Marketing & retention

People abandon food carts too: recovering 10–15% of lost orders

Someone built a cart at your restaurant tonight and didn't pay. That happens far more than most owners think — and a decent slice of it is recoverable within the hour.

Cart abandonment in e-commerce has hovered around 70% for years, and food ordering gets no exemption. People add the pad see ew, reach the checkout, and vanish. Say your ordering site collects 200 carts a week and turns 60 of them into orders. That leaves 140 carts, each one a person who wanted your food and left without it. Win back even 15 of those at a $30 ticket and you've added about $450 a week, all from people who were already on the page.

Food carts die for different reasons than shopping carts

Nobody browses ramen the way they browse sneakers, so the classic "just researching" abandonment barely applies. Food carts die for specific reasons, and most of them are fixable. The biggest is fee shock: a $24 cart turns into $33.80 once delivery, service fee, and tax land at the checkout, and it tops the list in every study that bothers to ask. Then there's forced registration. Ask a hungry person to create an account and verify an email, and they'll often just open a marketplace app instead. Checkout friction does its own damage, because retyping a card number with wet hands in a kitchen kills real orders. So does time panic, when a 55-minute estimate only shows up at the very end. And then the group chat: whoever built the cart is waiting on two friends to confirm, gets distracted, and dinner turns into cereal.

The recovery window is about an hour

Retail sends its cart emails the next morning. For food that's useless, because the person already ate. What you're racing is hunger, and hunger fades in roughly an hour. So the recovery sequence has to be fast and built for the phone, not the inbox.

Start with a push, 25 to 45 minutes in, for anyone on the app: “Your cart's still here, and the truffle pasta is patient. Weren't you hungry?” One tap brings the cart back exactly as it was. Skip the discount here; at this point most people just needed the reminder.

If the customer is known but doesn't have the app, follow at 60 to 90 minutes with an SMS: a short line and a link that reopens the same cart. Texts get opened around 98% of the time, which makes SMS the right rescue channel for the website crowd. The mechanics are in our SMS guide.

Only if the first two go nowhere do you send one last message, the next day around 11:30, and only now do you attach a small incentive: “Yesterday's cart, today's lunch? Free delivery on it.” Then stop. Three touches per cart is the ceiling; go past it and you're teaching people to tune you out.

One rule worth holding to: no discounts in the first two messages. If a coupon reliably shows up whenever someone abandons a cart, your sharpest regulars will start abandoning on purpose, and retailers learned that lesson the expensive way. A plain reminder is enough to bring back the people who were simply distracted, and that's most of them. The sweetener belongs in the next-day message, when the original hunger is gone and the order actually needs a reason to happen.

Carts built
140/wk
Lost without recovery
140
Recovered at 10%
14 ≈ $420/wk
Recovered at 15%
21 ≈ $630/wk
Same restaurant as the intro: 140 abandoned carts a week at a $30 average ticket. Those recovery percentages add up week over week.
How many carts died on your site this week?

Dots tracks abandonment and runs the recovery sequence automatically — we'll show you the funnel live.

Book a demo

The best recovery is a checkout that doesn't leak in the first place

Every cart you recover is really a patch on friction that shouldn't have been there. The prevention list is short. Put delivery fees and the time estimate on the menu page instead of the final screen, because people bail on the surprise far more than on the fee. Let customers check out as guests with just a phone number, and let the account get created automatically from the order. Save cards and addresses so a repeat order takes three taps. Keep the whole thing on one screen. That's the gap between an ordering site that converts at 3% and one that hits 15%, which we pulled apart in why most restaurant websites convert at 3%; it's the standard we build our ordering websites to.

What to measure so this doesn't become folklore

Track four numbers in your analytics every week: carts built, carts paid, recovery messages sent, and orders that recovery brought back within 24 hours. The ratio that matters is recovered-to-abandoned. Once the sequence runs on its own, 10–15% is a realistic steady state; anything under 5% usually means the messages fire too late or the cart link is dropping items when it restores. Test it yourself once a month. Build a cart at 18:00, walk away, and time how long it takes your own phone to buzz. If it never does, you've found the leak, and it's costing you every evening you don't fix it.

All of that (the tracking, the push, the SMS, the next-day message) runs as triggered automation inside the platform's marketing tools. You set it up once and leave it. Across the 3M+ orders Dots has processed, triggered messages like these beat broadcast campaigns pretty reliably, because they react to something the customer just did.

One thing to do this week: find your abandonment number. Most owners have never looked at it. It's the cheapest revenue you'll recover all year, because the customer already chose the food.

Stop losing orders at the last step

A checkout built to convert plus automatic cart recovery — live on your brand in about two weeks.

Book a demo

Related reading

👋 Let’s get acquainted!

Tell us about your restaurant and we’ll show you the platform on your own numbers.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.