Klaviyo Sunset Flow: When to Actually Suppress Subscribers

Search “sunset flow Klaviyo” and page one hands you four different answers: sunset after 30 days, or 90, or 180, or only the subscribers who never once engaged. Two of those answers come from Klaviyo itself, and its help doc and its blog do not agree.

None of them shows what the decision is actually worth, or what suppressing a subscriber costs you.

This post gives you one threshold, the reasoning behind it, and the first-party numbers the other guides leave out, from 4,000+ DTC campaigns across the 40+ accounts we run.

What a Sunset Flow Is in Klaviyo

A sunset flow in Klaviyo is an automated sequence sent to subscribers who have stopped engaging, with one last attempt to bring them back before the flow suppresses them. Suppression stops the sends without deleting the profile.

It exists for three reasons, and tidiness is not one of them. Dead weight on your list drags sender reputation, which lowers inbox placement for the subscribers who do want to hear from you.

It also protects deliverability during high-volume periods, when mailing a large unengaged segment is most likely to trip spam filters. And because Klaviyo bills on active profiles, every unengaged subscriber you keep mailing is a line item on your invoice.

Most readers arrive here for the threshold, not the vocabulary, so that is where the rest of this post goes. The short version: engagement is the trigger, suppression is the outcome, and the entire debate is about where you draw the line between them.

Klaviyo Sunset Flow vs Winback Flow: Why the Order Matters

These two get conflated constantly, and the order is the part people get wrong.

A winback flow tries to make a lapsed subscriber buy again. A sunset flow decides whether you keep mailing them at all.

Winback runs first. Sunset is what happens when winback fails.

So how much rescue effort is a lapsed subscriber worth? In our flow benchmark data across 14 accounts with a full flow taxonomy over a trailing 12 months, the winback flow earns $0.32 revenue per send, seventh of ten flows.

For comparison, from those same 14 accounts, Cart Abandonment earns $1.50 per send, Welcome $1.01, and Loyalty or Referral $1.00. A winback send is worth roughly a fifth of a cart-abandonment send.

Winback is also not some neglected opportunity waiting to be unlocked. 79% of those 14 accounts already run one, which makes this an over-rated flow, not an under-built one.

The takeaway for your build is blunt. Do not pour weeks into a nine-email rescue mission for people worth 32 cents a send, because two or three honest emails followed by the suppression decision does the same job.

What Klaviyo Officially Recommends, and Where It Breaks

Klaviyo ships a prebuilt Sunset (Email) segment, and it is a reasonable starting point. Here is the definition exactly as Klaviyo writes it:

Created is at least 180 days ago · Received Email at least 5 in the last 72 weeks · Opened Email 0 times over all time · Clicked Email 0 times over all time · Active on Site 0 times over all time · Viewed Product 0 times over all time · Checkout Started 0 times over all time · Placed Order 0 times over all time

It does the job, with two gaps worth adjusting for before you rely on it. Notice that every engagement condition reads “0 times over all time.”

The first gap is scope. Because the criteria are all-time, the segment catches only people who never did anything at all, so it can miss the subscriber who bought twice in 2024 and has gone quiet since.

The second is that it leans on opens. Apple Mail Privacy Protection pre-fires the tracking pixel whether or not a human opens the message, so opens now run high across the whole industry, not just in your account.

The nuance we would add, and it is our published position, is that opens still have a job. They work as a liveness signal, meaning did a real person ever touch this profile, even though they no longer work as a performance signal, meaning was this campaign better than that one.

So keep opens in the mix. Just stop treating an open as proof a campaign worked, and add behavioral signals alongside it, which is the same logic behind how we define engagement segments.

How to Build the Sunset Segment in Klaviyo

This is the part you can act on today. Build the segment on a window of real inactivity rather than all-time zero, and exclude the people you would regret suppressing.

Here is the segment logic we use, as a build:

1. Received at least one email in the window. If you never mailed them, you cannot call them unengaged.
2. Zero opens and zero clicks in the window. Both conditions, not either, so a single false-negative open does not rescue a genuinely dead profile.
3. Zero site activity and zero product views in the window. This is the behavioral signal opens cannot give you.
4. Has not placed an order in the last 12 months. A buyer who ignores marketing email is still a customer, not a sunset candidate.
5. Was not created in the last 90 days. New subscribers are still inside the welcome window and have not had a fair chance yet.

Notice that clicks sit here as one signal among several, never the sole test. The next section shows why that single choice protects a healthy list.

