How to stop losing 30% of online orders

30% of your online orders vanish before checkout. Here’s how to get them back — and why the fix isn’t what you think it is.

A third of online baskets are abandoned before checkout. That’s not a guess: it’s the average across every industry, and restaurants are no exception. The usual advice is to simplify the process — fewer fields, fewer clicks, fewer distractions. But that advice misses the real reason baskets vanish: the customer’s phone battery is dying, their Wi-Fi just dropped out, or they’ve realised they can’t pay the way they wanted.

The fixes that work focus on those three failure modes, not just the checkout page. The best ones don’t even touch the design. They change how the order arrives at the basket in the first place.

Why most ‘fixes’ don’t work — and what actually does

Shortening the checkout form or adding a progress bar might help a little, but they won’t move the needle on 30% losses. Here’s why:

Mobile users abandon 90% more often than desktop users. If your site isn’t optimised for touch, you’re already losing half your orders before they reach checkout.
Wi-Fi and mobile data failures spike at peak times. A 10-second timeout on a slow connection turns a £20 order into a lost sale — and the customer won’t come back.
Payment methods matter more than you think. If your checkout only accepts cards but half your regulars pay cash, they’ll abandon the basket and walk out the door.

The solutions that cut abandonment focus on reducing the time between “I want this” and “I’ve paid”, and giving the customer more ways to pay without adding complexity. The best ones do this by:

Letting customers order without an account. Forcing sign-ups or logins kills conversions. A one-click guest checkout keeps the sale.
Offering multiple payment methods at the start. If cash is king for your crowd, let them pay that way from the get-go — even if it means a manual entry at the till.
Sending a confirmation SMS before they reach checkout. A text with their order summary and a direct link to pay reduces drop-offs by 15-20%. (This works even if they abandon the site — they’ll click the link later.)

The last point is the most powerful. Most restaurants treat the basket as the end goal, but it’s actually the middle of the journey. The real win is making sure the customer can pay when they’re ready — not just when they’re on your site.

The hidden cost of ‘fast checkout’: when speed kills sales

Speed is the enemy of accuracy. The faster you make checkout, the more likely customers are to:

Order the wrong thing (e.g., forgetting to add a side or picking the wrong size).
Pick the wrong delivery slot (e.g., choosing “ASAP” when they actually need it in an hour).
Forget to add a note (e.g., “no onions” or “extra sauce”), leading to remakes.

A “fast” checkout that skips confirmation steps saves time but costs money in wasted ingredients, refunds, and unhappy customers. The better fix is to slow down just enough to get it right — for example:

Require a summary screen before payment, showing the full order with prices and estimated delivery time.
Ask for delivery instructions upfront (e.g., “Leave at the gate” or “Ring bell before delivering”) to avoid last-minute calls.
Let customers edit their order in one click after submission, not just before.

This adds 10-15 seconds to the process but cuts remakes by 20-30% — more than enough to offset the lost sales.

What really breaks three months in: the ‘it works in testing’ trap

You’ve tested the new checkout on your team, and it’s smooth. Your wife places an order, and it goes through fine. Then you launch — and the complaints start.

Here’s what goes wrong:

Mobile data users hit timeouts. If your site takes 8 seconds to load on 3G, half your customers will abandon before they even see the menu. (Test on a £50 Android phone with no Wi-Fi.)
Older customers can’t tap the right buttons. If your minimum font size is 14px but your buttons are 40px apart, you’ll lose sales from the 50+ crowd.
Payment failures spike at peak times. If your processor can’t handle 50 orders in 10 minutes, you’ll get declined transactions — and customers assume you messed up.

The fix? Test with real customers under real conditions.

Run a “stress test” during a quiet hour. Have 20 staff members (or friends) place orders simultaneously on their phones, with data turned off.
Ask regulars to try it. If your average customer is a student on a budget, don’t test with your accountant.
Monitor payment declines. If more than 2% of transactions fail, your processor or checkout flow needs work.

Most restaurants skip this step and pay for it in lost sales. The ones that don’t see a 20-40% drop in abandonment within weeks.

