Abandoned cart recovery on Shopify

Abandoned cart recovery on Shopify is the practice of bringing back shoppers who added items and left without buying. Shopify's built-in abandoned checkout email only reaches shoppers who reached checkout and entered an email address. Everyone who left before that point is invisible to it — and on most stores, that is the larger group.

This is the distinction that decides whether a recovery programme works: an abandoned checkout and an abandoned cart are different events with different sizes and different fixes. Shopify's native tooling addresses the smaller one. Most advice about send timing addresses the smaller one too.

What is the difference between an abandoned cart and an abandoned checkout?

An abandoned checkout is a session where the shopper began the checkout flow and gave you an email address, then left before paying. Shopify creates a record for it, and can email that address automatically. An abandoned cart is a session where the shopper added something and left at any earlier point — from the product page, the cart drawer, the cart page — usually without ever identifying themselves.

The consequence is blunt: you cannot email an anonymous shopper. No app changes that. What an app can change is how many shoppers are still anonymous by the time they leave.

Why do most recovery programmes plateau?

Because they start at the last step. The usual sequence is: install a recovery app, configure a three-email sequence, tune the send delays, then argue about discount size. All of that operates on the population Shopify already had an email address for. It cannot grow that population, so the programme improves for a month and then flattens, and the reported recovery rate becomes a number nobody can move.

The higher-leverage work sits earlier, in this order:

How do you capture an email without a popup on arrival?

By asking at a moment when the shopper wants something in return. An entry popup asks a stranger for their address before you have given them a reason; it converts poorly and trains people to dismiss overlays. The alternatives all trade something of value:

• Back-in-stock alerts. The shopper wants the product and it is unavailable. Wanting the notification is the reason to give an address, which is why this converts far better than an unprompted ask.
• Save this cart / email me my cart. Framed as a convenience for the shopper rather than a capture for you, and genuinely useful on a cross-device purchase.
• Price or restock watch on a wishlist. The same trade, extended over a longer horizon.
• Accelerated checkout. Shop Pay, and wallet buttons generally, identify the shopper earlier in the flow — a capture mechanism most stores do not think of as one.

Each of these moves shoppers from the anonymous side of the line to the reachable side before they abandon, which is the only thing that grows the population a recovery sequence can work on.

What actually causes checkout abandonment?

Before optimising recovery, it is worth removing the reasons people leave. The recurring causes are unglamorous and mostly fixable:

• Cost revealed late. Shipping, duties or taxes appearing at the final step reframes the price the shopper agreed to. Showing them earlier converts better than discounting later.
• A forced account. Requiring registration before purchase adds a step with no benefit to the buyer.
• A slow or broken step. Particularly on mobile, and particularly on the address and payment steps. Worth measuring rather than assuming.
• Payment method missing. The shopper's preferred method is not offered — market-specific and easy to miss from head office.
• Delivery uncertainty. No date, or a date that arrives too late to matter.

Every one of these produces an abandonment that a recovery email then tries to reverse with a discount. Fixing the cause is cheaper and permanent.

How should the email sequence be structured?

Once the first two levers are in place, sequencing is worth doing properly. The durable principles, rather than specific hour counts that vary by store and category:

1. Send the first message while intent is still live. Hours, not days. Its job is a reminder, not a pitch.
2. Do not discount in the first message. A meaningful share of abandoners return without any incentive; discounting immediately pays them to do what they were going to do anyway.
3. Make the second message useful rather than louder. Answer the objection — delivery time, returns policy, sizing — instead of repeating the reminder.
4. Reserve any incentive for last, and treat it as the price of the sale rather than a lever to pull harder.
5. Stop. A sequence with no end converts nobody and costs deliverability.

Test the intervals against your own data. Category matters enormously: a considered purchase and an impulse purchase have different abandonment half-lives, and a sequence copied from a case study in another vertical is a guess wearing a number.

What should you measure?

Recovery rate alone is misleading, because it is calculated against the reachable population and therefore improves when that population shrinks. Track three things together:

• Identified rate — what share of abandoning sessions you can contact at all. This is the number the first lever moves, and the one most stores never look at.
• Recovery rate — conversion within that reachable group.
• Incremental revenue — recovered revenue minus what the discount cost and minus what would have returned anyway. The only one of the three that is unambiguously a business result.

What to do next

Find your identified rate. If you cannot calculate it, that is the finding: you are optimising a sequence without knowing what fraction of the opportunity it can reach. Then fix the largest abandonment cause you can name from your own checkout data before touching send delays.

We build RecoverlyAI for cart and checkout recovery, and Wishlist+ for the back-in-stock and save-for-later capture that grows the reachable population in the first place. If you would rather fix the checkout than recover from it, that is CRO work, and it is usually the better first spend.