Klaviyo Segments for Shopify: 7 Worth Building First

Quick answer: Build these seven Klaviyo segments first, in this order: engaged subscribers, recent buyers, repeat buyers, lapsed customers, high-intent browsers who have not bought, location (Australia, the US and everyone else), and SMS-consented profiles. The first two protect your sender reputation and your margin, the next three earn the revenue, and the last two stop the wrong message reaching the wrong person. Each takes a handful of conditions in Klaviyo’s segment builder, and every time window below is a starting point to tune to how often your customers actually reorder, not a standard.

Most “segment your list” advice hands you twenty ideas and no order to build them in. For a Shopify store doing $150K to $1M, you need the few segments that decide whether your email programme costs you money or makes it, built in a sensible order, with rules you can still read in six months.

This guide ranks seven segments by what each protects or earns and gives the Klaviyo rule for each. I took the Klaviyo and Shopify rules and condition names from their own help pages, read as at 31 October 2026, so check your screen if the labels have changed. Where I give a window such as 30, 90 or 120 days, that is my starting point. Neither platform sets it.

One scope note. Prism3 does not run email, ads or SMS for clients. We work on conversion after the click, organic and AI-search visibility, and the measurement that makes your numbers trustworthy. Email still matters here because it is part of the system that sets your cost per customer and your customer lifetime value, which is the argument in how email lifts customer lifetime value. These segments are the targeting layer under that argument: they decide who gets which message.

The seven, ranked by what they protect or earn

Protective segments go first because once the engaged and recent-buyer segments exist, any campaign can exclude the people it should not reach. Location and SMS come last because they refine the first five rather than replace them.

Before you build

Segment, list or flow?

Klaviyo’s help centre separates the three. Lists are static and grow only when people subscribe or are added manually. Segments are dynamic: profiles join when they meet the conditions and leave when they stop. Klaviyo describes segments as a way to pick out cross-sections of your audience for one-time campaigns; ongoing messaging belongs in flows. Several of the seven below work best as the include or exclude list on a campaign, or as a filter inside a flow, rather than as something you email on its own.

Klaviyo segments or Shopify segments?

Shopify has its own customer segments, built with ShopifyQL, and they update automatically. Its filters include customer_countries, number_of_orders, last_order_date and rfm_group. Dates use whole days in your store’s time zone, and predicted_spend_tier only works once a store has made more than 100 sales.

You can send a Shopify segment to Klaviyo (Customers, Segments, Use segment, Klaviyo), but Klaviyo’s help page says the result is a list that reflects a point-in-time sync and does not update as membership changes in Shopify. To refresh it you repeat the process and get a new list. So my rule is to build the seven in Klaviyo, where the sending happens and the segments keep updating, and to use Shopify segments to explore your customers and sanity-check the counts.

Three data checks before any rule

1. Consent came across. Klaviyo says the Shopify integration syncs whether a customer opted in to email or SMS marketing. Open a few profiles and confirm the status matches what the customer actually agreed to.
2. Country is filled in. Klaviyo fills country and address fields from Shopify. Look at which spellings appear in your own profiles before building segment 6.
3. Onsite events are firing. Order and shipping events sync from Shopify, but Active on Site, Viewed Product, Viewed Collection and Added to Cart are onsite events that need Klaviyo’s onsite tracking set up. Klaviyo also notes there can be two versions of Viewed Product, one from its app embed and a Shopify-branded one from Shopify pixel tracking, so check the source before you build on one. Segment 5 depends on this.

1. Engaged subscribers: protects your sending reputation

The job: make sure regular campaigns go to people who still want them.

Klaviyo’s help pages do not define “engaged”: its AI segment builder lists “Engaged profiles last 30 days” only as a sample prompt, with no criteria. The definition is yours to write.

