The most common thing I see in Klaviyo audits is brands sending every campaign to their full list. Every sale, every product launch, every newsletter — everyone gets it, every time.
That's not email marketing. That's a broadcast. And it has three predictable consequences: deliverability degrades over time, unsubscribes climb, and revenue per email sent plateaus even as the list grows.
Good segmentation fixes all three. It also unlocks the kind of revenue that most brands think only comes from growing the list — when in reality it comes from sending better messages to the right people.
What segmentation is actually for
Most people think about segmentation as a way to personalise content. That's part of it, but the bigger purpose is relevance at scale. The more accurately you can match a message to where someone is in their relationship with your brand, the higher every metric climbs — open rate, click rate, placed order rate — and the lower your unsubscribes go.
Klaviyo gives you two types of segmentation tools: lists (static, opt-in-based) and segments (dynamic, rule-based). Lists are for consent tracking. Segments are where the revenue work happens. If you're doing campaign segmentation with lists, you're doing it wrong.
The 6 segments every DTC brand needs
1. Champions (VIP buyers)
These are your top 10–15% by lifetime value. They buy frequently, they open emails, they refer friends. The mistake most brands make is lumping them in with general campaigns and giving them the same 10% off email they give everyone else.
Build it in Klaviyo: Placed Order count ≥ 3 AND Total spent ≥ [your top-tier threshold] over all time. Adjust thresholds based on your AOV and order frequency.
What to send: Early access to new products, loyalty rewards, personal notes from the founder. No mass-discount emails — these people already buy at full price.
2. Active engaged (non-VIP)
Opened or clicked email in the last 90 days AND placed at least one order. This is your core health segment — the people who are genuinely paying attention. All your regular campaigns should go to this group by default.
Build it in Klaviyo: Opened Email at least once in last 90 days OR Clicked Email at least once in last 90 days, AND Placed Order at least once over all time.
3. Engaged but never purchased
Opened or clicked in the last 60 days but zero purchases. These people are interested but unconverted. Sending them the same campaign as buyers is a waste — they need conversion content, not retention content.
Build it: Opened or Clicked in last 60 days AND Placed Order zero times over all time.
What to send: Best-seller spotlights, social proof, FAQ content, and a time-limited first-purchase offer. Do not send them VIP content — it creates the wrong expectation.
4. Lapsed buyers (at-risk)
Purchased before, but hasn't opened or clicked in 91–180 days. This is your win-back window. Beyond 180 days, most of these contacts are gone; the win-back window is tight.
Build it: Placed Order at least once over all time AND has not opened email in last 90 days AND last active between 91–180 days ago.
What to send: A short "we miss you" sequence — usually 3 emails. Open with a genuine update (new products, what's changed), follow with social proof, close with a one-time offer if they haven't re-engaged. If they still don't engage, suppress them before they hurt your deliverability.
5. High-value one-time buyers
Purchased once with a cart value significantly above your AOV, but haven't bought again. This segment has high repurchase potential and is regularly overlooked. A one-time buyer who spent £180 on a single order is worth a different conversation than someone who bought a £15 item once.
Build it: Placed Order count = 1 AND Total spent ≥ [1.5× your AOV].
What to send: Product education, complementary product suggestions, and a replenishment reminder if the product category supports it. The goal is a second purchase — unlock that and LTV climbs sharply.
6. Suppression segment (unengaged)
This one doesn't make you money directly — it protects the revenue of every other segment. Sending to people who haven't opened a single email in 180+ days actively damages your sender reputation, which reduces deliverability for your best subscribers.
Build it: Has not opened email in last 180 days AND Has not clicked email in last 180 days AND Subscribed at least 180 days ago.
What to do: Run one final win-back campaign. If they don't engage, suppress them. Remove the fear — a smaller, healthier list outperforms a large, unengaged one every time.
Common segmentation mistakes
Using engagement windows that are too wide. If you define "active" as opened in the last 365 days, you're still sending to people who last opened an email 11 months ago. Tighten it to 90 days and watch your open rates jump.
Not refreshing segments before campaigns. Klaviyo segments are dynamic but they don't recalculate in real time for large lists. Before sending a major campaign, trigger a manual refresh so the segment reflects current behaviour.
Building too many segments and paralysing on strategy. Six well-built segments that you actually use consistently beat twenty segments that sit unused. Start with the six above and add complexity only when you have a specific reason to.
Forgetting to exclude existing customers from acquisition campaigns. If you send a "first-time buyer" discount to someone who's already bought, you either look disorganised or you've just given away margin for nothing. Always exclude Placed Order ≥ 1 from new subscriber campaigns.
Building the segments: practical notes
When building segments in Klaviyo, use all conditions for exclusions and any conditions for engagement signals. Most engagement rules should use "any" — someone who clicked but didn't open is still engaged. Most exclusion rules should use "all" — you want to be precise about who gets removed.
Test each segment before you use it in a campaign. Click into the preview, check 10–15 profiles at random, and verify they match the intent of the segment. Klaviyo's logic can produce unexpected results if you have messy historical data or a custom property that wasn't set consistently.
Segmentation is ongoing work, not a one-time setup. Review your segments quarterly. What counts as "active" changes as your sending cadence changes. What counts as "high value" changes as your AOV shifts. Keep the definitions calibrated to your current business, not the one you had 18 months ago.