How to Build a High-Converting Klaviyo Abandoned Cart Flow

An abandoned cart flow should recover purchase intent without immediately training customers to wait for discounts.

The strategy is simple: remind first, remove friction second, escalate only when necessary.

Most underperforming cart flows fail in one of two ways. They either send a single reminder and stop, or they lead with a discount and permanently reset what customers are willing to pay.

Cart Abandonment vs Checkout Abandonment

Cart abandonment starts when someone adds a product to their cart and leaves. Checkout abandonment starts when someone begins the checkout process and leaves.

Checkout generally signals higher intent, because the customer has already committed to the purchase decision and started handing over details.

Established brands should usually separate the two. They can begin from a similar content framework, but the triggers and exclusions have to be correct so a customer never receives both journeys for the same session. As testing produces data, the two journeys can diverge in tone, timing and incentive logic.

Recommended Klaviyo Trigger and Filters

The cart flow triggers on the add-to-cart event, and the checkout flow triggers on started checkout.

For the cart flow, useful exclusions include:

• placed an order since entering the flow;
• started checkout since entering the flow, so high-intent customers move to the checkout journey instead;
• a sensible cooldown so repeat browsers do not re-enter constantly.

For the checkout flow, remove the started-checkout exclusion. It conflicts directly with the trigger and will stop the flow from sending.

For Shopify stores, verify which event your setup is actually receiving before building conditions on top of it. Cart and checkout events can be delivered by different integrations, and a flow built on the wrong event looks correct while quietly under-triggering.

How Long Should You Wait?

For high-intent cart and checkout events, roughly 15 minutes to 1 hour is a reasonable range to test for the first message.

Thirty minutes is a common starting point, not a universally optimal delay. The closer the event sits to the purchase decision, the more a faster follow-up tends to make sense, because you are catching a live session rather than reopening a closed one.

Test the delay directly. It is one of the few flow variables where a change can move recovery rate on its own.

Email 1: Simple Reminder

The first email should remind, not sell and not discount.

Include:

• the abandoned product, shown clearly;
• one obvious path back to the cart;
• relevant risk reversals such as returns, guarantee or shipping terms.

Most abandonment is distraction, not rejection. A phone call, a closed tab or a delivery at the door ends far more sessions than price objections do, and those customers need a reminder rather than a concession.

Email 2: Human Check-In

The second email works well as a short, personal message rather than a designed template.

Write it in the voice of the founder or the customer service team, keep it to a few lines, and offer help with the things that genuinely stall a purchase:

• sizing and fit;
• delivery timelines;
• specifications and compatibility;
• anything they were unsure about.

A plain-text style email often reads as more personal, but it does not guarantee any particular inbox placement. Use it because it suits the message, not as a deliverability trick.

Email 3: Handle the Real Objections

By the third message, assume something specific is holding the purchase back and answer it directly.

Common objections worth addressing:

• shipping cost or speed;
• return policy;
• product quality or durability;
• fit, sizing or suitability;
• price relative to perceived value;
• guarantees and warranty;
• trust in a brand they do not know yet;
• how the product compares to the alternative they are considering.

Answering objections protects margin far better than discounting does. A discount removes the objection by paying for it, and it teaches the customer that waiting is rewarded.

Split New and Returning Customers

New and returning customers abandon for different reasons, so they should not automatically receive identical incentives.

A first-time visitor has no experience with the product, the shipping or the service. After a reminder and objection handling, a financial nudge may genuinely be what closes the order.

A returning customer already knows the brand delivers. Giving them the same incentive frequently means paying for an order that was likely to happen anyway.

A reasonable split looks like this:

Returning customers

Support, reminders and relevance. Add a smaller incentive only where the margin and customer value justify it.

New customers

A time-limited offer can be tested later in the sequence, after the reminder and objection-handling messages have had their chance.

We do not recommend a fixed discount percentage. The right incentive depends on margin, AOV and how often the brand already promotes.

Example Flow Structure

This is a starting framework, not a template to copy without adaptation.

1. 1Reminder with the abandoned product
2. 2Personal check-in offering help
3. 3Objection handling
4. 4Conditional split: new vs returning customer
5. 5New customer offer, if economically justified
6. 6Final reminder before the offer expires, if an offer is used

The structure should change with average order value, margin and buying cycle. A considered high-AOV purchase can support a longer, more educational sequence. A low-margin impulse product usually cannot support an incentive at all.

What to Test

• initial delay before the first message;
• how email and SMS are coordinated across the sequence;
• plain text vs designed templates;
• which objection you lead with;
• incentive size;
• incentive timing within the sequence;
• returning-customer branch logic;
• total number of emails.

Cart recovery is one of a small number of automations worth building first. For the wider priority order, see the five ecommerce email automations to prioritize first.