If you sell something people buy more than once, your biggest growth lever isn’t acquisition. It’s getting the customers you already paid for to come back. Most brands know this. So they switch on a post-purchase flow, set a couple of delays, add a discount, and move on. Six months later the revenue is underwhelming, and the instinct is to blame the channel. Email is tired, people say. It isn’t. The channel is fine. The decisions are the problem.
The channel isn’t the problem
Email and SMS are extraordinary at one thing: getting a message to the right inbox, at scale, cheaply, with consent handled. That is execution, and the tools that do it are very good at it. What they don’t do is decide. They don’t know that this customer reorders every nineteen days and that one every forty-five. They don’t know that this buyer never needs a discount and that one won’t move without one. They send what you told them to send, on the schedule you set, to everyone in the segment. Precision was never their job.
So when a post-purchase program underperforms, it’s rarely because the message went to the wrong place. It’s because it was the wrong message, at the wrong time, to the wrong person, with the wrong offer. That is a decision failure, not a delivery failure, and no amount of better copy or nicer templates fixes a decision problem.
What post-purchase flows usually get wrong
Almost every underperforming flow makes the same four mistakes:
- One timeline for everyone. A flow that fires on day 30 treats a coffee buyer and a supplement buyer identically, even when their real reorder windows are weeks apart.
- The same product to everyone. A best-seller recommendation block ignores what this specific customer is actually likely to buy next.
- A discount by default. Blanket incentives train full-price buyers to wait for a deal and quietly erode the margin you worked to protect.
- Send, always. Most flows have no concept of doing nothing. Every eligible customer gets the email whether or not it helps.
A program underperforms not because the email went to the wrong inbox, but because it was the wrong message, at the wrong time, to the wrong person.
What actually drives LTV
The brands that get real lift out of post-purchase don’t send more. They send smarter. Four decisions do most of the work:
- Reorder timing. Find the day a customer is genuinely likely to run low and rebuy, by product, and reach them then. Not day 30. Their day.
- The next right product. Use what customers actually buy after a given first purchase, not a generic popularity list that’s the same for everyone.
- Offer depth that fits the person. Protect margin on full-price loyalists. Spend incentive only where it changes behavior, not where the sale was coming anyway.
- Knowing when to stay quiet. Choosing not to send is a decision too, and a valuable one. It protects attention, deliverability, and margin.
Stop asking “what should we send?” and start asking “what should happen next for this specific customer?”
Decisions, not sends
None of this means tearing out your stack. Keep it all: Klaviyo, your SMS tool, your subscription app. They are the execution layer and they’re good at it. What’s missing in most brands is the layer above them. The part that looks at each customer’s history and decides what should happen next, then hands the existing tools a clear instruction.
That is the difference between a post-purchase program that looks busy and one that actually grows lifetime value. The busy one sends. The effective one decides, then sends. Same channel, same customers, completely different result.
So if your post-purchase revenue has plateaued, resist the urge to send more. Look at the decisions underneath the sends first. That is almost always where the growth is hiding.
Key takeaways
- The channel is rarely the problem. The decisions behind it usually are.
- Generic timing, generic products, and blanket discounts quietly cap retention revenue.
- Real LTV growth comes from per-customer timing, product, and offer decisions.
- Sometimes the best decision is to send nothing at all.