Learning Log
Abandoned cart emails: what to say before you offer a discount
An abandoned cart is usually someone still deciding, not someone who said no. The flow's job is to show up at the right moment with the right answer, and the answer isn't always a discount.
Most abandoned cart flows I'm handed open the same way. You left something behind - then a code.
It works well enough that nobody questions it. A discount does recover carts. But it also teaches every customer who gets one that leaving a cart is how you unlock a lower price, and it assumes the only reason anyone hesitated was money.
Across three brands we run cart flows for - a fashion accessories brand, a heritage sweets brand and an outdoor gear brand - the reason people hesitate is rarely just money. And the right flow looks different for each of them.
Why carts get abandoned: still deciding, not saying no
Picture what's actually happening when a cart is left. Someone has found a product they like enough to add. Then they stop.
Most of the time they haven't rejected it. They're browsing, they can't quite decide, and they're comparing - your site in one tab, two others in the tabs next to it. They're checking whether the same thing is cheaper elsewhere, whether a competitor's version looks better, whether they really need it this week.
That changes what a cart email is for. It isn't a reminder that they forgot something. It's your one chance to rejoin a decision that's still in progress, and it's competing with every other brand they had open.
So two things matter more than the discount: when you show up, and what question you answer when you do.
It depends on the industry you're in
There's no single right cart flow, because there's no single way people buy. Before deciding anything about timing or offers, work out where your product sits between two kinds of purchase.
Impulse purchases. In fashion and accessories, fast buying is everything. The customer saw it, wanted it, and was one tap from buying. What stopped them is usually momentum, not doubt - a notification, a meeting, a second thought about spending. The email's job is to bring the impulse back before it cools: the product, big and beautiful, back in front of them while they still remember the feeling.
Considered purchases. In outdoor gear, specialist food, anything technical or gifted, the customer is comparing and they have questions. Will it fit? Will it last? Is it fresh? Which one should I pick? Is this the right gift? Showing them the product again does nothing if the question is still open. The email's job is to answer it.
Our three brands sit at different points on that line. Vibes Shades is close to pure impulse: sunglasses are a wanting purchase, and the flow leads with the product. Tahan Outdoors is at the considered end: people buying gear for heat and rain compare specs and want to know it'll survive. Rohit's Mithai sits in between - mithai is often bought for someone else, so there's desire, but also questions about choice, freshness and what will land well.
Everything that follows - timing, who gets the flow, what the discount comes with - shifts depending on where you sit.
Test 1: send time - 2 hours, 4 hours or 1 day
The first email's timing is the biggest lever most flows never test.
The test is three arms: 2 hours, 4 hours and 1 day after the cart is abandoned. The question is which one captures attention best and closes the sale fastest.
Each arm has a logic:
- 2 hours catches people while they're still comparing. The other tabs may still be open. For an impulse product, this is often where the feeling is still warm.
- 4 hours gives them room to finish comparing without letting the day end. Fast enough to still be relevant, slow enough not to feel like being chased.
- 1 day suits a considered purchase, where the customer may genuinely need to think, check with someone, or come back with a clear head.
I have instincts about which wins for each category. I don't trust them enough to skip the test, and nor should you. The point of running three arms is to find out when your customer is still in the decision, not to confirm what a benchmark said.
How to set it up:
- Change only the delay. Same subject line, same content, same offer in every arm. If you move two things at once, you can't tell which one worked.
- Measure orders from the flow and time to purchase, not opens. A fast send that gets opened and ignored hasn't done its job.
- Run it long enough to collect a meaningful number of abandoned carts in each arm before calling it.
Test 2: include or exclude people still in the welcome flow?
This is the question almost nobody asks.
New subscribers are, by definition, still surveying. They've just met you. They're browsing, adding things to a cart to see the total, and thinking about it. That makes them more likely to abandon a cart than someone who already knows the brand.
So should they get the cart flow while they're still inside the welcome flow?
There's an argument both ways. Two flows firing at once can talk over each other - a story email from the welcome sequence and a "you left this behind" email landing on the same day, pulling in different directions. But the cart email might also be exactly the nudge that turns a curious new subscriber into a first order.
I don't know which is true for every brand, and I'd be suspicious of anyone who says they do. So A/B it:
- Arm A: new subscribers in the welcome flow also receive the cart flow.
- Arm B: they're excluded from the cart flow until the welcome flow has finished, or the welcome flow pauses while the cart flow runs.