The rule (starting point):

• If someone can or cannot receive marketing: can receive, email marketing, with the filter “because person: Subscribed” (the mistakes section explains why)
• AND at least one of:
  • What someone has done: Clicked Email at least once in the last 90 days
  • What someone has done: Placed Order at least once in the last 90 days
  • What someone has done: Active on Site at least once in the last 90 days

Why not opens? Klaviyo’s help page on opens says Apple Mail Privacy Protection prefetches its tracking pixel and that open rates will be inflated, so an “opened in the last 30 days” rule counts some people who never saw the email. Klaviyo suggests a custom report with an MPP property to see how much of your opens are affected. The page names no ready-made segment filter for it, so check what your own builder offers. Clicks are not perfectly clean either: Klaviyo’s email metrics page says automated bot clicks can be included in Clicked Email. That is another reason to mix in orders and site activity, which are harder to fake.

The windows are a judgement call. You will see 30, 90 and 120 days quoted in agency guides as engagement windows. They are rules of thumb, not standards. A workable starting set:

To tune them, ask how often you send and how often customers buy. If you email twice a month, a 30-day window gives someone about two chances to click. If your product is a once-a-year purchase, a 90-day window drops buyers who are fine, just not due. Including Placed Order and Active on Site protects those customers: a person who buys but never clicks is not unengaged.

A worked example (illustrative, not a benchmark). Say 10,000 profiles can receive your email and 3,000 were engaged in the last 90 days. A campaign earns 120 clicks, all from the engaged group. Sent to all 10,000, that is a 1.2% click rate. Sent to the 3,000, it is 4.0%. The email did not improve; the denominator changed. Pick one audience definition and compare campaigns on it. Without this segment you cannot tell a weak email from a diluted list.

What about the people who fall out? Klaviyo’s sunset-flow guide gives a strict recommended definition for the group to suppress: profiles created at least 180 days ago that have received at least five emails in the last 72 weeks, with zero opens, clicks, site visits, product views, checkouts started or orders over all time. It recommends no more than three emails in the re-engagement flow, and warns that repeatedly sending to inactive inboxes can hurt deliverability. Your engaged tiers are a softer, earlier line than that. For why mailbox providers care, see the Gmail and Yahoo sender rules for Shopify stores.

2. Recent buyers: protects margin and patience

The job: be the exclusion list that stops the wrong email reaching someone who just paid.

Picture the customer who paid full price on Tuesday and gets “10% off your first order” on Thursday: an avoidable discount and a small insult. This segment keeps them out of acquisition offers, abandoned-cart nudges and win-back messages.

The rule (starting point):

• What someone has done: Placed Order at least once in the last 30 days

Tuning the window. Thirty days is a placeholder. Tie it to how long the order takes to arrive and how long the product takes to use. Klaviyo’s Shopify data reference lists Fulfilled Order, Confirmed Shipment and Delivered Shipment among the events it syncs, so if parcels take a while to reach customers, consider defining “recent” from delivery rather than order date, after confirming those events fire for your carriers.

Use it mostly as a subtraction on campaigns and flows where an order should end the conversation. For the post-purchase sequence itself, a flow is the better tool: see the post-purchase flow for Shopify and Klaviyo.

3. Repeat buyers: earns the most per send

The job: find and look after the customers who already came back.

Klaviyo’s own VIP example is everyone who has purchased at least X times or spent over a certain amount, and it leaves X to you. That is the right instinct, because the threshold depends on your store.

The rule (starting point):

• What someone has done: Placed Order at least 2 times over all time (repeat buyers)
• A tighter VIP segment: Placed Order at least 3 times, or total spend over a figure you choose

Choosing the line. Use Shopify’s number_of_orders filter to count how many customers sit at 2, 3, 4 and 5 or more orders, and look at the shape. Where does a meaningful group sit? Set the VIP line there. Shopify’s rfm_group filter (values such as CHAMPIONS, AT_RISK and DORMANT) is good for exploring, and orders_placed MATCHES (count >= 3, date >= -6m) shows the form a count rule takes in ShopifyQL. To see how many of your customers repeat at all, read how to calculate and raise your repeat purchase rate.

Predictive options. Klaviyo’s “Predictive analytics about someone” conditions include Historic CLV, Predicted CLV, Total CLV, Predicted number of orders and Average time between purchases. They come with conditions: Klaviyo says CLV predictions need at least 500 customers who have placed an order, at least 180 days of order history with orders in the last 30 days, and some customers with three or more orders. A profile that does not qualify shows a message saying so. Klaviyo also says predictions are not exact for individuals and are most reliable averaged across many customers.

