Customer retention: how to measure it, and how to bring customers back.

What customer retention means for an online store

A subscription business knows exactly who is still a customer: they are paying. A store does not. A customer who ordered a serum in March and nothing since is not cancelled, just quiet. So for a store, retention means buying again within a window that fits your product, and the first job is choosing that window.

A window that fits is roughly how long a typical customer takes between orders. A coffee or skincare brand might measure month to month. A furniture or eyewear brand might need a year. Pick one and keep it, because a retention rate is only useful compared with itself.

• Retention is the behaviour: customers buying again.
• Churn is the other side: customers who stopped. See customer churn: how to measure it and win customers back.
• Loyalty is the attitude behind it: a customer who would choose you even when a cheaper option is one tap away. Loyalty usually shows up as retention, but a customer can be retained by habit or convenience without being loyal.

How to calculate customer retention rate

The textbook formula is (E - N) ÷ S × 100, where S is the number of customers at the start of the period, E the number at the end and N the new customers won during it. Subtracting the new ones leaves only the customers you kept.

For a store, the cleanest way to apply it is to compare two periods of equal length: of the customers who ordered in the first period, how many ordered again in the second.

1. Pick the periodUse your repeat window: a month, a quarter or a year. Quarters suit most fashion and beauty stores.
2. Count the customers in the first period (S)Unique customers with at least one completed order. Match customers by phone number as well as email, since many shoppers in the region order with a phone number only.
3. Leave out orders that never became salesCancelled orders and cash-on-delivery parcels refused at the door are not purchases. Counting them inflates both periods and hides the real number.
4. Count who came back (E - N)Of the customers in the second period, keep only those who were also in the first. New customers and people returning after a long gap are not retained customers, so they stay out.
5. Divide, and repeat every periodRetained customers ÷ first-period customers × 100. Run the same calculation every period so you compare like with like.

A worked example in EGP

Take an illustrative skincare store in Cairo selling on Shopify, about half of its orders cash on delivery. Between January and March, 2,000 customers received and paid for at least one order. Between April and June, 2,300 customers ordered: 1,760 for the first time and 540 returning. Of those 540, 40 had last ordered before January, so they are reactivated, not retained.

Retention rate for the second quarter: 500 ÷ 2,000 × 100 = 25%. One in four of last quarter’s customers bought again.

The store and its numbers are an example, to show the calculation. They are not a benchmark.

The metrics around retention, and how they connect

Retention rate tells you whether customers come back. The other numbers tell you what that is worth. The figures below continue the same illustrative store over 12 months: 6,000 customers, 8,400 orders and 12.6M EGP in revenue.

They form a chain. Better retention raises purchase frequency. Frequency times order value is what a customer is worth. And what a customer is worth sets how much you can afford to pay for the first order: if a customer brings 840 EGP of gross profit in a year, an acquisition cost above that loses money in year one, however good the ads look.

What a good customer retention rate is

It depends on what you sell and how often people need it, so a single target number does more harm than good. The most useful published figures for stores come from Bluecore’s 2025 Customer Growth Benchmarks, which covers more than 100 retailers across seven verticals using 2024 data, and defines retention as the share of the previous year’s buyers who buy again.

Bluecore’s report does not break out the Middle East, and it measures year on year. Treat it as a rough sense of scale, not a target for a store in Cairo or Riyadh. Figures from Bluecore as tabulated by Shopify, checked October 2026.

Set your own baseline instead

• Calculate your retention rate for each of the last four quarters with the same rules. The trend matters more than the level.
• Group customers by the month of their first order and follow each group. On Shopify, the customer cohort analysis report in the admin does this for you.
• Split it by first product bought. Customers who start with a refillable often come back at a very different rate from those who start with a one-off.
• Split cash on delivery from prepaid. If COD first orders retain worse, the problem may be the first delivery, not the product.

Why retention matters for stores in Egypt, Saudi Arabia and the UAE

• You pay for every first order. It comes through ads, influencers or marketplaces, and each one is paid for. A second order comes through a message to a customer you already have.
• Returning customers are worth more than one order. Bain’s Fred Reichheld found that in financial services a 5% increase in customer retention produced more than a 25% increase in profit. It is an old figure from another industry, but the reasons he gives carry over: returning customers buy more over time, cost less to serve and refer others.
• The first COD order is the riskiest one. A customer who has already received a parcel and paid at the door has shown you that the address, the phone number and the intent are real. See cash on delivery in MENA e-commerce.
• Your customers are already on WhatsApp. The phone number is on every order, so you can reach a past buyer where they already talk to you, instead of hoping they open an email. See WhatsApp broadcasts and a WhatsApp CRM built on orders.

