Sunset Flow (Klaviyo): Clean Your List Safely

A sunset flow (Klaviyo) is what stands between a list that quietly drags your inbox placement down and one that doesn’t. This guide covers how to define the unengaged segment by behaviour instead of guessing at a day count, how to run the re-permission sequence, why suppression and deletion are different decisions in a sunset flow, and the mistake that sunsets recent buyers who simply don’t open email.

What is a sunset flow (Klaviyo), and why does it matter?

A Klaviyo sunset flow is an automated sequence that finds profiles who have stopped engaging with your email, gives them a short, direct chance to opt back in, and suppresses the ones who don’t respond. It exists to protect the sending reputation of your whole domain, not to punish individual subscribers.

Inbox providers like Gmail and Yahoo weight engagement — opens, clicks, and the inverse, complaints and spam-folder moves — heavily in deciding whether your next send lands in the inbox or the spam folder. A list carrying a large, permanent block of addresses that never open anything drags the average engagement rate down for every subscriber on it, including the ones who buy from you every month. A sunset flow removes that drag on a rolling basis instead of letting it accumulate.

A win-back flow is a different flow entirely: it targets lapsed purchasers and tries to bring the sale back, usually with an offer. A sunset flow targets email engagement specifically, and its metric of success is deliverability health, not revenue recovered. The two flows can pull from overlapping audiences, but conflating their logic is how teams end up either emailing dead addresses forever or discounting to people who were never going to buy again anyway.

How do you define the unengaged segment?

Define “unengaged” by behaviour over a rolling period combined with a minimum number of sends received in that period — not by a single flat day count applied to every profile regardless of how much email they actually got. A profile that joined your list three weeks ago and received two campaigns hasn’t had a fair chance to prove itself unengaged; a profile that’s received twenty sends over the same window and opened none of them has.

In Klaviyo’s segment builder, that means combining a “Did Not Open” (or “Did Not Click,” for a stricter definition) condition over your chosen window with a second condition requiring a minimum count of campaign or flow emails received in that same window. Requiring both conditions — not opening, and having had enough opportunity to — is what keeps genuinely new subscribers out of a segment meant to catch profiles that have had a real, sustained chance to engage and didn’t take it.

Whether “Did Not Open” or “Did Not Click” is the right base condition depends on how much you trust open-tracking on your list. Apple Mail Privacy Protection and similar features inflate open rates by pre-fetching images regardless of whether a human read the email, which makes “Did Not Open” a noisier signal than it used to be. Many teams now lean more heavily on click behaviour, or on a combination of the two, rather than trusting opens alone to define engagement.

The step most teams get wrong: sunsetting recent buyers who don’t open email

The most common mistake in a Klaviyo sunset flow is building the unengaged segment purely on email behaviour — opens and clicks — with no exclusion for purchase activity. That segment definition catches a real category of customer that has nothing wrong with it: someone who buys from your store, is happy with the product, and simply doesn’t open marketing email. They order through a direct link, a saved bookmark, a search, or an app notification, and your campaigns sail past them unopened every time.

Sunset that person and you’ve suppressed an active, revenue-generating customer from ever receiving another campaign, sale announcement or back-in-stock alert — not because anything is wrong with the relationship, but because your segment logic mistook “doesn’t open email” for “doesn’t want to hear from us.”

The fix is a second, excluding condition on the same segment: exclude any profile with a placed-order event, or an active-on-site event, inside your engagement window, before the “did not open” condition is allowed to qualify them. In Klaviyo’s segment builder this is an AND-NOT relationship — the profile must fail the engagement condition AND must not have the purchase or site-activity condition — built as a segment with both a qualifying and an excluding condition group, not two separate segments joined loosely by process.

Any other signal that indicates a live relationship even without email opens extends the same exclusion logic: a recent support ticket, an SMS click if you run SMS separately, or a loyalty-program action, if you track one. Anything that proves the person is still transacting or interacting with your brand belongs in the exclusion list before the sunset flow ever sees that profile.

How do you run the re-permission sequence?