What to do with it. Early access and new products first. Do not discount by default; they already buy.

4. Lapsed customers: earns back what you already paid for

The job: reach people who bought, have been quiet longer than they normally are, and are not caught by another segment.

The rule (starting point):

• What someone has done: Placed Order at least once over all time
• AND Placed Order zero times in the last X days
• AND can receive email marketing (subscribed)
• Optional: AND zero Checkout Started in the last 7 days, so you do not email a lapsed customer who is mid-purchase

The only judgement is X, and it should come from your own reorder gap, not from a number in a blog post.

A worked example (illustrative). Suppose your product is a consumable people use up in about eight weeks, and your order data shows most repeat orders land six to ten weeks after the previous one. My own rule of thumb, not a platform standard, is to call someone lapsed at about one and a half times the typical gap. Here that is twelve weeks, or 84 days: Placed Order at least once over all time, and zero times in the last 84 days.

Now apply a lazy “lapsed means 90 days” rule to a store selling something people buy once a year. Nearly every customer is “lapsed” within a quarter, and your win-back email arrives asking why they have not been back during a perfectly normal year. On a monthly-use product the same rule is far too late. Same rule, wrong window, opposite errors.

To find your gap, use Shopify’s customer cohort analysis report. It groups customers by the date of their first order and shows a chosen metric for each cohort over the weeks, months or quarters after it, so you can see when repeat orders land. Klaviyo’s profile also shows an average time between orders figure where predictive analytics is available, a useful cross-check.

Two tiers beat one. Add a second segment at two to three times the gap for the deeply lapsed, with a lighter touch, and feed the oldest into the sunset thinking from segment 1. The flow that sits on top is covered in the winback flow for lapsed Shopify customers.

How membership changes. For relative-time rules, Klaviyo says someone who orders again is removed close to real time, but profiles who age into or out of the window through time alone are added and removed once every 24 hours. There is no event for “did not buy”, so nothing fires sooner.

5. High-intent browsers who have not bought

The job: reach people who looked closely and did not order.

The rule (starting point):

• What someone has done: Viewed Product at least 2 times in the last 14 days (or Checkout Started at least once in the last 7 days)
• AND Placed Order zero times in the same window
• AND can receive email marketing (subscribed)

Know what the segment can and cannot see. Klaviyo links onsite activity to a profile only once it identifies the browser, and its help page says a visit alone does not make a browser known. It can identify a browser after someone clicks a Klaviyo email or SMS link or submits details through a Klaviyo form. So your “high-intent browsers” are the browsers Klaviyo already knows, a subset of everyone who browsed. Klaviyo also says the data inside an Active on Site event cannot be used for segmentation or flow filtering, so use Viewed Product, Viewed Collection, Added to Cart or Checkout Started when you need product detail.

Segment or flow? The always-on reminders belong in flows; see the abandoned cart flow for Shopify and Klaviyo. A segment does different jobs: a one-off campaign to people who viewed a particular collection, a check on how large the group is, and an exclusion on discount sends so a broadcast does not compete with the flow.

A caution on discounts. If many high-intent browsers never buy, the leak is often before the email, in the product page or the clarity of the offer, and a discount hides that rather than fixing it.

6. Location: Australia, the United States and everyone else

The job: keep the right message, currency, delivery promise and season in front of the right country.

If you sell in both Australia and the US, one send to everyone says something wrong to half of them. A December email about summer shipping cut-offs lands differently in Sydney than in Chicago, and a delivery promise that is true for Melbourne is false for Boston.

The rule (starting point), one segment per market:

• Properties about someone: Country equals Australia
• Properties about someone: Country equals United States
• A third segment for “other”, so nobody falls through

Klaviyo’s location-based segment page says location can come from synced address data such as a billing address, or from IP geolocation estimates, and that for profiles who have not purchased it is based on IP geolocation. So a subscriber who has never bought may be placed by IP, which can be wrong. The page also warns that profiles may store spelling variants, using “UK” and “united kingdom” as its example. Check how Australia and the US appear in your own profiles (AU, Australia, United States, USA, US) and include every variant you find. In Shopify the equivalent is customer_countries CONTAINS plus the ISO country code, and a customer with several addresses can fall into more than one location segment.

