How many emails should an abandoned cart flow have?

Most recovered orders come back on the first message, a smaller share on the second and fewer still on the third. That shape is consistent enough that adding a fifth and sixth email rarely changes revenue while reliably raising complaints.

Send the first within a couple of hours. Soon enough that the shopper still remembers the product, late enough that you are not interrupting somebody who stepped away and is coming back. Sending within minutes reads as surveillance and often reaches people who were still shopping.

Send the second around a day later, and the third around a day after that. The gaps matter more than the exact hours. Three emails in one evening feels like pressure; three across two days feels like a service.

Then stop. If the person has not returned after that, they are better served by your normal campaign schedule than by more messages about a basket they have moved on from.

The first is a reminder and nothing more. Show the items with images, make the return link obvious, and assume the reason for leaving was a distraction rather than a decision. No discount, no argument, no urgency.

The second addresses hesitation. This is where the actual reasons live: delivery cost or speed, whether it will fit, whether it can be returned, whether the store is trustworthy. Answer those directly with policy details, a size guide, a review, or a link to help. This is usually the message with the most room for improvement in an average flow.

The third closes. Stock scarcity if it is true, a reminder that the basket will not be held indefinitely, or an incentive if you have decided to use one. Never invent scarcity you cannot substantiate, because it is checkable and it damages the relationship when it is not real.

A fourth, if you add one, works best as a change of angle rather than a repeat: a different product from the same category, a customer story, or a simple offer of help from a person.

The largest gains usually come from segmentation rather than from more emails. A basket worth a small amount and a basket worth several hundred deserve different treatment, different copy and different incentives.

Split by value first. High-value baskets justify a longer flow, a more considered second message and sometimes a personal follow-up. Low-value baskets rarely justify a discount at all, because the margin cannot carry it.

Split by customer history second. Someone who has bought three times before does not need to be reassured about your returns policy and should certainly not be given a code, since they were likely to come back anyway. A first-time visitor needs the reassurance and may need the incentive.

Split by what was in the basket third, where your catalogue has genuinely different categories. A message that speaks to the product type outperforms a generic one, and most email platforms can do this with the product data already in the event.

The flow depends on capturing an identity before the basket is abandoned. That means an email address collected at the start of checkout, from a previous session, from a signup form, or from a click in an earlier email. Without one there is nobody to send to.

Check the trigger. Platforms differ in whether the flow starts from a cart event, a checkout event or both, and the difference determines how many people enter it. Many stores run a checkout abandonment flow and believe they are running a cart one, which explains a smaller audience than expected.

Exclude people who completed the order. A recovery email arriving after someone has bought is the most damaging failure in this flow, and it usually comes from a delay in the purchase event or from an exclusion rule that was never set.

And respect consent. Where marketing consent is required, transactional-style reminders and marketing messages are not the same category, and the rules differ by market. Get this settled before the flow goes live rather than after a complaint.