How D2C brands measure customer retention
Cohort analysis, repeat purchase rate and retention rate are the three measures that tell a D2C brand whether buyers come back. Here is how each works and how to read it.
The short answer
D2C brands measure customer retention with three core tools: cohort analysis (grouping customers by first purchase month and tracking what share buys again), repeat purchase rate (the share of customers with more than one order in a period) and retention rate (active customers at the end of a period divided by those at the start). Read together, they show whether loyalty is improving.
Key takeaways
- D2C brands measure retention mainly through cohort analysis, repeat purchase rate and retention rate, each answering a different question.
- Cohort analysis groups customers by first-purchase month so a brand can compare newer buyers with older ones at the same age.
- Repeat purchase rate should be read against the product's natural repurchase cycle, not a generic benchmark.
- A rising revenue line with a falling retention rate usually means the brand is buying growth it cannot keep.
- Retention only creates value when repeat orders carry a positive contribution margin after product, shipping and payment costs.
Customer retention, in plain words, is the share of buyers a brand keeps. A D2C (direct-to-consumer) brand sells through its own website, app and stores rather than mainly through marketplaces, so it owns the customer record. That record is what makes retention measurable at all. Retention is not the same as satisfaction. A happy customer who never returns is not retained. Retention is a behaviour: did this person buy again, and when?
For Indian D2C brands, the question has become sharper as acquisition costs rise. Comet, the Bengaluru sneaker label, spent ₹21.5 crore on advertising and promotion in FY26 against ₹74 crore of operating revenue, according to company filings reported by media. Neeman's spent ₹45 crore on advertising and marketing against ₹180 crore of revenue in the same year, according to its annual financial statements. When a brand spends that much to get a first order, the second order decides whether the money was well spent. Here is how D2C customer retention metrics actually work.
What are the core D2C customer retention metrics?
Most retention reporting rests on four numbers. Each answers a different question, and none is useful alone.
- Repeat purchase rate — the share of customers who have placed more than one order in a given period. It answers: how many buyers come back at all?
- Retention rate — the number of customers active at the end of a period divided by the number active at the start. It answers: is the base growing or leaking?
- Cohort retention — the same idea applied to a fixed group of customers who first bought in the same month. It answers: is the brand getting better at keeping people?
- Purchase frequency and time between orders — the average number of orders per customer and the typical gap between them. It answers: when should the brand expect a second order?
Two supporting metrics sit alongside these: customer acquisition cost (CAC), the marketing spend needed to win one new buyer, and contribution margin per order, what is left after product, shipping and payment costs. Retention only creates value when the repeat order carries a positive contribution margin.
How does cohort analysis work in practice?
A cohort is a group of customers defined by a shared start date — usually the month of their first purchase. Cohort analysis tracks each group separately over time, so a brand can compare the January cohort with the June cohort instead of staring at one blended average that hides everything.
The method is simple. Take every customer whose first order was in a given month. Then count how many of them ordered again in month one, month two, month three and so on after that first purchase. Plot the share returning at each step. A healthy cohort curve drops sharply after the first month and then flattens — a small core keeps coming back. A curve that keeps sliding to zero means the brand is renting customers, not keeping them.
For Indian brands selling across channels, cohorts need one adjustment. SNITCH said in a company statement that its website and app contribute 35-40% of overall business, marketplaces around 15% and offline retail the remaining 45-50%. A customer who first buys on a marketplace and later buys on the brand's own site is a repeat customer, but the marketplace order may not carry an identifiable customer record. Brands that want clean cohorts usually match orders by phone number or email across channels.
What does repeat purchase rate tell a founder that revenue does not?
Revenue can grow while retention falls. A brand that doubles ad spend will usually add first-time buyers faster than it loses old ones, so the top line rises even as the repeat rate drops. That is why repeat purchase rate is a check on growth quality.
Read it in context. A cookware brand like Ember, which sells pans and Dutch ovens, should expect a long gap between orders — a second Dutch oven is not a frequent purchase. A grooming brand such as H.I.M, which recently launched an eight-product range priced ₹299 to ₹899 across its website, Amazon and Blinkit, lives on consumables that run out. The same repeat rate means very different things for the two. Compare a brand's repeat rate with its own past, and with products of similar repurchase cycle, never with a random benchmark.