Use country as an AND condition inside other segments, so lapsed Australia is a different send from lapsed US, and match the currency, delivery message and offer to the market.

The rules differ, so apply the stricter one to everyone. In Australia, the Spam Act 2003 (section 16) says you must not send a commercial electronic message unless the recipient consented (with some exceptions), and section 18 requires a functional unsubscribe facility. The ACMA explains what counts as consent, how quickly you must action an unsubscribe and what an SMS opt-out should look like at acma.gov.au/avoid-sending-spam; read that page before you set your own rules. In the US, the FTC’s CAN-SPAM guide for business focuses on opt-outs: a valid physical postal address in each message, and opt-out requests honoured within 10 business days. This is general information, not legal advice, and US SMS has separate rules I have not covered. The practical point: do not build a looser consent rule for one country than the other. Gate every marketing send on a clean “subscribed” condition, keep a record of how and when each person consented, and do not treat a purchase alone as permission to market to someone. Check what the ACMA says about inferred consent before you rely on it.

7. SMS-consented profiles

The job: hold the list of people who can be texted, so a text only ever goes to them.

The rule (starting point):

• If someone can or cannot receive marketing: can receive, text message marketing (Klaviyo’s channel-consent article calls the same channel SMS, so match the label in your account)
• AND, for market-specific texts, Country equals Australia or the United States

Klaviyo stores consent separately for each channel, so a profile can be subscribed to SMS and unsubscribed from email. Its channel-consent page says to add the consent condition with AND rather than OR in most cases, and that SMS profiles are subscribed, unsubscribed or never subscribed, with filters for how and when they subscribed or unsubscribed.

Treat this as a consent segment first and a revenue segment second. Under Australia’s Spam Act, marketing texts need the recipient’s consent just as emails do, and the ACMA’s page at acma.gov.au/avoid-sending-spam sets out its examples of consent records and SMS unsubscribe wording. General information, not legal advice. The Shopify detail is in SMS marketing consent in Australia for Shopify.

Use it narrowly. Texts suit short, time-bound messages to people who asked for them, not a second chance to reach email subscribers.

Putting them together: send recipes

A segment sends nothing on its own. The value comes from combining includes and excludes. These are starting recipes, not a template.

No recipe sends to “all subscribers”, and the engaged segment appears in most of them, which is why it comes first.

Six segment-builder mistakes to avoid

Each of these comes from Klaviyo’s or Shopify’s own documentation.

1. Using OR with a negative condition. Klaviyo’s own example: “Country doesn’t equal” United States OR Canada includes nearly everyone. Use AND to exclude both.
2. Forgetting the subscribed filter. By default, “can receive, email marketing” can include never-subscribed profiles that have an email address and no global suppression, and segments do not require opt-in unless you add a subscription condition. Add “because person: Subscribed”.
3. Ignoring that Klaviyo works in UTC. The segment builder uses UTC regardless of your account time zone, and “last 30 days” starts at midnight UTC. For a Sydney store in daylight saving, that boundary falls at 11am. Shopify’s segment dates use your store’s time zone, so the two tools will disagree on edge days.
4. Expecting a segment to fire a flow. “Placed Order zero times” does not trigger a flow on its own, and manually updating a segment does not add its members to segment-triggered flows. Use a metric trigger with profile filters.
5. Segmenting on data Klaviyo cannot read. Properties nested in “Extra” or “Details” arrays cannot be segmented, text values in event filters are case sensitive, and date operators do not work on event properties. A wrong data type (a list where the data is text, say) can also leave a segment with zero profiles.
6. Treating the Shopify sync as live. A list created from a Shopify segment is a one-time copy.

