Every D2C brand has a bestseller. It leads the homepage, carries the ads, and tops every revenue report. What almost no one measures is what happens to the customers who enter the brand through it.
In a recurring model the first order is not a transaction. It sets the ceiling on everything that follows: if the first product disappoints, the second order rarely happens. Brands spend heavily to win that first customer, then leave the most important variable in the equation unmanaged – which product the customer arrives through.
A bestseller that was costing a brand its customers
A US client of ours sells soups in pouches. One meat-based soup was the clear bestseller summing up to 30% all first sales, and by default it was everywhere – homepage, landing pages, ads, banners. It sold the most, so it got promoted the most, and the promotion made it sell more.
Another soup, a cheese-based one, barely moved (2% of all first sales). It was not promoted, it was more demanding to fulfill, and on every standard report it looked like a product to phase out.
Entry product analysis in Sublime inverted the picture. Customers who entered through the bestseller came back at a rate well below the store’s average – something in that first experience was suppressing the second order. Customers who entered through the cheese soup returned 5x more often and were worth 8x more over the following year, despite the higher cost of serving them.
The bestseller was not a bad product. It was a bad door.
Meat-based soup: ROAS 1.76, 30% of first sales, LTV: 81
Cheese-based: ROAS 0.76, 2% of first sales, LTV: 663
What the analysis measures
Product reports measure products: how much a SKU sold, at what margin. Entry product analysis measures the customers a SKU brought in.
Every new customer is assigned to the product they bought first, and that cohort is followed forward. Three numbers do most of the work.
Share of new customers. How much of your acquisition actually runs through this product. A great repeat rate on 0.4% of acquisition is a curiosity. The same rate at 30% is a strategy.
Margin-based value over time. Margin, not revenue, over a fixed window such as twelve months. Products with different fulfillment and discount profiles are not comparable on revenue, and entry products are often sold at a deliberate loss.
Repeat behavior in the early window. Whether the cohort places a second order within a period read against your own reorder cycle. It is the earliest reliable signal, long before twelve-month value exists.
Two caveats. Cohorts need enough customers to be readable – forty first-time buyers tell you nothing, and treating that as a finding is how budget gets reallocated on noise. And young products have no twelve-month history, so their long-term value is a projection and should be labeled as one.
Cutting the analysis by acquisition source makes it considerably more useful. The same entry product performs differently depending on whether the customer came from paid social, search or a referral, because the expectation they arrive with is not the same. Product and channel are one decision, not two.
The entry mix is a lever
Most brands manage a product mix. Very few manage an entry mix.
Once you know which products produce returning customers, the promotional question changes: not which product converts best, but which product should be the front door. That decides creative, landing pages, bundles, starter sets, and which SKU your acquisition campaigns point at.
It also changes what an unprofitable first order means. Fast growth in a recurring model requires overpaying for customers up front, and that only works if retention pays it back. An entry product with a higher cost to serve and a materially higher repeat rate is not a margin problem — it is an investment with a known return, provided someone measured the return.
That requires acquisition data and customer data in one place: which product the customer entered through, where they came from, and what they did over the following year. Useless separately, decisive together.
The usual question is which product to promote. The better one is which product should win you customers – and which you should only sell. If you want to see which doors your customers are walking through, book a demo.
