Platform
Reading what shoppers actually do after the first purchase
Palette processes anonymised purchase and reorder streams across Indian e-commerce to find categories where consumer pull is already real, before any brand commitment is made.
Four signals the platform reads in every category
Palette does not measure awareness or first-purchase intent. It reads the signals that only appear when a shopper returns.
Reorder Rate
The share of first buyers who place a second order in the same category within a defined window. The benchmark is calibrated to Indian e-commerce baselines, so the comparison reflects actual shopper behaviour rather than general benchmarks.
Repeat-Buy Interval
How long shoppers wait between a first and second purchase. A consistent interval is more useful than a high reorder rate alone, because it tells us whether a consumption pattern is forming around the category or whether returns are scattered and unpredictable.
Basket Co-occurrence
Which other categories and product types appear alongside the target in the same shopper's purchase history. Co-occurrence patterns reveal the broader routine the product sits inside and point toward the right positioning before brand architecture decisions are made.
Shopper Cohorts
Repeat buyers are segmented by geography, purchase frequency, and price tier to identify whether demand is concentrated in one city or one promotional window, or whether it is distributed across independent shopper groups who return without a discount trigger.
The Shift
From "Can we sell this?" to "Will this buyer come back?"
Most Indian consumer brand decisions start with first-purchase demand: social response, trial basket size, launch wave volume. The purchase-sequence read changes the starting question. A category with a strong reorder rate and a consistent repeat-buy interval already has a customer. The brand's job is to show up reliably for that customer, not to manufacture a demand that does not yet exist in the purchase data.
What the Read Separates
Genuine consumer pull from trial spend and discounting
Categories inflated by trial spend or heavy discounting show a pattern in the data: reorder rate drops when promotional support falls, repeat-buy intervals lengthen when the price is not reduced, and second-order basket sizes shrink. None of these signals disqualify a category on their own, but they indicate that what looks like demand is largely a response to pricing activity rather than a genuine consumer habit forming around the product.
The categories Palette acts on are the ones where shoppers return at a similar rate and a similar interval regardless of whether a promotion is running. That is the signal worth building a brand around.
Curious how the platform informs a category decision?
Read about how Palette uses the demand read to select categories and decide whether to build from scratch or back an existing founder.
Read Our Approach