Test every segment before it sends

1. Open five profiles inside the segment and check each fits the rule, then five you expected and cannot find.
2. Check the count against Shopify. Recent buyers in Klaviyo should be close to Shopify’s customers with an order in the same period. Differences usually come from UTC boundaries, cancelled orders or guests.
3. Give a new segment time to populate. Klaviyo notes large segments can take a while, and a manual update can take up to 15 minutes.

If you use Klaviyo’s AI segment builder to draft conditions, its help page says you remain responsible for the final definition, so read every timeframe and AND/OR connector.

Where segments stop

Segments only choose among people you already have. They cannot turn a visitor into a subscriber or make a first order cheaper. If your high-intent browsers segment is tiny, ask whether tracking is firing or too few visitors are being identified. If your lapsed segment is large and your repeat segment small, that is a retention and product problem no segment solves.

Two places where this meets the rest of the system. Cost per customer: the more visitors who buy on the first visit, the fewer you need to win back, and conversion after the click is the part we work on. The whole-system view of paid, organic, conversion and email shows where email sits. And trusting the numbers: an engaged-only audience raises your click rate without improving the email, so compare results on the same audience, and check how revenue is credited across tools (the attribution guide under Keep reading covers it).

Key takeaways

• Build seven segments, in order of what they protect or earn: engaged subscribers, recent buyers, repeat buyers, lapsed customers, high-intent browsers, location, SMS-consented.
• Klaviyo’s help pages do not define “engaged”. Write your own rule from clicks, orders and site activity, because Apple Mail Privacy Protection inflates opens and bot clicks can be counted.
• Treat 30, 90 and 120 days as starting windows and tune them to how often you send and how often customers buy.
• Set the lapsed window from your own reorder gap, for example about one and a half times the typical gap, not a fixed 90 days.
• “Can receive marketing” can include never-subscribed profiles by default, so add the subscribed filter every time.
• Klaviyo’s builder uses UTC and Shopify uses your store time zone, so small count differences are expected.
• Location is for relevance, not for loosening consent. Apply the stricter rule to every market; this is not legal advice.

FAQs

How many Klaviyo segments does a Shopify store need?

Start with the seven here. The help pages I read state no limit on the number of segments, and one segment can hold up to 100 conditions, so the practical limit is your ability to maintain them. If you cannot say in one sentence what a segment is for, delete it.

What counts as an engaged subscriber in Klaviyo?

Klaviyo does not set a definition, so you write the rule. A sensible one counts recent clicks, orders or site activity, because open rates are inflated by Apple Mail Privacy Protection. Start with a 90-day window and tune it to how often you email and how often your customers buy.

How often do Klaviyo segments update?

Klaviyo says segmentation requests are processed as close to real time as possible, so a purchase or click can change membership shortly after. Rules using relative time, such as “in the last 30 days”, also change as time passes, and those time-only changes are processed once every 24 hours. A manual update can take up to 15 minutes (as at 31 October 2026).

Can I send different emails to Australian and US customers?

Yes. Build one segment on Properties about someone, Country equals Australia, and another for the United States, and use them as AND conditions in campaigns. Check how each country appears in your profiles, because Klaviyo warns about spelling variants. Keep one consent standard across both: Australia’s Spam Act (explained by the ACMA) turns on consent for marketing messages, and the FTC’s CAN-SPAM guide sets opt-out and postal address rules in the US. General information, not legal advice.

Do I need Klaviyo’s predictive analytics to segment well?

No. Rules based on order count, time since last order and engagement do most of the work and are easy to check against Shopify. Klaviyo’s CLV predictions need at least 500 customers who have ordered and at least 180 days of order history, among other conditions.

Keep reading

• The post-purchase flow for Shopify and Klaviyo
• Repeat purchase rate on Shopify: how to calculate and raise it
• Email attribution across Klaviyo, GA4 and Shopify

Want to know where first orders are being lost?

Segments decide who gets which message; they do not show where visitors leave before they join a list. Book a free 30-minute diagnostic: we look at where your store loses buyers after the click, using your own numbers, and tell you what to fix first. We do not run ads or email. No pitch. No email gate. If you want to see how we work on conversion, start with our Shopify CRO service.