Black Friday Cohort Analysis: Are Last Year's Buyers Worth Chasing Again?

You are about to decide how hard to push acquisition this Black Friday: how deep to discount, which products to feature in the Black Friday campaign, how much to spend on ads. Last year's Black Friday sales produced a big revenue number, but you probably do not know how many of those new customers became repeat customers, or whether they only came back for the next discount. A Black Friday cohort analysis of last year's order data answers that before your team commits this year's budget.

Key takeaways

• Treat it as a customer retention check: compare last year's Black Friday first-time buyers with a normal month's first-time buyers at the same age, 30, 60, 90 and 180 days after their first order.

• A similar headline repeat rate can hide a problem. Split second orders into full price and discounted before you call the cohort healthy.

• Three splits do most of the work: first product, discount depth on the first order, and new versus returning buyers during the sale.

• Lead this year's sale with the first products that brought Black Friday buyers back at full price, and stop discounting deeply the ones that did not.

• Budget follow-up only for buyers whose first product usually earns a second order, not for everyone who bought in the sale.

Are Black Friday shoppers worth acquiring?

They are worth acquiring when enough of them place a second order at a normal price within the same window as your usual first-time buyers. If they only come back for the next sale, or not at all, you paid customer acquisition costs for one discounted order, not a customer.

The first order rarely covers what it cost to win the customer, and keeping an existing customer usually costs less than finding a new one. Customer lifetime value is built on the next purchase and the ones after it, and a customer who has ordered twice is more likely to order again than a one-time buyer. The second order is where the math works or breaks.

Many brands judge the sale by revenue on the day. The critical number is what that revenue turned into. The answer is rarely yes or no for the whole cohort. It usually depends on what they bought first and how deep the discount was, which is why the splits below matter more than the headline number.

How do I compare Black Friday customers with customers from other months?

Line up the cohorts by age, not by calendar. Count how many customers from each cohort placed a second order within 30, 60, 90 and 180 days of their own first order, and compare those rates side by side.

A cohort is a group of customers who share a starting point and are followed over time. Grouping customers by the date of their first purchase gives you acquisition cohorts, which is what this comparison uses. The metrics to compare are the retention rate (the share who come back for a second purchase), its mirror image, churn (the share who never return), and the revenue each cohort brings in over time. The guide to cohort analysis covers the idea in more depth, and the cohort analysis glossary entry has the short definition.

Calendar totals mislead. Black Friday buyers run straight into the holiday season, December gifting and January sales, so "orders in January" says little about whether they became customers. Matching ages removes that noise: every cohort gets the same 90 days to come back.

For the comparison cohort, pick a month without a big promotion, such as the October before the sale or a quiet month in spring. If your store is seasonal, pick the month closest to normal for you.

Building the Black Friday cohort analysis from your order export

You need one row per order line or per order, with these columns:

• order id, customer id, order date

• product id or product name, quantity

• order value, at full and at discounted price if your platform exports both

• discount code or discount amount

• sales channel or campaign tag, if you have one

Then work through it in a spreadsheet:

1. For each customer id, find the first order date (the minimum order date).

2. Flag the Black Friday cohort: customers whose first order falls in your sale window, for example Black Friday to Cyber Monday, or the whole cyber week if your Black Friday promotions ran that long.

3. Flag the comparison cohort the same way, using the month you chose.

4. For each customer, find the second order date and compute days from first to second order. This is the repeat purchase timing.

5. Mark whether the second order came within 30, 60, 90 and 180 days.

6. Mark whether the second order used a discount.

7. Count per cohort, and divide by the number of customers in the cohort.

Remove first orders that were fully returned before you count. A Black Friday order that came back is not a customer, and gift returns make this more common after the sale than in a normal month.

Only compare ages that every customer in the cohort has reached. If you run this in September, last year's Black Friday cohort is old enough for the 180-day view, and so is any comparison month before March.

The three splits that separate bargain hunters from loyal customers

A single rate for the whole Black Friday cohort blends very different customers. Split it three ways.

• First product. Group customers by the product in their first order. Some products lead to a second order much more often than others, and the gap is usually wider than any gap between cohorts.

• Discount depth on the first order. Bucket first orders into no discount, light discount and heavy discount. Customers who arrived on the deepest discount are the ones most likely to wait for the next sale.

• New versus returning. Separate genuine first-time buyers from existing customers who bought during the sale. Existing customers inflate the cohort's repeat numbers and belong in a different analysis.

Two optional cuts help if your data allows. By acquisition channel, if orders carry a channel or campaign tag: it shows which channels brought customers who came back, not just volume. And behavioral cohorts, grouped by what customers did during the sale, such as which coupon they used or whether they bought a bundle.

Keep each split to groups with enough customers to trust. A few dozen customers per group is a sensible floor. Merge small products into their product family rather than reading noise.

Do discount-acquired customers come back at full price?

Some do and some only return for the next discount. Check the second order of each Black Friday first-time buyer: did it use a discount code, and how does its average order value compare with second orders from your normal-month cohort?

Build a small grid with cohorts as rows and second-order type as columns:

Read it this way:

• Healthy: the Black Friday cohort's full-price second-order share is close to the comparison month's.

