Your add-to-cart rate looks healthy. Your order count doesn't. Somewhere between the two, cart abandonment is eating the difference, and the frustrating part is that the number tells you nothing about which of the fifteen things in your checkout caused it. This piece starts after the shopper has already added to cart, and it's written for Canadian stores specifically, because shipping, duty, currency and payment habits create failure points that American checkout advice doesn't cover.
What is cart abandonment, and what counts as a normal rate?
Cart abandonment is what happens when a shopper adds an item and leaves without paying. Most of it comes from cost surprises and checkout friction, not from changing their mind. The top three causes are shipping and fees revealed too late, forced account creation, and a checkout that asks for too much.
The formula is simple: cart abandonment rate equals one minus completed purchases divided by carts created, expressed as a percentage. So 300 carts and 90 orders is a 70 per cent abandonment rate.
That 70 figure isn't a coincidence. Baymard Institute, which aggregates dozens of published studies rather than running one survey, puts the documented average at roughly 70 per cent. Before you panic at your own number, though, two caveats. Much of that abandonment is people comparing prices and window shopping, which no design change will recover. And cart abandonment isn't the same measurement as checkout abandonment.
| Metric | What it measures |
|---|---|
| Cart abandonment | Everyone who added an item and didn't buy, including browsers who never intended to. |
| Checkout abandonment | Only those who started the checkout process and dropped out. This is the number you can actually fix. |
Store owners conflate the two constantly, then compare their cart number against someone else's checkout number and conclude they're broken. Track both, separately, before you change anything.
Why Canadian shoppers abandon carts
Here they are in rough order of impact, with the Baymard research doing the ordering where it has data and Canadian specifics filling the gaps it doesn't cover.
- Shipping cost revealed too late. Looks like: strong cart adds, sharp drop at the shipping step. Do: show full landed cost in the cart, or a clear estimate with an explanation.
- Duty, tax and cross-border surprises. Looks like: abandonment concentrated on US-fulfilled orders. Do: state duty and tax handling before checkout, not at the end.
- Prices shown in the wrong currency. Looks like: high bounce from Canadian traffic on product pages. Do: detect and display CAD, and say which currency you charge in.
- Forced account creation. Looks like: a drop at the login screen. Do: offer guest checkout, and make it the visually prominent option.
- Too many form fields and too many steps. Looks like: partial form completion, especially on mobile. Do: cut every field that isn't needed to ship and charge.
- Missing preferred payment methods. Looks like: abandonment at the payment step with no error. Do: add wallet payments and, for Canadian buyers, consider Interac options.
- A checkout that fails on a phone. Looks like: a large mobile-to-desktop conversion gap. Do: complete a purchase on your own phone, on cellular data, today.
- No visible trust, return or contact information. Looks like: hesitation at the final step, often high time-on-page before exit. Do: put returns policy and a contact route next to the pay button.
Three of those have hard numbers behind them. Baymard's checkout UX research found 39 per cent of users abandoned a checkout because of extra costs like shipping, taxes and fees, 21 per cent because delivery was too slow, and 19 per cent because they were forced to create an account. Those are self-reported reasons, which means they're better at identifying causes than at sizing them, but the ordering has been stable across years of the same research.
Two of these are commonly overstated. Trust badges, on their own, do very little; visible returns terms and a real contact route do considerably more. And "reduce the number of steps" is a proxy goal that gets chased for its own sake. A one-page checkout with thirty fields is worse than a three-step checkout with well-designed forms.
The Canadian specifics deserve their own note. Nearly four in five of the Canadian stores we've audited showed no landed cost anywhere before the shipping step, which is the single most fixable thing on this list. Roughly half were missing at least one payment method their own customers told us they expected. Duty and cross-border mechanics get their own treatment in cross-border ecommerce into the US, and payment infrastructure in recurring billing on Canadian websites.
“many sites are still including too many form fields in their checkouts”
Edward Scott, Research Lead, Baymard Institute
Checkout optimization: the fixes worth making first
Reordered as a work plan rather than a diagnosis. Effort is what it costs you, not what it costs an agency.
| Fix | Effort | Who does it |
|---|---|---|
| Turn on guest checkout | Minutes | You, in platform settings |
| Show shipping cost or a rate estimate in the cart | An hour | You, in shipping settings |
| Set currency to CAD for Canadian visitors | An hour | You, or your platform's market settings |
| Add wallet and local payment options | Half a day | You, plus your payment provider |
| State duty and tax handling before checkout | Half a day | You, writing the copy |
| Remove unnecessary form fields | A day | Developer, if fields are theme-level |
| Make guest checkout visually prominent | A day | Developer or theme editor |
| Put returns and contact info at the payment step | A day | Theme edit |
| Rebuild a checkout that fails on mobile | Two to three days | Developer or studio |
Work top to bottom. The first three cost you an afternoon and address the two causes with the largest documented share of abandonment. Anyone selling you a rebuild before you've done them is selling you a rebuild. Which platform you're on changes what's settings-level and what isn't, and Shopify versus Webflow for ecommerce covers where those lines fall.
What a low-friction Canadian checkout actually looks like
Walk your own checkout against this sequence. Where it diverges is where you're losing people.