Once the unengaged segment is defined correctly, trigger a short flow of emails against it, spaced a few days apart, each asking plainly whether the recipient wants to stay subscribed. This is not the place for a normal promotional tone or a discount code buried three paragraphs down — the entire job of this sequence is to generate one click, either “keep me subscribed” or an unsubscribe, and a long or cluttered sequence just adds more unopened sends to the exact problem the flow exists to solve.

Put a clear, one-click “keep me subscribed” link in every email in the sequence, and make the unsubscribe link equally visible rather than buried in small footer text — a re-permission email that makes leaving hard defeats the purpose of the exercise, which is separating people who genuinely want your email from people who don’t, cleanly.

Set the flow’s exit logic so any qualifying response — a click on “keep me subscribed,” or an open of any email in the sequence, depending on how strict you want the bar to be — routes the profile out of the sunset flow and resets its engagement clock. Everyone who doesn’t respond by the end of the sequence moves to suppression.

Suppression versus deletion in a sunset flow: which one, and when?

Suppression and deletion are different decisions with different consequences, and a sunset flow should default to suppression.

Suppressing a profile — adding it to a suppressed list or applying Klaviyo’s global exclusion — stops all campaign and flow sends to that address while keeping the underlying profile, its purchase history, its lifetime value and its event data intact. That matters for three reasons: attribution reporting still counts their historical orders correctly, segmentation built on lifetime spend or product history still works, and if the profile re-engages later — a new purchase, a click from a transactional email — you can route it back into active sending without having lost anything.

Deletion removes the profile from Klaviyo entirely: no history, no re-permission path, no attribution record. That’s the right call for a genuine data-deletion request, or as a deliberate retention-policy decision applied on a schedule — not as the automatic endpoint of a sunset flow. Treating “didn’t respond to two emails” as grounds for deletion throws away order history and attribution data for no deliverability benefit beyond what suppression already delivers.

How does a sunset flow interact with where subscribers come from?

A sunset flow drains unengaged profiles, but if your sign-up forms and pop-ups keep feeding new unengaged profiles in at the same rate, you’re running the flow in place rather than making progress. Before or alongside building the flow, check which acquisition sources — a pop-up offering a broad discount, a co-registration partner, an in-store SMS-to-email capture — are producing the highest share of profiles that later land in your unengaged segment, and either tighten the offer, add a confirmation step, or pause the worst-performing source.

A Klaviyo sunset flow also depends on your wider data hygiene: if Klaviyo isn’t receiving accurate order or site-activity events from Shopify in the first place, the purchase-exclusion condition in your segment definition won’t work reliably, and you’ll sunset buyers regardless of the exclusion logic. Confirm your order and site-activity sync is accurate before trusting any engagement-based segment built on top of it.

How do you verify the flow worked?

After the flow has run for a full cycle — long enough for the re-permission sequence to complete and the suppression step to fire — check three things in Klaviyo: the unengaged segment’s current size relative to its size before the flow launched, the sending-domain-level engagement rate on your next few campaigns, and a spot check of the suppressed list for any profile with a recent order date, which would indicate the purchase-exclusion condition isn’t set up correctly.

A working sunset flow shows a smaller, more responsive unengaged segment over time, a rising open and click rate on subsequent campaigns as the suppressed block stops dragging the average down, and zero recent purchasers appearing in the suppressed list. If purchasers are showing up suppressed, the exclusion condition from the “step most teams get wrong” section above needs to be checked first.

A sunset flow is a lifecycle flows problem before it’s a deliverability problem — the segment logic, the exclusion conditions and the suppression path all sit inside the same flow architecture that runs your welcome series, abandoned cart and win-back sequences, and building it in isolation from the rest of your flows is how the purchase-exclusion step gets missed in the first place. If your flow build doesn’t already account for how much revenue passes through it, the flow revenue calculator is a starting point for seeing what a properly maintained flow programme is worth protecting. Pointerflow builds and maintains lifecycle flows — including sunset and re-permission logic — for scaling ecommerce brands that can’t afford a flow architecture nobody is watching.