A customer buys once. The order ships, the thank you email goes out, then nothing.
Shopify repeat customers are the cheapest growth you own, and most stores watch them slip away without a fight. Across ecommerce the average repeat purchase rate sits at about 28.2% (Rivo, 2025). For every hundred people who trust you with a first order, roughly seventy two never place a second.
The reason usually isn't the product. It's timing. Most stores fire the same reorder email to everyone on day 30, or push a win back to the whole lapsed list on the first of the month. Klaviyo will send it beautifully. What it won't do is decide who should get it, and when, one person at a time. That is the gap agentic marketing has started to close, software that works each customer's own rhythm instead of the calendar's.
How many Shopify repeat customers should you expect?
A good repeat customer rate on Shopify runs between 20% and 40%, and the ecommerce average lands at about 28.2% (Rivo, 2025). Below 20% you're leaking faster than you should be; above 40% usually means a subscription or a consumable people genuinely run out of.
The average hides a lot, though. Repeat rates swing from 9.9% in luxury to 65.2% in grocery, so the only benchmark worth anything is your own category. If you sell coffee or supplements, 28% is nothing to celebrate. If you sell sofas, it's a strong year.
Start by finding your real number. In Shopify, open Analytics and read your returning customer rate over the last twelve months. Write it down. Everything below is about moving that one figure.
Why does the second order matter more than any other?
Because it's the hinge the rest of the relationship swings on. In a study of 18 million shoppers, only 32% of customers ever made a second purchase, but of the ones who did, 53% came back for a third and 64% of those went on to a fourth (RJ Metrics, 2015).
Read that curve again. The hardest jump in the whole journey is from one order to two. Once someone buys a second time they are on a different track, and the work to keep them drops with every order after.
So the email that earns you the most isn't the tenth one to a loyal regular. It's the one that turns a first time buyer into a two time buyer. The window that decides it opens the moment the first parcel lands, because a buyer is never warmer towards you than in the days after the thing arrives and actually works. That is when a pointed second offer lands, the exact item that pairs with what they bought, or the refill of the thing they'll finish, framed as the obvious next step and not a discount grab. Wait six weeks and hope a newsletter catches them, and you've handed the moment to the two in three who never reorder.
The hardest jump in the whole journey is from one order to two. Win it and every order after comes easier.
How do you time a replenishment email on Shopify?
You time it to the customer, not the calendar. A 40 day tub of protein bought by someone who trains five days a week runs dry weeks before the same tub bought by someone who trains twice a month, so a fixed day 30 refill flow is wrong for both of them.
The move is to read each customer's own rhythm from the gap between their orders, then land the refill nudge a few days before they run out. Refill due Friday, reminder on Wednesday, while the decision is still easy and before they've wandered off to a competitor's shelf.
Most stores can't do this one person at a time in Klaviyo, because flows fire on a fixed delay and a segment, however tight, is still a group being treated the same. There's a blunter version you can ship on Monday: pull your top three consumables, find the median number of days between the first and second order for each, and set that product's refill flow to that gap minus a few days. A per product timer beats one global timer every time, and you can sharpen it later.
Which flows actually drive repeat purchases?
The automated ones, by a distance. Across more than 183,000 brands, Klaviyo's flows drive about 41% of email revenue from just 5.3% of sends (Klaviyo, 2026). Campaigns do the shouting; the quiet, behaviour triggered flows do most of the earning.
Four flows carry repeat purchase: the reorder or replenishment nudge, back in stock, the browse follow up, and the win back for customers who've gone quiet. Each one waits for a real signal from a real person instead of a date on a wall calendar.
The leak is that nearly all of them run on fixed rules and one set of copy. A cart left for two days gets the same three email sequence whether it's a £15 refill someone simply forgot or a £400 first order they're still thinking hard about. Same timing, same words, two completely different customers. And the win back that fires at a flat 90 days will chase a monthly buyer who lapsed at week six long after you've lost her, and nag a quarterly buyer who was always going to come back. That sameness is where the flow revenue quietly drains.
The blast
- Day 30 refill email to the whole list
- One cart sequence for a £15 refill and a £400 order
- Win back fires flat at 90 days
Agentic
- The nudge lands the week her tub runs dry
- £15 refill and £400 order read as two people
- Win back timed to her own rhythm
What would it take to run every customer one to one?
More decisions than a person can make by hand, every day, for every name on the list. That's the work we're building at PilotX, the agentic marketing platform for consumer brands: four agents that run each customer instead of one flow running everyone.
Discovery learns a buyer's rhythm from their own behaviour. Decision picks the next best move and, just as often, chooses to wait. Delivery writes and sends it on brand, through Klaviyo or on its own, across email, SMS, push and WhatsApp. A Supervisor watches what worked and sharpens the next call. It goes live on Shopify and Klaviyo in about half an hour, and keeps Klaviyo as the sender if you'd rather it did.
Everything runs against a control group you set, a slice of customers held back from all of it, so the lift you read is yours and not the season doing the work. Modelled against that control group we're aiming at up to 50% more revenue, against the roughly 15% that segment and token personalisation tops out at. That figure is modelled, not measured, and we're early with our first brand live. The mechanism underneath it is old fashioned, though: right person, right message, right time, at a scale a human team can't hand run.
You don't need any of that to start this week. Find your returning customer rate, pick your three biggest repeat sellers, and fix the timing on their refill flow. If you'd rather see where the second order is leaking before you touch anything, our free Revenue Leak Audit maps the gaps across your Shopify and Klaviyo setup and puts a pound figure on them. When you want to see the per customer version working, that's how PilotX runs across Shopify and Klaviyo, and the 14 day pilot is built to prove the lift before you commit.
The second order is where the store you have turns into the store you're trying to build. Most brands never really fight for it. That's the opening.