Almost every store we take over has more Klaviyo flows than it needs. Fifteen, twenty, most half-built, several overlapping, and four of them doing effectively all of the work. The problem is not that the others are bad. It is that each one is a thing that can break quietly, and nobody audits twenty flows.
The four that earn
Abandoned checkout. Someone entered payment details and stopped. This is the highest-intent moment in the whole store and it is worth more than everything below it combined. Build it first.
Welcome. Triggered by a signup, usually from the popup. Its job is not to sell, it is to set expectations and get the first open, because a subscriber who ignores the first three emails will ignore all of them.
Browse abandonment. Viewed a product, did not add to cart. Weaker intent than the checkout flow and it still earns, provided you exclude anyone already in a cart or checkout flow so people are not receiving three emails about one visit.
Post-purchase. The cheapest revenue in ecommerce is a second order from someone who already trusts you. This flow is also where returns and support tickets get prevented, by telling people what happens next before they have to ask.
That is the set. Everything after it is optimisation, and optimisation before those four exist is rearranging furniture in a house with no roof.
Why that order specifically
Because it is descending intent, and intent is the only thing that reliably predicts whether an email earns.
People build in the opposite order, starting with the welcome flow because it feels like the beginning of the customer journey. It is the beginning of the relationship and the end of the queue for revenue, because a new subscriber is a stranger and someone who abandoned a checkout is not.
The part nobody says about flow revenue
Klaviyo will tell you a flow generated a number. Treat that number as an upper bound rather than a measurement.
Attribution windows mean an email that arrives an hour before someone was going to buy anyway gets full credit for that order. Some proportion of abandoned checkout revenue is people who had left a tab open and were always coming back. Nobody knows the exact proportion, including us, and anybody quoting you one is guessing.
This is not a reason to skip the flows. It is a reason not to make decisions as though the attributed figure were incremental profit. The useful comparison is what happened to total revenue in the months after the flows went live, not what the flow dashboard claims.
Deliverability decides everything above it
A flow that lands in Promotions or spam earns nothing regardless of how good the copy is, and this is where most of the damage happens invisibly.
Two things matter more than any other setting. Authenticate your sending domain properly, which means DKIM and a dedicated sending subdomain rather than sending as your raw domain. And suppress unengaged subscribers on a schedule rather than emailing everyone forever, because continuing to send to people who never open is the single fastest way to teach an inbox provider that your mail is unwanted.
Sunsetting subscribers feels like throwing away a list you paid for. You are not throwing away revenue, you are protecting the deliverability of the people who do open.
What we do first on a new account
Before building anything, we check whether the four flows exist, whether they are actually live rather than in draft, whether their exclusions overlap, and whether the sending domain is authenticated. On the accounts we have taken over, at least one of those four has been wrong roughly every time.
Finding a flow that has been sitting in draft for eight months is a better first week than writing a new one.
If you want this looked at on your store, that is part of what our Klaviyo work and email and SMS service covers.