Segment before drawing conclusions. Repeat rate by first product bought, by city, by acquisition channel and by whether the first order was discounted. A pattern usually appears: one entry product retains far better than the rest, or discount-acquired customers rarely return at full price.
How should a brand interpret retention rate and cohort curves?
Retention rate is a ratio, so the definition matters. Decide whether a customer counts as retained if they ordered in the last few months, half a year or a year, and keep that definition fixed. Changing it mid-year makes every comparison meaningless.
When reading the numbers, watch three signals:
- The shape of the curve, not the first point. Month-one retention is always the highest. What matters is whether the curve flattens at a level that covers acquisition cost.
- Newer cohorts versus older ones. If the June cohort retains better than the January cohort at the same age, something in product, delivery or communication improved. If it is worse, growth is coming from weaker customers.
- Retention by channel. Miraggio said in a company announcement that it gets about 70% of its online business from marketplaces and 30% from its own website, and is targeting 60% of revenue from own channels. Own-channel customers usually carry richer data, so retention measured there is more reliable — but it is not the whole business.
A falling retention rate with rising revenue is the classic warning sign. It usually means the brand is buying growth it cannot keep.
What mistakes do brands make with retention metrics?
- Counting revenue, not customers. A few large buyers can lift revenue while most customers never return.
- Ignoring returns. In fashion and footwear, a returned order is not a purchase. Net orders of returns before calculating repeat rate.
- Blending channels. Mixing marketplace and own-site data without matching identities understates or overstates retention depending on the mix.
- Judging too early. A cookware or furniture brand cannot assess a cohort at 30 days. Match the observation window to the natural repurchase cycle.
- Chasing repeat orders with discounts. Discount-led repeat purchases can show healthy retention while destroying contribution margin.
What it means for Indian retailers
India's D2C brands are moving from pure online selling to hybrid models that combine websites, stores and delivery partners. SNITCH uses existing stores for 60-minute delivery in Bengaluru and Gurugram, estimating, according to the company, that express delivery adds ₹60-120 per order. Comet, per media reports, opened a store at Pacific Mall in Jaipur on 17 November 2026 and targets 20 stores by the end of FY27. Stores change retention maths: a shopper who has tried a shoe in person behaves differently from one who bought online. Brands should tag store visits and store purchases against the customer record so cohorts capture the full relationship.
Retention also decides how much a brand can afford to spend. According to company filings, Comet spent about ₹1.15 for every rupee of operating revenue in FY26; Neeman's spent ₹1.11, according to its annual financial statements. Both are buying scale. The brands that narrow those ratios fastest will be the ones whose repeat customers carry the fixed costs. For founders, the practical rule is simple: report cohort retention every month, segment by first product and channel, and judge every acquisition push by what the cohort does in month three, not week one.
In this story
- CometCompany · 5 stories
EmberCompany · 6 stories- H.I.MCompany · 4 stories
- MiraggioCompany · 13 stories
Neeman'sCompany · 7 stories- SNITCHCompany · 17 stories
Logos are trademarks of their owners and are shown to identify them.
Frequently asked questions
What is a good repeat purchase rate for a D2C brand?
There is no universal number. It depends on the repurchase cycle: a grooming brand with ₹299-₹899 products should see repeat orders far sooner than a cookware brand selling Dutch ovens. Compare a brand's repeat rate with its own past cohorts and with products of a similar cycle.
How is retention rate different from repeat purchase rate?
Repeat purchase rate is the share of customers with more than one order in a period. Retention rate is active customers at the end of a period divided by those at the start. Repeat rate measures whether buyers return at all; retention rate measures whether the customer base is holding or leaking.
Why does cohort analysis matter more than an average retention number?
A blended average mixes old and new customers, hiding whether the brand is improving. Cohorts track fixed groups over time, so a founder can see if the June cohort retains better than the January cohort at the same age, which points to real product or service gains.
Should marketplace customers be included in retention metrics?
Yes, but carefully. Marketplace orders may not carry a clean customer record, so brands match orders by phone or email across channels. Excluding marketplace buyers understates retention for brands like Miraggio, which said in a company announcement that about 70% of its online business comes from marketplaces.
What is the biggest mistake in measuring D2C retention?
Ignoring returns and contribution margin. A returned order is not a purchase, and a discount-led repeat order can look like loyalty while losing money. Net out returns and check contribution margin per repeat order before calling a cohort healthy.
