Shopify brands using Klaviyo often have exactly one segment: everyone. Every campaign goes to the full list, and every flow treats a first-time browser the same as a five-time buyer.
The results sit near the industry average. That's what happens when you send an average message to an average audience.
Segmentation is the fix. Well-built segments produce roughly 3–5x the revenue of unsegmented sends to the exact same list. Flow performance alone lifts 40–100% once segmentation replaces a one-size-fits-all approach.
This guide walks through exactly how to build that structure in Klaviyo, step by step. It covers the segments worth building first, how RFM fits in, and the rule for deciding what to keep.
Segments vs. lists: get this right first
Before building anything, understand the distinction Klaviyo draws. Lists are static — they only grow as people actively subscribe, through a signup form or an import. Klaviyo's own documentation is direct about this: a list holds whoever joined it, full stop.
Segments work differently. They're defined by conditions, not a fixed roster — the same core idea behind what email segmentation means. Membership updates automatically, growing as people meet the criteria and shrinking as they stop meeting it.
The practical rule to remember: use lists to capture consent and grow your core audience over time. Use segments for everything else — campaign targeting, flow triggers, suppression, and reporting. Sending campaigns straight to a list instead of a segment built from it usually leaves revenue on the table.
What segmentation actually buys you
The dollar case is straightforward, but the mechanism is worth understanding. Klaviyo's 2026 benchmark data shows automated flows generating nearly 41% of total email revenue from just 5.3% of sends. That's an 18x gap in revenue per recipient over broad campaigns.
That gap isn't about flows being better written — it's about relevance and timing. A welcome email means nothing to someone who's bought six times already. A win-back offer means nothing to someone who ordered last week.
Segmentation is what tells Klaviyo which message belongs in front of which person, at which moment.
Quick reference: the segment types worth building
- Engagement-based
- What it captures: 30/60/90-day openers and clickers
- Primary use: Deliverability, sunset flows
- RFM tiers
- What it captures: Recency, frequency, monetary value
- Primary use: VIP treatment, at-risk recovery
- Behavioral
- What it captures: Browse, cart, category affinity
- Primary use: Relevant campaign targeting
- Predictive (CLV)
- What it captures: Klaviyo's forecasted spend and churn risk
- Primary use: High-value protection, discount targeting
- Product-launch eligible
- What it captures: Category history + recent engagement
- Primary use: Early access, launch revenue
Step 1: Start with engagement-based segments
Engagement segments are the foundation, and they matter beyond targeting. They matter for deliverability. Klaviyo Signals and inbox providers both weight recent engagement heavily when deciding where your email lands.
Build at least these three to start:
- 30-day engaged — opened or clicked in the last 30 days. Your most responsive audience; safe to email more frequently.
- 90-day engaged — opened or clicked in the last 90 days but not the last 30. Still active, but cooling.
- Sunset candidates — no opens or clicks in 90–120+ days. Route these into a win-back sequence, then suppress non-responders.
A sunset flow targeting 90–120-day-inactive subscribers is one of the highest-leverage segments most accounts never build. It protects the sender reputation every other segment depends on.
Step 2: Layer in RFM tiers
RFM — recency, frequency, monetary value — is the backbone most serious Klaviyo accounts build on. It scores every customer on how recently they bought, how often, and how much. Those scores group customers into tiers that call for different treatment.
The five tiers worth building, adapted from common RFM frameworks for Klaviyo:
Champions / VIPs. Recent, frequent, high-spend customers. They've earned early access, loyalty perks, and a different tone — these aren't people who need convincing.
Loyal customers. Strong repeat buyers who aren't quite top-tier spenders yet. Good candidates for subscription upsells and referral asks.
New customers. First purchase within the last 30 days, no repeat yet. The goal here is converting purchase one into purchase two.
At-risk. Previously frequent, high-value customers who've gone quiet — typically 60–180 days since their last order. They already know and like the brand; they need a reason to come back, not a generic discount.
Dormant. No purchase in 180+ days. Light, infrequent win-back attempts make sense here. Heavy investment usually doesn't.
A quick note on how these tiers behave in practice. Champions and Loyal customers together are often under 15% of a list.
That small group typically accounts for a disproportionate share of revenue. Treating it like a first-time visitor is one of the most common ways brands leave money on the table.
