A small online coffee roaster spends a good share of its marketing budget winning first orders through ads and promotions. Plenty of those customers like the coffee. Most of them still don't order again, because nobody reminds them at the moment their bag runs out, and the supermarket is on the way home. The store sends a monthly newsletter to its whole list, with the same discount for everyone, and the people who were about to reorder anyway use it while the people who drifted away ignore it.
For many online stores, the second order is the one worth working hardest for. A customer who orders twice has shown they like both the product and the experience, and reaching an existing customer by email costs far less than winning a new one through ads. Yet repeat purchasing is often left to chance, or to a newsletter that treats every customer the same way.
This article covers what to learn from your own order data first, and what an automated repeat purchase setup does around Shopify and your email and text tools.
What one-time buyers cost
If you pay to acquire customers through ads, marketplaces or promotions, a one-time buyer often barely covers the cost of winning them. The profit in many store models comes from the second, third and fourth orders. A store with a low repeat rate has to keep paying for new customers just to stand still.
Blanket discounting adds another cost. Discounts sent to everyone often get used by loyal customers who would have paid full price anyway, which gives margin away without changing behavior.
Learn from your order data first
Find your repeat rate and timing
Export your order history from Shopify and work out what share of customers have ordered more than once, and the typical number of days between a first and second order. Break it down by the product bought first. These numbers vary enormously between stores and products, and they tell you where to focus.
Understand how products get used
For consumables like coffee, skincare, pet food or supplements, estimate how long a typical purchase lasts. A rough figure based on pack size and typical use is a good starting point, and the actual intervals of your repeat customers will refine it. For durable goods like furniture or clothing, repeat orders usually come from related products, accessories or seasonal needs, so the approach is different.
Segment simply
Start with a few groups: first-time buyers, repeat customers, customers whose expected reorder time has passed, and lapsed customers who haven't ordered in a long time. Each needs a different message.
Set a discount policy
Decide when a discount is used. One common approach is to keep discounts for win-back messages to lapsed customers, and rely on timing, relevance and convenience for everyone else.
Check consent and preferences
Make sure you know who has agreed to receive marketing emails and texts, and that your store and email platform agree on it. Retention messages to people who haven't opted in will hurt deliverability and may break the rules where you sell.
What the automated version does
Email and text platforms such as Klaviyo and Shopify Email can run automated flows from your store data. A repeat purchase setup built on them typically includes:
- A helpful post-delivery message. A few days after delivery, the customer gets something useful about what they bought: a brewing guide, care instructions, a sizing tip. No sales pitch.
- A replenishment reminder. Timed to when their product is likely to run out, based on what and how much they bought, with a link that rebuilds their last order in the cart so reordering takes seconds.
- A relevant suggestion. Based on what customers with similar first orders bought next, a single well-chosen recommendation instead of a catalog.
- A second-order nudge. If a first-time buyer passes their expected reorder time without ordering, a short, personal check-in asks how they got on with the product.
- A subscription offer. For consumables, customers who have reordered once or twice are offered a subscription on their usual interval.
- A win-back message. Customers well past their expected reorder time get a personal note from the store, and this is where an offer can make sense.
- Reply handling. When customers reply ("the dark roast was too strong for me"), the message is read and a suggested response is drafted, such as recommending a lighter roast, for someone to approve and send.
- Reporting. A monthly view shows repeat rate, time to second order and revenue from each flow.
The AI work is in the timing and the writing: estimating each customer's likely reorder date from their history and similar customers, choosing the most relevant suggestion, writing messages in the store's voice, and drafting replies. The flows and rules are set by you.
Our repeat business engine builds this on your store and email platform, and the ecommerce and retail page shows how it fits with listings, reordering and reviews.
A worked example
Take the coffee roaster from the opening. Its order export shows the typical gap between first and second orders for customers who do reorder, and that customers who buy the smaller bags tend to come back sooner than those who buy a kilogram. The store sets up a replenishment reminder for the best-selling blend, timed from those intervals and adjusted for bag size, with a link that puts the same order back in the cart. A post-delivery email with a brewing guide goes out first, with no offer attached. Customers who pass their expected reorder date get a short check-in from the founder asking how they're finding the coffee. Those who reply that it's too strong get a drafted response suggesting a lighter roast, which someone checks and sends. Customers who have reordered twice are offered a subscription. Only customers several months past their expected reorder date get an offer, as part of a personal win-back note.
What to watch out for
- Message overload. Several flows running at once can bombard a customer. Set a cap on how many marketing messages anyone gets in a week, and have flows check it.
- Discount dependency. If every message carries a discount, customers learn to wait for one. Use offers deliberately, mainly for lapsed customers.
- Consent rules. Email and text marketing are regulated, with rules on consent, identification and unsubscribing that vary by country. Texting has stricter consent requirements in many places, including under the Telephone Consumer Protection Act in the US.
- Personalization that feels intrusive. Referring to someone's purchase is helpful. Referring to how often they browse a page they didn't buy from can feel uncomfortable. Keep personalization to what customers would expect you to know.
- Measuring what the flows actually add. Some customers who order after a reminder would have ordered anyway. Holding back a small random group from each flow shows how much the flow really changes behavior.
- List quality. Sending to old, unengaged addresses hurts deliverability for everyone. Suppress addresses that haven't engaged in a long time.
How to tell whether it's working
Track the repeat purchase rate for new customers in each monthly cohort, the median time to second order, and revenue per customer over their first six months. Compare customers who received each flow with the holdout group to see what the flows actually add. Watch unsubscribe and spam complaint rates for each flow, and the share of orders that used a discount, since rising discount use without a rise in repeat rate means offers are being given away. Subscription sign-ups and retention are worth tracking separately once they're offered.
Where to start this week
Export your orders and calculate two numbers: the share of customers who've ordered more than once, and the median days between first and second order for your top three first-purchase products. Then set up one replenishment reminder for your best-selling consumable, timed from that median, with a one-click reorder link. Run it for two months, ideally with a small holdout group, and compare repeat rates.
That single flow will tell you more about your customers' reordering habits than any newsletter, and it gives you a template for the rest of the setup. Once it's working, adding the post-delivery message, the second-order nudge and the win-back is a matter of applying the same approach to the rest of your catalog.
If you'd like a quick read on where repeat business sits among the other jobs worth automating in your store, run the free PATH report on your website. To talk it through, book a call.