Measurement

How to reduce cart abandonment: fix the leak you actually have

Cart abandonment is three separate leaks. Measure each stage in Shopify, fix costs and the account wall first, shorten checkout, then recover.

A miniature shopping cart on a yellow surface with sugar cubes spilling out of the basket.
Photo: Sergey Meshkov / Pexels

Seventy percent of online shopping carts are abandoned. That number is quoted everywhere, and on its own it is close to useless: it tells you that carts leak, not where, why, or what to fix first. Two stores with the same rate can need opposite changes.

This guide splits abandonment into the three stages where it happens, shows how to measure each one in Shopify, and orders the fixes by what shoppers actually leave over: late costs, forced accounts and a long checkout first, recovery messages last. It ends with a four-week order of work.

Cart abandonment is three different leaks

A cart is abandoned when a shopper adds an item and leaves without paying. Shopify counts it at checkout: the Abandoned checkouts report lists checkouts that were started and never completed. That is a narrower definition than most articles use, and the difference matters.

Between the product page and the order confirmation there are three exits: product page to cart, cart to checkout, and checkout to payment. Each has its own rate and its own causes.

The average rate hides the reason

Baymard Institute, which aggregates published studies, puts the average documented cart abandonment rate at 70.22% across 50 studies, updated in October 2026. When it asks shoppers why they left, the answers cluster into a few causes, and only one of them is price.

  • 40% said extra costs, such as shipping, tax or fees, were too high.
  • 19% did not trust the site with their card details.
  • 18% were asked to create an account when they did not want one.
  • 17% found checkout too long or too complicated.
  • 12% could not see the total order cost upfront.

Read that list again. Four of the five largest reasons are about information and steps, not about the product being too expensive. That is why a discount is usually the wrong first move: it pays shoppers to overlook problems you could remove for free.

Step 1: Measure each stage separately

Shopify's Conversion funnel report, under Analytics, shows sessions, sessions with add to cart, sessions that reached checkout and sessions that converted. Read it weekly and by device, then calculate three numbers:

  • Add-to-cart rate: sessions with add to cart divided by all sessions. A low number is a product page or traffic problem, not a cart problem.
  • Cart-to-checkout rate: sessions reaching checkout divided by sessions with add to cart. A low number is where cost, trust or an unanswered question stops the shopper.
  • Checkout completion: orders divided by checkouts started. A low number is a checkout form, payment or shipping problem.
  • Repeated support questions: the same question arriving at the cart is a page that needs an answer, and it usually shows up before the metric does.

Compare four full weeks, not two days. Small stores get noisy fast, and a campaign, a sale or a holiday can move every number at once. Segment mobile from desktop before you conclude anything: the same store often has two different leaks.

Step 2: Fix costs and the account wall first

These changes are cheap, reversible and aimed at the largest documented reasons. Do them before you touch design.

  • Show the shipping cost, or the threshold for free shipping, in the cart. A progress bar to free shipping also gives the shopper a reason to add one more item instead of leaving.
  • Turn on guest checkout and leave customer accounts optional or off. Accounts can be offered after the order, when the shopper has already decided.
  • Put a delivery estimate next to the total in plain language: a date range, not a processing-time formula.
  • Show a total that includes tax and shipping before checkout, or an honest estimate of it. “Calculated at checkout” is a surprise waiting to happen.

None of this needs a redesign, and all of it addresses the 40%, 18% and 12% reasons above. Measure the stage rates again two weeks later.

Step 3: Shorten the checkout itself

On Shopify the checkout is separate from your theme, so most of it is configured rather than designed. The goal is fewer decisions and fewer keystrokes on a phone.

  • Put express payment buttons, such as Shop Pay and the wallets your customers use, at the top of checkout and on the cart page. Returning shoppers should not retype an address.
  • Make optional fields optional. Company name, address line two and a phone number are the usual suspects when a form feels longer than the order.
  • Test autofill and address validation on a real phone, with one hand. If the keyboard type is wrong for a field, the checkout is slower than it looks on a desktop.
  • Show errors inline, next to the field, before the shopper presses pay. A generic error at the bottom of the page sends people back to hunt.

Then place one real order on a phone, on mobile data, as a first-time customer. Every extra tap you notice is a tap a shopper may not make.

Step 4: Answer the question that ends the visit

Many cart exits are not decisions against the product. They are a question the page does not answer: will this fit, does it work with what I own, when will it arrive, can I return it. A shopper with an open question rarely fills in a contact form to ask it. They leave.

A BuyScout® AI Store Agent is trained on a store's products and policies and answers inside the cart and on the product page, using page, browsing, cart and conversation context. It is a free Shopify app, set up in about a minute.

  • Answer the questions that stop a purchase where they are asked: size and fit, compatibility, specifications, delivery and returns.
  • Recommend the item that completes the order. If a shopper adds a camera, the agent can offer the tripod most buyers choose, as a bundle at a saving.
  • Explain policies in plain language instead of linking to a page the shopper has to read at the worst moment.

BuyScout reports that stores using the agent typically lift conversion 10–30%, grow average order value 15–25% and cut repetitive support tickets 41%; those are the vendor's own figures, not guarantees. The agent is free forever and GDPR and AI Act compliant, and it is meant to work alongside your other fixes, not replace them.

Step 5: Recover the carts that left

Recovery is the cheapest sale you can make, because the shopper already chose the product. It works best as a short sequence, not a single reminder.

  • After about an hour, send a plain reminder with the cart contents and a one-click link back. No discount yet.
  • After a day, answer the most likely objection: shipping cost and time, the returns policy, or stock and size. A short answer beats a bigger coupon.
  • After three days, offer an incentive if your margin allows it, and make it expire. Leading with the discount trains shoppers to wait for the next one.

Shopify can send one automated recovery email; set the delay and edit the copy under Settings, then Checkout. For a sequence, an email or SMS app is the simpler route. Exclude shoppers who already bought, and measure recovered revenue per send rather than open rate.

A four-week order of work

Change one thing at a time, and give each change two full weeks before judging it.

  1. Week 1: measure the three stage rates by device, and list the top five questions your support inbox receives at the cart.
  2. Week 2: fix the costs and the account wall: shipping visibility, guest checkout, delivery estimate, total upfront.
  3. Week 3: shorten checkout: express payments, optional fields, error messages, a real phone test.
  4. Week 4: launch the recovery sequence and put an answer in the cart for the five questions from week 1.

What to watch when you report

Report the stage rate you changed, not the blended abandonment rate. A cart-page fix will not move checkout completion, and a checkout fix will not change add-to-cart. Compare like with like: same device, same traffic source, same four-week window. If a change looks flat after two weeks, keep it if the reason is sound and the sample is small; a store with a few hundred sessions a week cannot detect a small lift, and it should not pretend to.

Two related guides go deeper: what a shopping cart is and how to make yours convert, and how to increase average order value.

Start with the measurement, not the redesign. Write down your three stage rates this week, fix the surprise costs and the account wall, and only then decide whether the cart needs a new design, a new app or a new discount. If you want the question answered inside the cart, BuyScout® AI Store Agent installs free from the Shopify App Store in about a minute.

Other articles

All articles

Try it Now

Make guided selling easier to evaluate

Free forever. Set up in 1 minute.