Customer retention strategies, each with the flow that runs it

Retention is mostly won or lost in a handful of moments after the first order. Each strategy below is a flow: it starts from an order event or a customer segment, and sends the message without anyone watching the phone.

1. Make the first order go well

A customer whose first parcel arrived on time, as described, is the one who orders again. Confirm the order on WhatsApp before you pack it, then tell the customer when it ships and when the courier is coming. See the order confirmation flow and the shipping and delivery updates flow. For the COD side, read COD order confirmation on WhatsApp.

On Odoo, Nonito receives order created, cancelled, packed and delivered events, and each can start a flow. Order dispatched and order updated events are stored but do not start a flow yet.

2. Ask how it went, after delivery

A few hours after delivery, ask for a rating in one tap. Happy customers become reviews; unhappy ones reach a person before they leave quietly. Seven brands on Nonito run the post-purchase feedback flow, which has reached 30,375 customers. It is run for the replies and reviews, not for revenue. When an order goes wrong and is refunded, the refund recovery flow asks what happened and offers credit toward the next order.

3. Remind them before they run out

For anything consumable, the best time for the next order is just before the last one runs out. Count from the delivery date, not the order date. In Nonito this is a segment, such as customers who bought a 30-day serum with 25 to 30 days since their last order, and a flow that starts from it.

4. Win back the customers who went quiet

Customers who used to buy and stopped are cheaper to bring back than strangers are to win. Reach them while they are cooling, not once they are lost, with one message and a reason to return: something new, something they liked, a reminder of what they bought. Keep discounts for the ones who do not respond. The customer churn guide covers how to spot them and what to send.

5. Tell them when it is back in stock

A customer who asks to hear about a sold-out size has told you exactly what they want. The back-in-stock flow sends an SMS to those people when the variant returns. On Nonito it has run for one brand so far: 1,212 customers alerted and 168 orders, a 13.0% conversion. One account is enough to report, not enough to call a range.

6. Treat your best customers differently

Your most frequent and highest-spending buyers deserve more than the same discount everyone gets. Early access to a launch, a gift with the next order or a referral link with a reward work better than another code. The referral flow and the cross-sell flow both start from a saved segment. Both are newer flows on Nonito, with no meaningful results to publish yet.

Nonito figures are real results from brands on Nonito, as of 13 September 2026. Yours depend on your product and pricing, so read them as a range, not a forecast.

How to measure retention in Nonito

• Lifecycle stages. Every night, Nonito scores each Shopify customer from their order history into one of eleven stages: New, One-Timer - Recent, One-Timer - Cooling, One-Timer - Churned, Returning, Loyal, Champions, Cooling, At Risk, Previously Loyal and Lost.
• Lateness is judged against each customer’s own rhythm. A repeat customer is measured against their usual gap between orders, and a one-time buyer against your store’s typical gap, rather than one fixed number of days for everyone.
• Value is kept separate. Each customer also gets a High, Mid or Low value tier, so a frequent small spender and an occasional big spender do not end up on the same label.
• Segments. Combine days since last order, days since first order, number of orders, average order value, total spend, city, products and collections bought, lifecycle stage, value tier, discount use and whether the customer can be reached by phone or email. A live segment re-checks every 6 hours.
• Flows from segments. A flow can start from a saved segment, so customers who move into At Risk or reach a replenishment window are picked up without anyone exporting a list.

Lifecycle stages and segment conditions run on Shopify order data. Odoo orders do not feed them today. To include Odoo customers, upload them as a list, or start a flow from one of the Odoo order and customer events.

The flow builder is on every paid plan, from Spark at $49 a month. Growth, at $129 a month, adds your own WhatsApp number, which marketing messages such as win-back need. Plan prices are in US dollars and exclude VAT. See pricing, and how store data comes in from Shopify and Odoo.

For how each flow performs across brands on Nonito, in EGP, see the flow benchmarks report.

Sources

• Bain & Company, Fred Reichheld: Prescription for cutting costs (2001) (checked Oct 2026)
• Shopify blog: Average customer retention rate by industry, citing Bluecore (checked Oct 2026)
• Bluecore: 2025 Customer Growth Benchmarks Report (checked Oct 2026)
• Shopify Help Center: Customers reports, including cohort analysis and RFM (checked Oct 2026)