- Cart. Item, price, shipping estimate and tax, all visible. The total the customer will pay, or an honest estimate with a note on how it's calculated.
- Guest checkout offered first. Not hidden below a login form. A clear "continue as guest" button, at least as prominent as signing in.
- Address entry with autocomplete. Postal code lookup where you can manage it. Province as a dropdown, not free text.
- Payment options above the fold. Card, wallets, and whatever else your customers use, visible without scrolling.
- One confirm action. One button, one label, nothing competing with it. No cross-sell decision between the customer and the purchase.
- Confirmation that answers the two questions. When will it arrive, and what happens if something's wrong. Both, on screen, before the email lands.
When we rebuilt GlowBrew's checkout around guest purchase and upfront landed cost, their checkout completion rate went from roughly 26 per cent to just over 34 per cent across the following quarter. Most of that came from the first two items on the list, not from anything clever.
Mobile deserves calling out separately. If your desktop checkout is fine and your mobile checkout isn't, the fixes are usually about tap targets, keyboard types and field count rather than layout, and mobile-first design covers the underlying approach.
Recovering the carts you still lose
Some abandonment is recoverable and most isn't, so keep this in proportion.
Reminder emails work best for distracted abandonment: someone who meant to buy and got interrupted. First message within a few hours, while the intent is warm. Whether you send one or a sequence depends on your margin and your tolerance for annoying people, and for a small Canadian store one well-written message usually beats three automated ones. On-site exit prompts can catch price-sensitive shoppers, but they're a discount mechanism dressed as UX.
The honest limit: recovery messaging doesn't fix UX-caused abandonment. Bringing someone back to a checkout that already failed them just exposes them to the same problem, this time with less patience. Fix the flow first, then recover what's left. We're deliberately not reviewing recovery tools here, because that's a different question with a different answer.
How to measure whether your checkout changes worked
Record four numbers before you touch anything: sessions, add-to-cart rate, checkout starts, and completed orders. The drop between each pair tells you where the loss sits, and you cannot interpret an after without a before.
Then separate cart drop-off from checkout drop-off from payment-step drop-off. They have different causes and different fixes. Compare mobile against desktop for each, since a gap there usually points straight at the problem.
On timing: give a change at least four full weeks, and preferably a full purchase cycle. Which raises the seasonal problem. October and November are terrible comparison windows because traffic composition changes underneath you, and holiday shoppers behave differently from your usual customers. If you're making changes now, compare this October against last October, not against September. Getting more of the right traffic into the funnel in the first place is a separate discipline, covered in ecommerce SEO for Canadian stores.
For context on the size of the prize: Statistics Canada reported that retail e-commerce accounted for 6.1 per cent of total Canadian retail trade in December 2025, up from 5.8 per cent in November. That's a narrower definition than the double-digit figures forecasters quote, since it counts only retailers classified under retail trade. It's the conservative number, and it's still growing.
Do it yourself or bring in a studio?
Most of the top of that fix table is genuinely yours. Guest checkout, shipping display, currency, payment methods: settings changes, an afternoon, no developer. Do them before you talk to anyone.
Theme and template work is the middle ground. Field removal, guest checkout prominence, trust information placement: doable if someone on your team is comfortable in the theme editor, worth outsourcing if they aren't, because a half-finished theme edit on a live checkout is an expensive kind of mistake.
Then there's the case where the checkout was built wrong. Signs: you can't change field order without breaking layout, mobile performance can't be recovered because the template is heavy by design, or every small change needs a developer. At that point you're patching around a structural problem, and a rebuild costs less over a year than the patches do. That's the conversation our ecommerce web design work and our web design service exist for. If the underlying question is whether you should be on your own store at all, your own store versus marketplaces is the earlier decision. And if friction is a site-wide problem rather than a checkout one, how design reduces bounce rate covers the pages before the cart.
Frequently Asked Questions
What is a good cart abandonment rate?
Baymard Institute puts the documented average at roughly 70 per cent, so anything meaningfully below that is decent. The more useful comparison is your own checkout abandonment rate over time, since cart abandonment includes browsers who never intended to buy.
Why is shopping cart abandonment a problem?
Because you've already paid to acquire that visitor and they've already told you they want the product. Abandonment at checkout is the most expensive place to lose someone. A meaningful share of it is caused by fixable design friction rather than genuine hesitation.
How to fix cart abandonment?
Start with the two causes that carry the most documented share: show the full landed cost in the cart, and offer a prominent guest checkout. Then cut unnecessary form fields, add the payment methods your customers expect, and test the whole flow on a phone.
Do abandoned cart emails actually work?
For distracted shoppers who meant to buy, yes, particularly a single message sent within a few hours. For abandonment caused by checkout friction, no. Returning someone to a checkout that already failed them tends to reproduce the failure.
Show the real cost early, let people buy without an account, ask for less, and accept the payment methods your customers already use. That's most of it. The rest is measuring properly and resisting the urge to judge a change during the one month of the year when your traffic doesn't look like itself.
We build and rebuild ecommerce stores for Canadian businesses, including the checkout flows that decide whether a full cart turns into an order. Worth sorting before the holiday run.
