A 58% average reorder rate against a 31% category baseline is the clearest product signal we have seen in Palette's early incubation work. That comparison comes from our internal pilot tracking across three Indian consumer categories. It does not mean every product with a high reorder rate will become a large brand. It does mean the product has cleared a test that launch-day sales cannot: someone chose to buy it again.
Consumer teams routinely give more attention to first orders than second orders. They monitor impressions, conversion, new-customer acquisition cost, launch-week revenue and marketplace rank. These metrics are useful for understanding distribution and demand capture. They are much less useful for answering the question that determines whether a consumer business can compound: did the first purchase create a reason to return?
Trial is an event. Reorder is a behaviour.
A first order contains several possible explanations. The customer may have been curious, responding to a discount, influenced by a creator, replacing an urgently needed item, or simply experimenting with a new listing. Some of those explanations are commercially valuable. None proves that the product belongs in the customer's routine.
A reorder removes some of that ambiguity. The customer has experienced the product, incurred the effort of evaluating it, and decided that another purchase is worth making. In categories with a plausible consumption cycle, that decision is a closer proxy for perceived utility than a launch spike.
This is why reorder rate is more useful when read as a cohort measure rather than as a single dashboard percentage. Take customers whose first order occurred in the same week. Track how many return within a reasonable replenishment window, then compare that cohort with later cohorts, acquisition channels, selling platforms, pack sizes and product variants. The signal is not just the percentage. It is the shape of repeat behaviour.
A product with genuine pull usually produces a few recognisable patterns. Reorders appear across multiple acquisition sources, not only among customers acquired through a heavy introductory offer. The interval between first and second orders is consistent with the product's consumption cycle. Repeat rates hold as the customer base broadens. New variants may bring incremental trial without weakening the core product's repeat behaviour.
A product without that pull often looks different. The first-order curve is steep, then falls away. Reorders cluster around promotions or a single traffic source. Customers who do return take much longer than the expected usage period to do so. One large launch cohort creates an attractive blended number, while subsequent cohorts deteriorate.
Why launch spikes create false confidence
Launch performance is especially noisy in Indian e-commerce. A new listing can receive temporary visibility through platform merchandising, paid traffic, creator activity, introductory pricing or a concentrated burst of sampling. These interventions can be sensible ways to test a proposition. They can also make weak product-market fit look like strong demand.
Consider a synthetic example: a haircare brand launches a scalp serum on a major Indian e-commerce platform in early 2025. The brand sells 12,000 units in its first month after combining a launch discount with creator-led traffic. The team sees strong conversion and increases inventory. Over the next two months, the first-order volume remains respectable, but only a small share of the original cohort returns within the product's expected usage window. The second purchase is also heavily discount-led.
There are several possible explanations. The product may not deliver a noticeable benefit. The usage instructions may be unclear. The pack may last longer than assumed. The customer may have bought it as a one-time experiment. Or the acquisition promise may have been more compelling than the product experience. Sales data alone cannot separate these explanations. Reorder data can narrow the field.
The important comparison is not simply 12,000 first orders versus a competitor's 8,000. It is the proportion of comparable first-order cohorts that return, the time they take to return, and whether they return at full or reduced discount. A smaller launch with stronger repeat can be the healthier commercial asset.
What e-commerce data makes measurable
Historically, repeat purchase was easier to observe in offline channels through distributor movement, retailer feedback and household panels. Those sources remain valuable, but they do not always connect an individual first purchase to a later purchase. Indian e-commerce creates a more direct view of the customer journey across order history, dates, quantities, variants, discounts and platform.
That does not make the data automatically clean. Customer identifiers can vary across platforms. A household may use more than one account. Marketplace orders do not always reveal whether a product was consumed, gifted or returned. A customer may reorder from a general marketplace after discovering the brand through another channel. These issues require careful matching and conservative interpretation.
Still, at sufficient volume, the data allows a brand to ask practical questions that were previously difficult to answer at scale:
- What share of first-time buyers returns within the expected replenishment window?
