Abandoned Cart Emails That Work: Timing, Copy and the Mistakes That Cost Orders

Most carts are abandoned for reasons that have nothing to do with wanting the product less. Shipping cost appeared at the last step. The checkout asked for an account. The phone rang. The person was comparing two stores and yours was the tab they left open. A shopper who got as far as the cart has told you more about their intent than any ad click ever will, and an email that arrives while the intent is still warm is the cheapest sale in ecommerce.

The mechanics are simple. Your store sends an abandoned checkout event when a shopper with a known email address starts checkout and does not finish. That event starts a flow, the flow sends an email, waits, and sends another. A goal of placed order lets a person out the moment they buy, so nobody is nudged about a cart they already paid for. In Brainito the ready-made Abandoned Checkout Recovery flow does exactly this: one email on the spot, a 20-hour wait, a second email, exit.

What follows is how to make those two emails earn their place, and how to avoid the three mistakes that turn cart recovery into an annoyance.

The first email is a helpful nudge, not a sales pitch. The shopper may have been interrupted, may have hit a problem, or may simply need to see the product again. Send it within the first hour; the ready-made flow sends it immediately, and anything past a few hours loses the moment.

• Subject: plain and specific. "You left something in your cart" or "Still thinking about the [product name]?" Avoid urgency and avoid exclamation marks; the person has not done anything wrong.
• Body: the product image, name and price, pulled from the event, with one button back to the cart. One sentence that says you saved it for them.
• Remove the obstacle: a line about shipping, returns or how to reach a human. If checkout problems are common, say "if something went wrong, reply to this email" and mean it.
• No discount yet. Offering money off in the first email trains repeat customers to abandon carts on purpose. Many people just need the link.

About 20 hours later, if there is still no order, the second email goes out. By now the person has either decided against it or is on the fence, and the fence is where a small push helps. The template's hint for this email is a modest offer, and a modest offer is right: free shipping or a small percentage, with a deadline measured in days, not hours.

If you would rather not discount, give a different reason to decide: a review or two of the exact product, a note on how many are left if stock is genuinely low, or an answer to the objection you hear most. What matters is that the second email adds something the first did not. Repeating the first email with a louder subject line is what people mean when they complain about cart emails.

Keep the exit rule honest. The goal node checks for placed order, so if they bought between the two emails, the second never sends. If they did not buy after the second, let them go. A third and fourth email recover very little and cost you goodwill with the people who had already decided.

For shoppers who gave SMS consent at checkout, a short text the next morning often does more than a second email, because it is read within minutes and the link opens straight to the cart on the phone they are holding. In the flow builder you can replace the second email with an SMS step, or branch: email the people who opened the first message, text the people who did not.

The rules are stricter for texting. The step must check SMS consent, the number must be registered for the destination country, and in the US and Canada the message has to land inside quiet hours, 8am to 9pm in the recipient's local time. Brainito enforces all three: a contact without SMS consent is skipped with the reason recorded, an unregistered number cannot send, and a text scheduled at midnight waits for the morning. Registration and consent for texting are covered on the SMS marketing page.

• Emailing people who already bought. This happens when the flow has no goal, or when the store's order event is not connected. Nothing erodes trust faster than "you forgot something" an hour after the receipt. Connect the store, set the goal, test it with a real purchase.
• Sending to people who did not consent. A checkout email address is not automatically marketing consent in every jurisdiction. Cart recovery for an existing customer is widely treated as a transactional follow-up; for a first-time visitor, your checkout should state that you will email about their cart. Check the rules where you sell.
• Discounting every cart. Shoppers learn fast. If the second email always carries a code, the cart becomes a coupon dispenser and your margin pays for it. Reserve the offer for first-time buyers, or alternate it with social proof and plain reminders.

A fourth, smaller one: a cart email that shows the wrong product or a blank image because the event did not carry the data. Send yourself a test with a real cart before you turn the flow on.

Each email in the flow costs 25 credits per 1,000, the same as a campaign, so recovering 1,000 abandoned carts costs at most 50 credits and usually less, because the goal removes buyers before the second email. With the store connected, Brainito attributes each order to the flow email that earned it, so the report shows carts entered, emails sent, orders and revenue side by side.

Read it monthly. If the first email recovers plenty and the second recovers almost nothing, drop the discount and test a plain reminder. If neither recovers much, the problem is usually upstream: shipping costs revealed late, a checkout that demands an account, or a payment method missing. The automation platform page lists what the flow builder can do, and if you are weighing Brainito against the ecommerce tool most stores start with, the Klaviyo comparison goes through the differences in pricing and channels.