- Champions / VIPs
- Typical treatment: Early access, loyalty perks, personal tone
- What to avoid sending: Generic discount codes
- Loyal customers
- Typical treatment: Subscription upsells, referral asks
- What to avoid sending: First-purchase-style offers
- New customers
- Typical treatment: Post-purchase education, usage tips
- What to avoid sending: Aggressive discounting too soon
- At-risk
- Typical treatment: Personalized win-back, not a blanket coupon
- What to avoid sending: Being ignored until they're dormant
- Dormant
- Typical treatment: Light-touch, low-cost win-back attempts
- What to avoid sending: Heavy investment or frequent sends
Step 3: Add behavioral and purchase-based segments
RFM tells you how valuable a customer is. Behavioral segmentation tells you what they actually want. That's what makes a promotional send feel relevant instead of generic.
Category affinity groups subscribers by what they've purchased or browsed. Someone who's only bought skincare shouldn't get your haircare promo. A workable condition: purchased, viewed 2+ product pages, or added to cart from a given collection within 90 days.
Cart and browse abandonment segments feed your highest-converting flows, not just campaigns. Abandoned cart flows average $3.65 in revenue per recipient, with top performers well above that. Our guide to Shopify abandoned cart email covers the full setup.
Post-purchase window groups anyone who bought in the last 30 days. This audience shows 2–4x the open rate of the broader list. It responds well to usage tips and complementary recommendations, not another discount.
Engagement decay flags subscribers whose recent open rate has dropped against their own 90-day baseline. It's a pre-churn signal. Catching it early with a re-engagement sequence can save 15–25% of would-be churners before they go fully dormant.
Product-launch eligible is worth building for any brand releasing new SKUs regularly. Combine category history in the last six months with two or more opens in the past 30 days, then give this group early access. Early sends to a tight segment like this often drive a large share of total launch revenue from a small fraction of the list.
VIP by spend threshold is a simpler cut than full RFM. It suits brands earlier in their Klaviyo maturity — customers above a set lifetime spend, treated to perks regardless of when they last bought.
Seasonal and geographic segments round out the list for brands with weather-dependent products or a multi-country customer base. A swimwear brand doesn't need to promote to Southern Hemisphere customers during a Northern winter. A brand shipping internationally shouldn't send a US-only holiday promo to every subscriber either.
Build these from address data and past purchase timing. Revisit them each season rather than setting them once and forgetting them.
Step 4: Use Klaviyo's predictive analytics
Once an account has enough order history, Klaviyo's predictive layer adds a dimension RFM can't: forecasting.
Predictive customer lifetime value combines past spend with a machine-learning forecast of future spend. It's available as historic, predicted, or total CLV — the same metric behind how to increase customer lifetime value.
Klaviyo's documentation gives a practical example: segment customers predicted to spend under a threshold, then target them with a discount.
This isn't available on day one. Klaviyo needs roughly 500 customers with completed purchases and 180+ days of order history before predictive analytics activates reliably. Treat it as a layer on top of RFM and behavioral segments — not a replacement for the foundational work.
Churn-risk scoring works similarly, surfacing a probability that a given customer stops buying. Layering churn risk onto your existing at-risk RFM segment sharpens who gets your limited win-back budget first.
Step 5: Apply the one-segment rule
Here's where most Klaviyo accounts go wrong. It's not too few segments — it's too many that never get used.
The rule that keeps a segment library useful: every segment needs an action attached. If you can't name a specific campaign you'd send to a segment this month, it shouldn't exist. Small, unused segments don't just clutter your account — they quietly drag down sender reputation without adding meaningful revenue.
Before building any new segment, ask what campaign or flow it triggers. If the honest answer is "not sure yet," hold off until you have one.
How to actually build a segment in Klaviyo
The mechanics are simpler than the strategy. In Klaviyo, go to Lists & Segments → Create List/Segment → Segment. Add conditions from categories like "What someone has done," "Properties about someone," and "Predictive analytics about someone."
Stack multiple conditions with AND/OR logic to combine criteria in one segment. For example: "placed an order" AND "in the last 90 days" AND "NOT purchased from Collection X." Klaviyo shows the live segment size as you build it — a fast way to spot a bad fit.
Name segments consistently, something like RFM: Champions or Behavior: Skincare 90d. Six months from now, anyone on the team should be able to tell what a segment does without opening it.
A worked example. To build a "Skincare category, engaged" segment: add "Placed Order" → "Product Category" → "Skincare" → "in the last 180 days." AND that with "Opened Email at least 1 times" → "in the last 30 days." That combination targets exactly the subscribers who buy the category and still actively open email — the group most likely to respond.
