Your second purchase rate is where most ecommerce brands actually die

Retention fails long before the loyalty programme gets involved

Retention gets talked about as a loyalty problem. The VIP tier, the customer on their sixth order, the subscription ticking along nicely. By the time somebody's on order six, retention has already worked. The programme is celebrating survivors.

The leak that actually costs brands sits much earlier and gets almost no attention. The gap between a first order and a second is where the money quietly leaves. Close it properly and every retention curve you build afterwards gets easier.

We hear the same sentence from founders every month in different words: acquisition looks great, so why does the business feel like it's standing still. The answer is rarely the ads. It's that the second order never got a plan of its own.

One order means they tried you, not that they chose you

Most first orders happen on a discount code, a good ad or a friend's recommendation. The customer's relationship is with the deal, not the product range, and definitely not your newsletter's opinion on the new season colours. Nothing about that order says they've decided to come back.

The mistake is treating that person as an existing customer the moment the confirmation sends. They get merged into the general list and get the same broadcasts as somebody four orders deep. Nothing in their inbox acknowledges that they're one order in and still deciding.

First-time buyers need their own track, built to answer the only question that matters to them yet: was that first purchase worth repeating. The brand story and the wider catalogue can wait until that's settled.

Time the second purchase push off the product, not the calendar

Most versions of the post-purchase flow run on fixed intervals. An email a week after delivery, another a week after that. That ignores the one variable deciding whether the email lands or gets ignored: what the person actually bought.

A consumable that empties in five or six weeks needs its push around week four, while there's product left and the customer's starting to think ahead. Send a generic thirty-day email and it either arrives too early to matter or lands after they've run out and gone looking on Amazon. A durable product needs a different second act. Built around what naturally follows it.

So one flow can't do this job. It needs branching off product category, or ideally the actual item, so the serum buyer and the one-off homeware buyer sit on different timers. Setting each timer off the replenishment cycle of what was bought is the single biggest fix most brands can make here.

Recommend what follows the first purchase, not the bestseller feed

Get the timing right and the content is where it falls apart next. Too many second-purchase emails default to bestsellers or a generic you-might-also-like widget. That's a missed connection wearing a personalisation label.

The stronger version reasons from the item itself. The coffee grinder buyer should see beans and filters, not another kitchen gadget. The buyer of one foundation shade should see the concealer built to sit alongside it, not a rival shade they've no reason to own. It's a mapping exercise, done product by product, and none of it is complicated.

Treat the pairing as a test, not a guess. The obvious complementary product is often not the one that converts, so let the email prove which pairing earns the click before you hard-code it.

A working first-to-second track has five things true at once

• It triggers for first-time buyers specifically, separate from the general newsletter and the catch-all thank-you sequence
• Timing is set per product category or replenishment cycle, never one fixed delay for the whole catalogue
• The recommendation is mapped to the item actually bought, not pulled from a bestseller feed
• The copy still assumes the customer's undecided, instead of borrowing the loyal-customer voice
• If the second order doesn't come, the customer graduates back into general lifecycle messaging instead of sitting in limbo

Most brands already have some version of this live, and very few clear all five at once. It's usually the timing and the recommendation logic that fail. Fixing those two is a far smaller job than rebuilding the flow from scratch.

A report like that will happily show a post-purchase flow as live and earning. It'll never tell you that every buyer inside it sits on the same timer, which is the actual fault in most accounts.

Open the flow and check two things before you build anything

Open your post-purchase flow in Klaviyo and look at exactly two things. First, whether every product category shares one timing. Second, whether the product block in the second or third email is a generic feed or something mapped to what the customer bought.

If both come back as one generic template, you've found the gap costing you the most repeat customers. It's the pattern we see on most of the accounts we take over, and it's a smaller fix than most of the retention work brands go looking for first. The loyalty tier and the subscription push both get easier once it's closed.