Then see which closes the deal. If you've built a welcome flow that tells the brand story properly - the argument in Lab Note 08 - this test tells you whether the cart email helps that story land or interrupts it.
When it isn't about price: close the uncertainty
Sometimes the hesitation isn't price at all. It's a question, or a bit of uncertainty the product page didn't resolve.
What size am I? Which colour suits me? Will this stove work at altitude? How long will the sweets keep? Is this box big enough for a family gathering?
A discount doesn't answer any of those. It just makes an unanswered question slightly cheaper.
What answers them is a person. So in the cart flow, offer a direct line: WhatsApp support, or another direct channel where they can ask and get a reply from someone who knows the product. Put it in the email itself, not buried in a footer - Not sure? Message us on WhatsApp and we'll help you choose.
Tahan's cart flow already does this. Outdoor gear is the most considered purchase of our three brands - people want to know a tent will hold in a tropical downpour before they pay for it - so the cart email gives them WhatsApp support to ask exactly that, from someone who knows the kit.
A conversation closes what a discount can't, and it doesn't cost you margin. It also tells you, over time, what your product pages are failing to explain - which is worth more than the sale.
When the discount comes in, it never travels alone
None of this is an argument against discounts. Promotions close decisions, and a cart flow is a sensible place for one.
The argument is about what goes with it. A discount on its own is just a number, and a number is easy to compare against the other tabs. When the offer appears, it should be packaged with three things:
- Features - if there are any worth naming. What the product is and does.
- Benefits - what changes for the customer when they have it.
- What they get if they buy now - the price, the item secured before it sells out, being ready for the occasion they're buying for.
The mix shifts by category. In fashion, lead with the product and the buy-now reason; for an impulse product, a fast offer can be the right call. For a considered purchase, lead with features and proof, and let the discount be the final push once the questions are answered.
It's the same principle as Lab Note 04: a reason to buy, plus a promotion for a limited time. Never a percentage on its own. And it's why leading every email with money off ends where Lab Note 05 describes - customers learn the next email is just another sale.
A test plan you can copy
If you run one cart flow and have never tested it, this is the order I'd go in:
- Name your category. Impulse, considered, or somewhere between. Write down what you think stops people. This is your hypothesis, not your answer.
- Test send time. 2 hours vs 4 hours vs 1 day. Change nothing else.
- Test the welcome-flow overlap. Include vs exclude new subscribers.
- Add a direct line. WhatsApp or similar, in the email body. Watch what people ask.
- Then decide where the discount sits - and when it appears, package it with features, benefits and what they get by buying now.
One test at a time. Move everything at once and you'll get a result you can't explain, which is the same as having no result.
The limits
- Three brands, three categories. The reasoning travels; the winning arms won't. Your 2-hour winner might be someone else's 1-day winner.
- This is a test plan, not a results report. Apart from Tahan's WhatsApp line, these are tests to run, not tests we're reporting on. The value is in what to test and why, and you'll learn more from running it on your own list than from reading anyone's benchmark.
- A direct line has to be staffed. Offering WhatsApp support and replying the next afternoon is worse than not offering it.
Open your cart flow and read the first email as if you'd just left the cart with two other sites open. What does it give you that the other tabs don't?
Frequently asked questions
What's the best time to send an abandoned cart email?
It depends on how your customers buy. Impulse purchases like fashion usually reward a faster send, while considered purchases may need a day. The only reliable way to know is to test it: run 2 hours, 4 hours and 1 day against each other with everything else held the same, and measure orders and time to purchase rather than opens.
Should you offer a discount in an abandoned cart email?
It can work, but not as the opening move and never on its own. Many carts are abandoned because of a question or uncertainty, not price. When a discount does appear, pair it with the product's features, its benefits and what the customer gets by buying now.
Should new subscribers in a welcome flow also get abandoned cart emails?
New subscribers are still getting to know the brand, so they abandon carts more often. Whether the cart flow helps or interrupts the welcome sequence varies by brand. A/B test it: include them in one arm, exclude them in the other, and see which produces more first orders.
How do you A/B test an abandoned cart flow?
Change one variable at a time - the send delay, who receives the flow, or the content - and keep everything else identical. Measure orders placed from the flow and time to purchase, and let each arm collect enough abandoned carts before deciding.
Do abandoned cart emails work differently for fashion brands?
Yes. Fashion is largely impulse buying, so the email's job is to bring the impulse back quickly with a strong product visual and a clear reason to buy now. Considered categories need the opposite: answers to the customer's questions before any offer.