• Discount-trained: the Black Friday cohort returns at a similar overall rate, but mostly with a code. They are waiting for the next deal.

• One-off: few second orders of any kind. The sale bought orders, not customers.

The discount-trained pattern is the one that hides. The headline repeat rate looks fine and the margin does not.

Discounts are not the problem in themselves. When you offer discounts on the right first products, they create customers who come back at full price. On the wrong ones, they teach customers to wait.

Which first products convert Black Friday shoppers into loyal customers?

The ones where a large share of first-time buyers place a second order within your normal second-order window, at a normal price. The products that encourage repeat purchases are usually the ones that get used up (coffee, supplements, pet food, skincare) or products that need a companion item.

For each first product in the Black Friday cohort, record three things: how many buyers came back, how quickly, and whether the second order was discounted. Then compare with the same first product in the normal month. A product that retains in October but not on Black Friday is being bought for the price, not for itself.

The products that drive repeat purchases article explains how to find these gateway products across your whole catalog, not only in the Black Friday cohort. If you would rather not build it by hand, Affinsy reads an uploaded order export and shows retention by first product against your store's baseline, along with cohort retention by month.

Worked example: one cohort, two first products

An illustration with round numbers, not a benchmark.

A coffee store led last year's Black Friday with two products, both at a similar discount:

• Product A: a 1 kg bag of whole beans.

• Product B: a pod sampler box.

The sale brought 2,000 new customers, 1,000 on each product. For comparison, October brought 600 new customers. Here is what each group did within 90 days of their first order:

At the headline level, the Black Friday cohort looks as good as October: 30% of both placed a second order. The full-price column tells a different story, 18% against 25%.

The split explains it. Product A buyers came back more often than October buyers, and mostly at full price. Product B buyers came back half as often as Product A buyers, and four out of five of those who did used another code.

The decision for this year:

• Lead the sale with Product A, even at a lighter discount than last year.

• Stop featuring Product B as a hero offer. Use it as an add-on to Product A orders instead.

• Budget follow-up for Product A buyers who have not reordered by their usual window. Keep Product B buyers on the normal newsletter.

How do I use last year's result to plan this year's Black Friday campaign?

Lead with the first products that brought Black Friday buyers back at full price, keep the deepest discounts away from products that did not, and budget follow-up only for buyers whose first product usually earns a second order.

In practice:

• Product lineup. Feature the retaining products prominently in your Black Friday deals, even if they are not your deepest discounts. Products that produced one-off orders last year can still sell, as add-ons or bundle pieces, not as the headline.

• Discount depth. If heavy-discount buyers came back mostly with codes, try a lighter discount on the retaining products this year and compare.

• Follow-up budget. Plan the post purchase experience now: delivery updates, a usage email, and second-order reminders timed to each product's usual reorder window. That contact after the first order is what turns a sale into repeat business, and customer engagement in those weeks matters more than the size of the first discount. The purchase frequency guide shows how to find that window from your orders.

• Last year's non-returners. Black Friday purchases from customers who never ordered again are mostly a win-back question, not a new-customer one. Contact the ones whose first product usually retains; leave the rest. Check their first order too: a late delivery or a difficult return can end the relationship after one purchase, and no reminder fixes customer satisfaction after the fact.

Without this, the same discount event repeats every year and brings the same one-off orders. With it, you design the promotion around the products that bring customers back, which is how a one-time sales event starts building long term loyalty. The same result also tells you which products to stock deeper and which ones to use in bundles. For a broader read of repeat buying across all cohorts, not only Black Friday, the customer retention analysis article walks through the full process.

Next steps

• Pull last year's order export this week.

• Tag the Black Friday cohort and one normal month, first-time buyers only.

• Build the 30, 60, 90 and 180-day second-order view for both.

• Split by first product and discount depth, and separate full-price from discounted second orders.

• Pick the products to lead with and the buyers to budget a follow-up for.

If you would rather have the cohort comparison, retention by first product and the follow-up lists worked out for you, the 48-hour analysis delivers them from your own orders, in time for the sale when booked by 31 October.

FAQ

How many Black Friday customers do I need for a useful cohort analysis?

A few hundred first-time buyers give a directional read at the cohort level. For the first-product split, merge small products into their product family until each group has at least a few dozen customers, or the differences you see will be noise.

Does this work for products that do not run out?

Yes, with longer windows. Use the 180-day view as the main one and count any second order, not only a reorder of the same item. Accessories, refills and companion products are the usual second orders in these stores.

Do loyalty programs help keep Black Friday shoppers?

Not for the first step. For a first-time buyer, a well-timed reminder with the right product does more for the second order than points. Once you know which repeat buyers come back at full price, a loyalty program can reward that part of your customer base, for example with early access rather than just points. Customer loyalty earned at full price is brand loyalty worth rewarding; loyalty to the discount is not.

How often should I rerun the analysis?

At least twice a year: in September, before you plan the sale, and once the latest Black Friday cohort reaches 90 days. Use the same comparison month and the same ages each time so retention rates stay comparable. The repeat purchase rate guide covers the store-wide number to track between those checkpoints.