment?
• If a mug converts better, is the idea stronger, or is the buying context more giftable?

Product type separation does not need to be perfect in every account, but it matters whenever the seller wants clean decisions. A standard t-shirt campaign should usually not be judged against mug performance or hoodie performance as if all products carry the same economic and conversion expectations.