Most stores send one abandoned cart email. It goes an hour after the cart is left, it says some version of "you left something behind", and it recovers money. Then the flow is finished and nobody opens it again.
One email is a long way better than none. It is also the easiest part of the job, because the people it recovers were mostly coming back anyway. They got distracted, the email reminded them, they finished. That is a real result and it is not the interesting part of the flow.
The second and third emails are where the work is, and they are only worth sending if they say something different. Three sends of the same message is not a sequence. It is nagging on a timer.
The flow works because people abandon for three different reasons
A cart gets left for one of three reasons, and they need different answers.
They were interrupted. Phone rang, tab closed, they were on the train. Nothing is wrong. They intended to buy and the session ended.
They were not sure. About the size, the delivery date, whether it is right for their situation, whether they can send it back. Something specific stopped them and it has a specific answer.
They were not ready. They are still looking. Price is a factor, or they are comparing you with two other tabs, or this is a considered purchase and it was always going to take a week.
One email can only speak to the first group. The other two are a different conversation, and if you never have it, you never find out how many of them there were.
Email one, at about an hour: the reminder
Short. The cart, a link back to it, nothing else.
No discount here, and this is the single most expensive mistake in the flow. The people opening an email one hour after abandoning are the closest to buying anyone in your database. Handing them 10% is paying a group who had already decided, and worse, it teaches your repeat customers that abandoning is how you get the code. That habit is not hypothetical. Stores with a standing discount in email one grow a segment of customers who never check out on the first pass.
An hour is a reasonable default rather than a rule. If your product is considered and expensive, two to four hours reads as less frantic.
Email two, at about a day: the objection
This is the one that has to do actual work, and it is where most flows fall back on "still thinking about it?"
The job here is to answer the thing that stopped them, and you already know what that thing is. It is in your support inbox. Whatever question your customers ask most before buying is the question that killed the cart, and it is usually one of returns, sizing, delivery time, or whether it works for a particular situation.
So email two answers it. Not as a policy paragraph in the footer, as the content of the email. If the top question is "will this fit", the email is about fit. If it is "how long does delivery actually take", the email says the number.
A review does this well when it is a specific one. "Great product, love it" is decoration. A review from someone who says they were unsure about the size and it worked out is the objection being answered by someone who is not you.
Email three, at about three days: the one nobody writes
By day three, the reminder has been sent and the objection has been answered. If they have not bought, the truthful position is that they are not ready, and the flow has run out of things to say about the product.
Which is why the third email is usually one of two things, both wrong. Either it is the same cart image a third time with a louder subject line, or it is a discount fired at everyone still standing, including the people who were going to buy on Friday.
The third email should change the subject. There are three honest versions.
Widen it. They looked at one thing. Show them the two things next to it. Some abandoned carts are not a cart problem, they are a wrong-product problem, and the person who bounced off one item buys a different one the same week.
Answer the bigger question. Not "buy this", but the thing behind the purchase. If someone abandoned a running shoe, the email about picking a shoe for your gait is useful whether or not they buy today, and it is the email that makes them come back to you rather than to search.
Close it, and mean it. If a discount belongs anywhere in this flow it is here, on day three, to people who have now had two emails and not moved. Say it is the last email about it. Then make it the last email about it.
Any of these is better than a third reminder. What makes the difference is that the email admits time has passed instead of pretending it is still the hour after.
Where the extra emails actually come from
Everything above assumes the flow is being seen by everyone who abandons, and in most accounts it is not. Abandoned cart triggers on a browser event, and browser events go missing at a rate that surprises people the first time they check it. That gap is usually worth more than the third email is, and it is worth checking before you write anything.
Build the three emails. Then go and find out how many people are actually receiving them.
Should the third email always have a discount?
No, and defaulting to one is how a store trains its customers to wait. Use it when margin allows and when the first two emails have already done their jobs. If the third email is your only idea, the flow has a content problem that a discount will hide for about a quarter.
How long should the whole flow run?
Three emails across three days covers most of it. Past four or five days the intent has gone cold and you are into winback, which is a different flow with a different tone. Stretching abandoned cart to a fortnight does not recover more carts, it just annoys people who bought something else in the meantime.