A second example, for the sunset segment. Combine "Opened Email zero times" → "in the last 120 days" AND "Clicked Email zero times" over the same window. That combination catches subscribers who are genuinely unengaged, not just quiet during a slow month. It keeps the sunset flow from suppressing people still worth emailing.
Common mistakes to avoid
Building segments nobody sends to. An unused segment isn't neutral. It's clutter that makes the account harder to maintain and adds no revenue.
Segmenting only by RFM, never by behavior. RFM tells you who's valuable. Behavior tells you what to say to them. Both are needed for a promotional send to feel relevant.
Sending to "everyone who's ever purchased" as one group. A customer from three years ago needs a different message than one from last week, even though both technically qualify.
Skipping the sunset segment. Suppressing chronically unengaged subscribers protects deliverability for every other segment in the account. Skip it, and every send suffers a little.
Trying predictive analytics before the account qualifies. Klaviyo's CLV and churn scores need real order history to be reliable. Build the RFM and behavioral foundation first.
Letting segments go stale. A segment built six months ago on today's conditions can quietly drift out of date as the business changes. Revisit segment definitions whenever a product line, pricing tier, or customer base shifts meaningfully.
How Needle uses Klaviyo segmentation for DTC brands
Disclosure: Needle is our product, so we're not a neutral voice here — but segmentation is genuinely core to how we run Klaviyo for the brands we work with.
Needle connects directly to a brand's Shopify and Klaviyo accounts. It builds the RFM, behavioral, and engagement segments during onboarding, then keeps them current as new data comes in. Every campaign a brand approves is already matched to the right segment; founders aren't building condition logic themselves.
The results show up in the same revenue-per-send numbers segmentation is supposed to produce. TWOOAK's cost per order dropped from $41 to $19 under Needle, hitting a peak efficiency ratio of 41x. RTPTennis moved from roughly $3K to $4.8K in revenue per email send in its first month.
Email management starts at $499/month for five sends, set up and sent inside the brand's own Klaviyo account. Segmentation is included, not billed as an add-on.
Frequently asked questions
How many Klaviyo segments should a Shopify brand have?
Fewer than most accounts end up with. Start with 6–8 core segments — a few engagement tiers, the five RFM groups, and one or two behavioral segments. Add more only once each has a specific campaign or flow attached, not just because it seems useful in theory.
What's the difference between a Klaviyo list and a segment?
Lists are static and only grow through active subscription. Segments are dynamic, defined by conditions, and update automatically as customers meet or stop meeting those conditions. Use lists for consent and signup; use segments for everything else.
Do I need Klaviyo's predictive CLV to segment effectively?
No. RFM and behavioral segmentation deliver most of the revenue lift on their own. Predictive analytics only activates once an account has real order history — typically 500+ customers and 180+ days of data.
What's the highest-priority segment to build first?
A 30-day engagement segment and a basic RFM split — new, repeat, at-risk, dormant — cover most use cases immediately. Add behavioral segments like category affinity once those are in place.
How often should segments be reviewed?
Quarterly is a reasonable cadence for most Shopify brands. Check that every segment still has an active campaign or flow attached, and retire any that don't. Faster-growing brands may want to review monthly.
Can segmentation hurt deliverability?
Segmentation itself protects deliverability by keeping engaged and disengaged subscribers apart. What hurts deliverability is the opposite — sending broad, unsegmented blasts to a list full of unengaged addresses.
Should Shopify brands segment by product category even with a small catalog?
Yes, if there's any meaningful variation in what customers buy. Even a handful of categories stops irrelevant promotions from reaching uninterested subscribers, and setup cost is minimal.
Does segmentation work the same way for SMS as it does for email?
The same segments apply, but treat SMS more conservatively. It's a higher-permission, higher-attention channel. Reserve it for your most engaged and highest-value tiers rather than broadcasting as widely as email.
Conclusion: segmentation is the highest-leverage Klaviyo work
Most of the revenue gap between an average Klaviyo account and a genuinely top-decile one isn't copywriting talent. It's structure — engagement tiers, RFM groups, and behavioral and predictive layers, each tied to a specific campaign or flow.
Start with the foundation: a 30-day engagement segment and a basic RFM split. Add behavioral segments next, and only reach for predictive analytics once the account has the order history to support it.
None of this needs to happen at once. Build one new segment this week, and attach it to a real campaign. Compare the revenue-per-recipient difference against your last unsegmented send to the same list. If you'd rather have a team build and maintain this structure for you, see how Needle works.