A sunset flow is the procedure that stops you emailing people who stopped reading. It identifies chronically unengaged subscribers, gives them one last chance to re-engage, then suppresses the ones who do not. Most brands avoid it because the list gets smaller. The list was already dead. What shrinks is the part that was quietly damaging every send.
Use this SOP as a template
Copy or download the full procedure as Markdown or a Word file, then adapt it to your Klaviyo setup. Free, no signup, paste it straight into Notion, Google Docs, or ReccordSOP.
Mailbox providers score you on how recipients react. Send to thousands of addresses that never open, and inbox placement drops for the whole list, including the customers who do want to hear from you. Gmail and Yahoo now hold bulk senders to a spam-complaint rate under 0.3 percent, so an unengaged tail is a deliverability problem, not just a billing one.
Daily senders: no click or order in 90 days. Weekly: 120 to 180 days. Monthly: 12 months. The threshold is a number of sends without engagement, not a number of days.
Apple Mail Privacy Protection auto-opens messages, so open rate reports engagement that never happened. Clicks, site activity and orders are the signals that survive it.
Exclude anyone who subscribed in the last 30 to 60 days. A new subscriber who has not clicked yet is not unengaged, they are new.
Plenty of repeat buyers never click an email and order directly. Add 'has not placed an order in the last 180 days' to the entry filter or you will suppress paying customers.
Two or three emails over 10 to 14 days. Plain subject lines, a single obvious link, an explicit 'do you still want these?' preference option. No heavy design.
Suppress the profile or unsubscribe it. Deleting destroys purchase history and lets the address re-enter later through a form, restarting the damage.
Add the sunset segment as an exclusion on all campaigns and flows. A sunset flow that only stops campaign sends still leaks through automations.
Set it live continuously rather than as a one-off purge. Track inbox placement, spam-complaint rate and click rate on the remaining list, not list size.
Related guide
Five checks, a scoring rubric, and a quarterly cadence for catching broken triggers, aged-out segments, and revenue leaks.
Related guide
Sender reputation, list hygiene, and the sunset policy that keeps your flows out of the spam folder.
Record your screen while performing the klaviyo sunset flow. ReccordSOP turns the recording into a SOP with timestamped screenshots. When your live setup changes, drift detection flags it.
Generate this SOP freeAn automated flow that removes chronically unengaged subscribers from your active sending list. Profiles that have not clicked, visited or ordered within your engagement window enter the flow, receive a short last-chance sequence, and are suppressed if they still do not respond. The point is to protect inbox placement for the subscribers who are still reading.
Match it to how often you send, not to the calendar. A daily sender can judge someone at 90 days because that is roughly 90 chances to engage. A monthly sender needs 12 months to reach a comparable number of sends. Sunsetting a monthly-list subscriber at 90 days suppresses someone who has ignored three emails.
Revenue from the suppressed segment is usually near zero, since these are people who have not clicked or bought in months. The measurable effect normally runs the other way: better placement means the engaged majority actually see your sends. Run it on one cohort first and compare placement and click rate before and after.
No. Deleting loses order history and lifetime value, breaks your reporting, and does not stop the address returning through a signup form and starting the cycle again. Suppression keeps the record and keeps the address out of sends.
I built ReccordSOP after watching too many DTC ops teams lose months to undocumented workflows. These SOPs are battle-tested with Shopify operators running $1M to $50M brands.
Last reviewed June 1, 2026
We use essential cookies for sign-in and a small amount of analytics to improve the product. Privacy policy.