Commerce operations
Abandoned checkout
An abandoned checkout is a session where a shopper entered the checkout flow and provided identifying information but did not complete the purchase.
It is distinct from cart abandonment, and the distinction matters commercially: an abandoned checkout has contact details and a known intent, so it is directly recoverable. An abandoned cart usually has neither.
Recovery performance depends far more on timing and on the checkout link resuming the exact cart state than on the copy. A recovery email that drops the customer at a generic cart page recovers a fraction of what a resumable link does.
Four different events, routinely reported as one
Abandonment is not a single thing. It is a set of exits at different depths of the funnel, and what you know about the shopper — and therefore what you can do — changes completely at each step.
The commercially important line is the one where contact details appear. Above it you are doing advertising; below it you are doing direct outreach to a named person who told you what they wanted.
| Event | What the shopper did | What you know | Recovery route |
|---|---|---|---|
| Browse abandonment | Viewed products, never added to cart | A session, and nothing else unless they are already a known customer | Retargeting, or an on-site prompt before they leave |
| Cart abandonment | Added to cart, left before entering checkout | Cart contents tied to a session or cookie, usually no contact details | Only if the shopper is already identified from a prior visit or login |
| Checkout abandonment | Entered checkout and supplied an email, phone or address | Contact details, the full cart, and often the shipping destination | Direct email or SMS carrying a link that restores the exact cart |
| Payment failure | Submitted payment and the authorisation was declined | Everything above, plus the issuer decline reason | A payment fix, not persuasion — a new method or a retry, guided by the decline code |
Where the marker is set decides your abandonment rate
There is no standard moment at which a checkout becomes an abandoned checkout. Some systems create the record on the first keystroke in the email field, some when the shipping step is reached, some only once a payment method has been attached. Each choice produces a different denominator, so abandonment rates are not comparable across platforms and often not comparable across a replatforming.
Two artifacts inflate the count in almost every setup. The first is the same shopper appearing more than once: a session that starts on a phone and finishes on a laptop leaves an abandoned checkout behind and a completed order elsewhere, and unless records are deduplicated by customer and cart contents, both are counted. The second is automated traffic — crawlers and form-fill bots reaching the email step create records that no message will ever recover.
Suppression is the practical consequence. Before sending anything, check whether an order exists for that customer with that cart, in any session, and drop the record if it does. Emailing someone about the order they just placed is the single most damaging failure mode in this whole workflow, because it reads as either incompetence or surveillance.
Why shoppers abandon, split by what you can control
Ranking causes by frequency is less useful than ranking them by whether anything can be done. A large share of abandonment comes from people who were never going to buy on that visit — price checking, saving for later, showing something to a partner — and no sequence of emails changes that.
This is also why abandonment rate is a weak KPI. It is dominated by intent you cannot influence. Recovered revenue is the metric worth reporting, because it counts only the part the operation actually moved.
- Cost revealed late. Shipping, tax or import duty appearing for the first time at the final step is the most fixable cause on this list: show the total early, or offer a threshold you can afford.
- Forced account creation. A registration wall between a shopper and a purchase converts a buyer into a task. Guest checkout with an optional account after payment removes it.
- Length and friction. Every additional field and step is an exit opportunity. Address autocomplete, validation that corrects rather than rejects, and a single-page flow all reduce it.
- Missing payment method. A shopper who expects a wallet or a local method and does not find it does not switch to a card; they leave. This shows as abandonment and is really a payments coverage gap.
- Declined payment. Filed as abandonment by most reporting, but it is an authorisation problem. It needs the decline code, not a discount.
- Trust on a first purchase. An unfamiliar brand asking for card details benefits from visible return terms, contact details and recognisable payment marks — real ones, at the moment of the ask.
- Not actually in market. Comparison shopping, browsing, price anchoring. Unrecoverable, and the honest reason abandonment rate will never approach zero.
Recovery mechanics: the link does most of the work
The single largest determinant of recovery is what happens when the shopper taps the link. A link that restores the exact cart — variant, quantity, applied discount, chosen shipping method and entered address — and drops them at the payment step recovers far more than one that lands on a generic cart page and asks them to start again. Everything about the copy is second-order next to that.
The second determinant is that the link works without a login. A shopper who abandoned because checkout was too much work will not create an account to finish it.
Timing follows a shape rather than a benchmark: one message soon enough that the intent is still live, a second the following day, and a third a day or two after that, then stop. Test the intervals against your own data instead of adopting someone else's, because the right spacing depends on considered-purchase length, which varies enormously by category.
- Do not put a discount in the first message. A brand that reliably discounts abandoned checkouts teaches its customers to abandon, and the cost of that lesson is paid on every future order, including the ones that would have converted at full price.
- If a discount is used at all, put it last in the sequence and cap it. An unconditional recovery discount that never expires becomes a permanent price cut for the shoppers who notice.
- Branch on the reason. A checkout abandoned at the payment step after a decline gets a payment-method update path; a checkout abandoned at the shipping step gets clarity on delivery cost and timing.
- Respect consent and channel rules. Contact details entered to complete a purchase are not blanket permission to market, and SMS in particular is governed by consent rules that vary by jurisdiction.
- Suppress on completion, across sessions and devices, before every send.
- Report recovered revenue net of the discounts granted to recover it. A sequence that recovers 10% of abandoned value by giving away 20% has not necessarily made money.
Frequently asked questions
- What is an abandoned checkout?
- An abandoned checkout is a session in which a shopper entered the checkout flow, supplied identifying information such as an email address or phone number, and then left without completing the purchase. Because contact details and the full cart are known, it is directly recoverable by sending the shopper a link that restores their cart and returns them to the payment step.
- What is the difference between an abandoned cart and an abandoned checkout?
- An abandoned cart is a shopper who added items and left before entering checkout, so you usually have the cart contents but no way to contact them. An abandoned checkout is a shopper who got as far as entering an email, phone or address. The second is directly recoverable and the first generally is not, which is why the two should never be reported as one number.
- Why do customers abandon checkout?
- The causes an operator can act on are unexpected shipping, tax or duty appearing late, forced account creation, a long or error-prone form, a missing payment method the shopper expected, and declined authorisations recorded as abandonment. The causes an operator cannot act on are comparison shopping and browsing without purchase intent, which is why abandonment rate never approaches zero.
- How long should you wait before sending an abandoned checkout email?
- Send the first message while the intent is still live, typically within the first hour, then a second the following day and a third a day or two later before stopping. Treat this as a starting shape rather than a benchmark: the right spacing depends on how long your category takes to consider, so test intervals against your own recovery data.
- Do abandoned checkout emails work?
- They recover real revenue, but performance is dominated by mechanics rather than copy. A message whose link restores the exact cart, keeps the applied discount and shipping selection, and reaches the payment step without a login recovers substantially more than one that returns the shopper to a generic cart page. Timing is the second factor; wording is a distant third.
- Should abandoned checkout emails include a discount?
- Not in the first message. A brand that discounts every abandoned checkout trains shoppers to abandon deliberately, and that behaviour then applies to orders that would have converted at full price. If a discount is used, place it at the end of the sequence, cap it, expire it, and report recovered revenue net of what was given away to recover it.