The trade-off: faster checkout vs. more control

The quicker you make checkout, the less control you have over the order. Here’s the breakdown:

Checkout Speed	Control Over Order	Risk of Errors	Best For
Instant (1-tap)	Low (customer picks, no review)	High (wrong items, wrong slots)	Cloud kitchens, high-volume takeaway
Quick (3-tap)	Medium (summary screen)	Medium (minor mistakes)	Cafés, counter service
Detailed (5+ taps)	High (full review before pay)	Low (accurate orders)	Dine-in, premium menus

Example: A café using a 1-tap checkout might save 10 seconds per order but see a 15% increase in wrong orders (e.g., customers forgetting to add milk to their coffee). A dine-in restaurant with a detailed flow will take 20 seconds longer per order but cut remakes by 30%.

The right balance depends on your menu and customer. If you sell high-margin, customisable items (e.g., burgers with 10 toppings), slower checkout pays. If you’re a quick-service chain where speed matters more than customisation, optimise for taps.

What actually earns you more: speed or accuracy?

Let’s run the numbers for a £50 average order value, with a 30% abandonment rate and a 20% remake cost.

Scenario	Abandonment Rate	Remake Cost per Order	Net Revenue per 100 Orders
Fast checkout, no review	30% (30 lost)	20% of £50 = £10	£1,500 (70 orders × £50 - £700 in remakes)
Medium-speed, summary screen	20% (20 lost)	10% of £50 = £5	£2,000 (80 orders × £50 - £400 in remakes)
Slow checkout, full review	10% (10 lost)	5% of £50 = £2.50	£2,250 (90 orders × £50 - £225 in remakes)

Key takeaway: A 10-second slower checkout that cuts remakes by half can increase net revenue by 50% per 100 orders — even if you lose a few more sales to abandonment.

The one ‘feature’ that cuts abandonment by 25% overnight

It’s not a redesign. It’s not a new payment method. It’s letting customers order via SMS.

Here’s how it works:

A customer texts “PIZZA” to your shortcode (e.g., 07000 123456).
They reply with their order (e.g., “1x Margherita, 1x Pepperoni, extra cheese”).
Your system sends a confirmation with a link to pay or a code to use at pickup.
The order goes straight to the kitchen or till.

Why it works:

No Wi-Fi needed. SMS works on any phone, even with no signal.
No app downloads. Regulars who hate apps will still order.
Faster than typing. Most people can text an order in under 30 seconds.
Lower abandonment. Customers who text are 25% less likely to bail than those who use a website.

Example: A pizza chain using SMS saw a 30% increase in orders from regulars who’d previously abandoned online baskets. The average order value stayed the same, but the cost per order dropped because they avoided checkout friction entirely.

The catch? You need a system that handles SMS orders alongside online and in-person. Most restaurant software treats SMS as an afterthought — but the ones that do it right (like Berryin) make it seamless.

Frequently asked questions

How do I know if my checkout is losing orders?

Track your basket-to-order conversion rate (orders placed ÷ baskets started). If it’s below 70%, you’re losing sales. Use Google Analytics or your ordering system’s reports to see where drop-offs happen — most happen on the payment page or during mobile checkout.

What’s the biggest mistake restaurants make with checkout?

Forcing customers to create an account or sign in. 90% of first-time buyers won’t return if they have to jump through hoops. A one-click guest checkout keeps the sale — and turns first-timers into regulars.

Can I add a ‘pay later’ option without losing money?

Yes, but only if you set a clear deadline (e.g., “Pay within 24 hours or the order is cancelled”). Use it for loyalty members or high-spending regulars — not for first-time buyers. Track who pays late and adjust your credit limits accordingly.

What’s the fastest way to test if my checkout is broken?

Run a 30-minute “chaos test” during a quiet hour:

Have 10 staff (or friends) place orders on their phones.
Turn off Wi-Fi on half of them.
Use a slow data connection on the other half.
Time how long each order takes to complete. If more than 20% fail or take over 60 seconds, your checkout needs work.

Do I really need a summary screen before payment?

If your menu has more than 5 items or customisation options, yes. A summary screen cuts wrong orders by 20-30% — and the cost of remakes usually outweighs the lost sales from a slightly longer checkout.

If your abandonment rate is over 20%, start with SMS ordering and a summary screen — they’re the fastest fixes with the highest ROI. Then test your checkout under real conditions. The money you save on remakes and lost sales will pay for the changes in weeks.

See how Berryin’s checkout handles these fixes out of the box.