What is one sample worth?

The question

Every store sends samples, and every team wrestles with the same questions: is it worth it, and who should get one? This time we worked backward. Instead of following samples forward, we started from the creators who actually sold, and looked at whether, and when, they had received a sample.

Samples work

Based on our internal data, close to 90% of units are sold by creators who received a sample from the store before they made a sale. Half of all units come from creators who got their sample within the past month.

Who gets it matters more than how many

To see what a sample really adds, we compared creators who got one with creators in the same store, over the same period, selling about as much, who didn't. Sent to a creator already selling your products, one sample adds about 10 units on average. Sent to a creator who has never sold for you, fewer than 2. The same sample, in the right hands, is worth more than six times as much.

Yet in practice, about 90% of samples go to creators who have never sold for the store. New creators are worth developing, since today's newcomer can be tomorrow's top seller. But creators already selling for you shouldn't wait in the same line as everyone else.

Selling for you, but nobody's called

There's another group that gets little attention: creators who sell your products without ever asking for a sample. Since they never apply, review never sees them, and with modest follower counts, outreach based on followers skips them too. In our data there were over a thousand of them, and more than a hundred sold well.

Finding them takes no prediction at all. They're already in your order history. For a store, this is the cheapest, highest-intent outreach list there is.

We checked

You might ask: creators on a roll tend to request samples at exactly that moment, so wouldn't they have sold more anyway? That's why we compared against creators who were selling over the same period but didn't get a sample. After taking out what would have happened anyway, the gap is still clear. We're also running small controlled tests to measure it directly.

How OperaX runs this

• A fast lane for proven sellers. Their sample requests can be flagged and approved first, and new products can reach them before they ask.
• A weekly list of hidden sellers. OperaX can pull creators who sell without a sample out of your orders and line them up for outreach.
• See the result of every sample. Review, shipping, content and sales in one view, so the team knows which samples actually drove sales.

None of these rules are complicated. The hard part is applying them every day, across thousands of creators, without missing one. That's exactly the job of an AI operations team.