Ecommerce Customer Segmentation by Store Size

Ecommerce customer segmentation means sorting your customers into groups that each get a different message, offer, or level of service. Most online stores should start with purchase behavior: who has ordered once, who orders often, and who has gone quiet. Store size changes how many groups your team can run well. In our data, it doesn't change when first-time buyers come back, only how many do.

That last point matters for planning. A small store can copy the timing of a large store's follow-up without copying its number of segments.

What is ecommerce customer segmentation?

Customer segmentation is the practice of dividing customers into groups with shared traits or behavior, so each group gets treatment that fits. A segment is the resulting audience, such as first-time buyers from the last 30 days.

Done well, segmentation makes messages more relevant and cuts wasted discounts. Customers who would have bought anyway can stop receiving your deepest offers. Customers who need help with a product can get it before they lose interest.

Start from a decision, not a data field. Name the action first, such as "send a care guide after the first order", then define who should get it. A group that doesn't change what you send, offer, or spend is a report, not a segment.

The four types of customer segmentation

The four classic types are demographic, geographic, psychographic, and behavioral. Each answers a different question about the customer.

For an online store, behavioral data is the most reliable starting point because it comes from orders, not guesses. Three behavioral views do most of the work:

• Lifecycle: new, repeat, and lapsed customers, based on order count and time since the last order.
• RFM: recency, frequency, and monetary value scored together, so you can separate regular buyers from fading ones.
• Value: customers ranked by what they spend, for decisions about service, early access, and how much to invest in keeping them.

Browsing behavior and the channel a customer first came from are behavioral too. They help with abandoned carts and ad audiences, but order history decides who is a customer worth keeping.

Demographic and psychographic traits work best when customers give them to you, through a quiz, survey, or preference center. The guide to customer segmentation models compares rules, RFM, clustering, and prediction in more depth.

Do first-time buyers come back faster at bigger stores?

They come back more often, but on the same schedule. Across 1,269 Shopify stores, we tracked customers whose first order came between October 2024 and September 2025. We grouped stores by annual revenue and measured the share of first-time buyers who placed a second order within 30, 90, and 365 days.

At stores under $1m a year, 6.0% of first-time buyers ordered again within 30 days, at the median. At stores above $10m, it was 8.5%. Within a year, the figures rose to 17.4% and 24.4%.

Here is a follow-up sequence that matches that shape, whatever your size:

1. First orderThank you and product help
2. First monthMain follow-up and a reason to return
3. Following monthsA lighter reminder for slower buyers
4. End of the yearA win-back offer for buyers who never returned

Put the most effort where most second orders happen, then keep a lighter reminder for slower buyers.

Use your own buying cycle to set the exact gaps. A coffee store and a furniture store need different windows, even when both follow this order.

Which segments to start with at each store size

Size changes how many groups your team can build, maintain, and test. The right number is the one you can keep current and act on each week.

Under $1m a year

Start with three or four behavioral groups and automate them. A founder or a single marketer can't hand-build lists every week.

• First-time buyers in their first month: product help, then a reason to return.
• One-time buyers past the first month: a reminder or a relevant product, not an automatic discount.
• Repeat buyers: early access or a thank you, so your best customers don't only see sales.
• Abandoned checkouts: a short reminder flow, since these shoppers already chose a product.

$1m to $10m a year

Add value and product groups once the basics run without attention. RFM scores help you separate regular buyers from fading ones, and product groups let you send advice that fits what someone bought.

This is also where discount sensitivity becomes worth tracking. Keep customers who only buy on promotion out of full price launches, and keep loyal full price buyers out of your deepest offers.

$10m+ a year

Larger teams can run more segments, more channels, and proper tests. The risk shifts from too few groups to overlapping ones. Agree which message wins when a customer sits in several groups, and retire segments nobody uses. The customer segmentation strategy guide covers planning and ownership.

Whatever your size, segments should update themselves. A list saved last month has already put new buyers in the wrong group. See dynamic vs static segments for when a fixed list still makes sense. By the Numbers builds customer segments from order history, spend, and behavior and keeps them current, and can push them to Klaviyo, Mailchimp, Attentive, Meta, Google Ads, TikTok, and Reddit Ads.

Get the data right before you build groups

Segments are only as good as the records behind them. Check these before you trust a group:

• One customer, one record. Duplicate profiles split one person's history, so a loyal customer can look like two one-time buyers.
• Orders with a customer attached. Guest orders without an email or customer account drop out of behavioral groups.
• Clean order data. Test orders, staff orders, and wholesale accounts distort value groups.
• Consent by channel. Only email or text customers who opted in, and treat email and SMS consent separately.

Watch group size too. A segment too small to show a difference can't tell you whether a campaign worked, and one whose membership never changes is out of date.

Check that a segment changes results

A segment earns its place when the action you take with it changes what customers do. Hold back a random part of the group, send the campaign to the rest, and compare orders over the same period. Look at margin after discounts, not only revenue.

Compare against your own history, not against stores of another size. The benchmark above tells you where your second-order rate sits next to similar stores. It can't tell you that a campaign will close the gap. For the wider setup, from first segments to reporting, read the complete guide to customer segmentation in Shopify.

How we measured it

These benchmarks use anonymized, aggregated data from Shopify stores in the By the Numbers network. No individual store is identifiable from any figure. Stores that use an analytics app are not a random sample of all Shopify stores, so read the figures as a reference point, not as an average for the whole market.

Keep reading

If you want the full picture on customer segmentation, start with customer segmentation for Shopify.

Related reading: Customer segmentation models compared, How to build a customer segmentation strategy and Behavioral Segmentation: Store Examples.

By the Numbers builds this into the dashboard. See customer segments.