Shopify segmentation made simple

Shopify ships with a real segmentation tool, and most stores never open it. It lives under Customers → Segments in the admin, and it can group your customers by what they have bought, what they have spent, where they live, how they responded to email, and what Shopify's own models predict they will spend next. For a store doing everything by gut feel, it is the fastest available upgrade.

This guide covers what the built-in tools genuinely do well, the five segments worth building first, and the limits you will eventually meet. All of it works on every Shopify plan.

How Shopify segments work

A Shopify segment is a saved filter written in the segment editor: pick attributes, set conditions, and the segment updates itself as customers move in and out. That last part matters more than it sounds. A segment of "customers with no order in 90 days" maintained by hand is out of date the day after you build it; the same segment as a saved filter is correct every morning. In the language of this pathway, Shopify segments are dynamic segmentsDynamic segmentA segment defined by a rule that is re-evaluated continuously, so membership updates as customer behaviour changes. The opposite of a static list. "No purchase in 180 days, valid consent" stays correct every day it runs, because whoever matches today is in today.View in glossary, not static lists.

The attributes worth knowing about: orders placed and amount spent (the workhorses), date of last order, products purchased (by product, variant or tag), email subscription status, customer tags (anything your apps or your team have stamped on the record), location down to city level, and on most plans two predictive fields, predicted spend tier and Shopify's estimate of a customer's likelihood to purchase again.

Five starter segments, and the numbers to watch

1. First-time buyers, last 30 days. orders placed = 1 AND first order in the last 30 days. The number to watch is how many make a second purchase within 60 days. Second purchase is the single biggest fork in a customer's life: repeat buyers behave differently from one-timers in almost every category, so this segment exists to receive your welcome flow and your best reason to come back.

2. Lapsed customers. orders placed >= 1 AND last order more than 90 days ago (set the window to roughly twice your typical gap between orders; 90 days suits most, coffee is shorter, furniture much longer). Watch the reactivation rate of anything you send. A worked check: if this segment holds 1,200 customers with a £45 average order and a win-back email brings 4% of them back, that is 48 orders and about £2,160 from one send to people you had already paid to acquire.

3. High spenders. amount spent > X, where X is your top decile. To find X, export customers, sort by total spend, and read the value one-tenth of the way down. Watch this group's share of revenue; for most stores a tenth of customers produce a third or more of takings. These are the people who should never see your discount codes and should always see your new range first.

4. Discount-only buyers. Two or more orders, every one on a discount code (build it from your code naming convention, or tag these customers via your email platform). Watch what happens when you stop discounting to them. Some quietly pay full price, and the rest were never profitable customers so much as subsidised ones. This segment is where margin leaks hide.

5. Subscribed non-buyers. email subscription status = subscribed AND orders placed = 0. Watch conversion to first order. These people asked to hear from you and have never bought; they are the cheapest first orders you will ever win, because the acquisition cost is already spent.

Where the ceiling is

Shopify's segmentation sees what Shopify sees, and that boundary is worth understanding before it costs you money.

It does not see margin. Amount spent is revenue, so a £200 customer buying clearance stock outranks a £150 customer buying full price, and every "best customer" segment built on spend quietly repeats that mistake. It does not see your other channels: what people did with your emails lives in your email platform, ad audiences live with the ad platforms, and your marketplace or wholesale customers may not be in Shopify at all. And it cannot enrich: no household context, no area classification, nothing you did not collect at the checkout.

None of that argues against starting. It argues for knowing which decisions the data can support. Ranking customers by value? Shopify's numbers alone will mislead you the moment margins vary. Building a win-back flow? The built-in tools are ample. The general rule: the more a decision depends on joining data across systems, the sooner you meet the ceiling, and joining that data properly is what the rest of this pathway is about.