- Does repeat behaviour differ between marketplace, quick-commerce and direct channels?
- Do customers acquired through discounts reorder without the same incentive?
- Which pack size or product variant produces the strongest second order?
- Are repeat customers buying more units, the same quantity, or switching to another product?
- Does reorder performance hold across cities, price bands and customer cohorts?
At Palette, we analyse purchase and reorder data from Indian e-commerce to identify products with genuine repeat demand. The point is not to treat a dashboard result as a verdict. It is to connect demand signals with decisions about formulation, pack architecture, pricing, inventory and the kind of brand that should be built around the product.
The cost of optimising only for trial
Trial is not free, even when the first order is profitable on paper. A brand may spend on sampling, creator fees, marketplace visibility, discounts, returns, fulfilment and customer support before it learns whether the product can earn a second order. If the organisation optimises around first-order conversion alone, it can allocate more capital to the least durable part of the funnel.
The operational consequences arrive later. Inventory is committed against a demand curve that may not repeat. Forecasts assume a customer lifetime that has not been observed. Marketing teams are asked to replace churned buyers with new buyers at progressively higher acquisition costs. Product teams respond by adding variants, even though the core product has not demonstrated routine use.
We are not saying trial acquisition is unimportant: we are saying it tells you almost nothing about whether the product will survive its second month. Trial is how a product enters a customer's consideration set. Reorder is evidence that it earned a place in the customer's behaviour.
This distinction also changes how teams set targets. Instead of asking only how many customers a campaign can acquire, the team can ask what quality of cohort it is likely to create. A lower-volume channel with better repeat may deserve more attention than a high-volume campaign that produces one-time buyers. A pack that converts less effectively but leads to more predictable replenishment may be better for working capital and retention.
How to read a genuine pull signal
There is no universal reorder rate that proves product-market fit. Consumption cycles vary widely. A snack, face wash, supplement and household cleaner should not be judged on the same calendar. The right window is anchored to realistic usage, not selected to produce a flattering chart.
We look for several signals together. First, repeat should be visible in successive cohorts, not concentrated in the earliest and most heavily subsidised customers. Second, the time to reorder should make behavioural sense. Third, the pattern should survive reasonable changes in discounting and traffic source. Fourth, repeat customers should show some continuity in product choice, even if they expand into adjacent products later.
It is also important to separate customer retention from forced replenishment. A subscription can create a second shipment without creating genuine preference if cancellation is difficult or the customer is not actively choosing the product. Conversely, a long replenishment interval may understate satisfaction in categories where usage is slow. Returns, cancellations, product reviews and customer service contacts should sit alongside reorder data.
There is a further complication for Indian brands: the same product can behave differently by geography, affordability and channel. A premium pack may repeat well in one customer segment and fail in another. A product may be genuinely useful but unavailable when the customer is ready to replenish. Stockouts can suppress reorder rates, while a marketplace recommendation can artificially concentrate them. A weak rate is therefore a prompt for investigation, not an automatic reason to abandon the product.
The strongest teams treat reorder analysis as an operating loop. They identify the cohort, inspect the product and channel conditions, form a hypothesis, make one change, and observe the next cohort. Over time, this produces a more reliable view than any launch report. It also helps distinguish a product that needs better distribution from one that needs a better proposition.
The metric that earns the right to scale
For an early consumer brand, the most valuable outcome of reorder analysis may be restraint. It can show that a product deserves more inventory, broader distribution and a larger acquisition budget. It can also show that a promising launch is masking weak satisfaction, an unsuitable pack, poor availability or a proposition that works only at a discount.
That is why reorder rate should be reviewed before scale decisions, not after them. A launch tells you that the market noticed the product. A reorder tells you that at least part of the market found it worth keeping in its routine. The distinction is where product development and brand building become financially meaningful.
If you are building in a consumer category and want to see what your reorder data looks like before committing to scale, we are happy to take a look: the useful decision is not whether you can buy another launch, but whether customers are already choosing to come back.