What “Suppress Profiles” Actually Does in Klaviyo

Suppressing is not deleting, and the difference matters for both your bill and your data. When you suppress a profile in Klaviyo, it stays in your account with its full history, stops receiving email, and drops out of your active-profile count.

Deleting wipes the profile and everything attached to it, so if that person ever signs up again you have lost what you knew about them. Suppress by default, and delete only for compliance requests like a GDPR erasure.

The broader build errors tend to surface the moment we open an account, and they line up with what we check in a Klaviyo flows audit.

Why Defining “Unengaged” on Clicks Will Suppress a Healthy List

This is the mistake almost nobody warns you about, and it comes from an honest instinct. Marketers reach for clicks because MPP wrecked opens, so a click feels like the one signal left standing.

Here is the trouble. Across 31 active senders over a trailing 12 months, the median DTC click rate in our portfolio is 0.68%.

Sit with that number for a second. On a typical campaign, about 99 of every 100 recipients do not click a thing.

So a sunset segment defined on clicks over a short window will flag people who are opening, reading, and buying, then route them toward suppression. You would be culling your list on a signal that even healthy subscribers rarely produce.

Clicks are simply too rare to work as a liveness test on their own. Pair them with site activity and orders over a long window, so a low-click reader who is still shopping never lands in the sunset segment in the first place.

That is the whole reason step two of the build pairs clicks with site activity instead of trusting a click count alone.

30, 90, or 180 Days: What the Top Guides Actually Say

Here is where the advice fractures. Four of the most-cited sources on this topic, two of them Klaviyo's own, land on four different thresholds.

You can read Klaviyo's own sunset flow documentation for their version, but notice the table cannot even agree on what to measure, let alone for how long.

Our recommendation is 90 days of zero engagement to enter the flow, then suppression 14 days after the last email if nothing changes. Two reasons, both about the calendar.

Thirty to 45 days sits inside the normal purchase cycle for most DTC categories, so you would sunset people who simply have not hit their reorder moment. A quiet skincare customer at day 40 is not dead, they are between bottles.

At the other end, 180 days is half a year of paying Klaviyo to mail a subscriber who has shown no pulse. Ninety days is the point where lifecycle timing and cost stop pulling against each other.

What Suppression Costs You

Now the counterweight, because suppression is not free and most guides pretend it is. Cutting a subscriber removes them from every future campaign, so cut too early and you are deleting revenue, not protecting deliverability.

This is where our own position gets uncomfortable. Our unsubscribe calibration research, across 126 broadcasts over a trailing 12 months, found that the brands mailing hardest tend to earn more revenue per recipient, not less.

Put that next to the sunset decision. A subscriber sitting just outside your engagement window may still be worth mailing, which is precisely why the threshold belongs at 90 days and not 30.

Aggressive suppression feels responsible, and past a point it just shrinks the top of your revenue funnel. Suppress the genuinely dead, and leave the merely quiet alone.

Small Engaged List vs Large Unengaged List

This question comes up constantly, usually framed as a philosophy. It is really a math problem, and the answer is per-cohort, not per-list.

A smaller engaged list wins on two fronts. Deliverability improves when a higher share of your sends reach people who open and click, and your Klaviyo bill drops because you stop paying for profiles that never convert.

But “shrink the list” is the wrong instruction. The right one is to suppress the cohort that has gone dark on every signal while keeping the quiet-but-alive subscribers who still visit the site or buy.

A 200k list that is 60% engaged beats a 200k list that is 25% engaged, and it also beats an 80k list you built by suppressing everyone who missed your last three subject lines. Make the call cohort by cohort, on behavior, not against a target list size.

Making that call every quarter is ongoing work, not a one-time cleanup. It is a large part of what our Klaviyo team handles day to day.

Common Sunset Mistakes

Most sunset damage traces back to a handful of repeatable errors. These are the ones we see most.

• Sunsetting on opens alone in the MPP era. An “open” no longer proves a human is there, so you keep dead profiles while cutting live ones.
• Sunsetting profiles that never received a campaign. They are not unengaged, they are unmailed, and suppressing them tells you nothing about your list.
• Running sunset before winback. Sunset should fire only after a real re-engagement attempt, and how the winback flow should be built decides whether that attempt was even fair.
• Suppressing purchasers because they ignore marketing email. A repeat buyer who never opens is still revenue, and keeping post-purchase subscribers engaged is a flow problem, not a suppression one.
• Never suppressing at all. Let the dead weight pile up and your primary-inbox placement eventually drops for everyone, including the subscribers you most want to reach.

Frequently Asked Questions