Abandoned Cart Recovery in WooCommerce: Plugin Setup and AI Copy

Install a free cart recovery plugin such as Cart Abandonment Recovery for WooCommerce, turn on its consent checkbox, schedule three emails at roughly one hour, one day and three days, and ask an AI assistant to draft the copy from your brand notes. Send a discount only in the last email, then measure recovered orders against a group you did not email.

The plugin does the capturing, timing and stopping. The AI writes words. What neither can do is decide whether the reminder is lawful for your customers or whether the revenue it claims was truly recovered, and those two checks are the parts most stores skip. In the worked month further down, the plugin's report claims $1,664 recovered; a simple holdout suggests the emails themselves earned about $640 of it, which is still worth having.

Follow me on Instagram@sagnikteaches

Pick a plugin: two free ones that do the basics

Two free plugins from the WordPress plugin directory cover what a small store needs. Both were updated in October 2026 when this page was written, but check the listing for the current version and for what your WooCommerce version supports before you install.

Connect on LinkedInSagnik Bhattacharya

For a store with a few hundred orders a month, either will do. If you want the larger install base and a plain checkout consent box, start with the first. If you want a cart-level view of the most abandoned products, the second shows it. The second also lets you exclude carts by IP address or email domain, which keeps your own test carts and staff orders out of the figures.

Subscribe on YouTube@sagnikteaches

If you have been looking at AI add-ons for the store more widely, what AI assistant plugins do and what to review explains where this fits.

Set up the plugin in an evening

Do this on a staging copy of the site if your host offers one, because the plugin touches checkout.

1. Back up and install. Take a backup, then add the plugin from Plugins, Add New, and activate. The plugin adds its own admin screen under the WooCommerce menu.
2. Set the cut-off time. This is how long a cart sits idle before it counts as abandoned. Cart Abandonment Recovery's changelog gives its default as 20 minutes; Abandoned Cart Lite lets you set your own. Too short and you email people who are still on the page.
3. Turn on the consent checkbox. In the same settings area, enable the GDPR option so checkout asks permission before the address is stored for reminders. Write the label in your own words (see the consent section below).
4. Exclude what you do not want chased. Exclude your own admin users and, if the plugin offers it, orders in certain statuses and products you do not want reminders for (for example made-to-order items).
5. Create three email templates with the delays in the next section. Leave them as drafts until the copy is ready.
6. Check that scheduled tasks run. Reminder emails depend on WordPress's scheduled tasks. If your site has little traffic or a caching layer, the tasks may fire late. Cart Abandonment Recovery switches to Action Scheduler automatically when WordPress's own scheduler (WP-Cron) is disabled, but ask your host if you are unsure.
7. Send a test. Abandon a cart yourself from a private window with an address you own, wait through the cut-off, and check the email arrives, the restore link rebuilds the basket, and the sequence stops once you complete the order.

Also check your sending domain's email authentication (SPF, DKIM, DMARC) with your host or email provider. Cart emails sent from a domain with no authentication land in spam, and no amount of good copy fixes that.

Consent and the rules on cart reminder emails

This is the step that deserves care, and it is not legal advice. If you have customers in the EU or in markets with similar rules, two sets of rules apply: the GDPR, which covers storing and using the email address and any browsing data you capture, and the ePrivacy rules, which cover sending electronic marketing (PECR, the Privacy and Electronic Communications Regulations, is the best-known national version). Whether a cart reminder counts as marketing depends on its content and context, and the rules give no single answer. A reminder that only helps someone finish an order has the best case for being a service message; adding promotional content tips it towards marketing. PECR also has a so-called soft opt-in for marketing similar products to people whose details you collected during a sale or negotiations for a sale, provided they were offered a simple opt-out when the details were collected and in every message; whether an abandoned checkout counts is a question for your adviser. A cautious position for a small store:

• Capture the email address only through a clear checkbox at checkout that says what you will do with it, such as "Email me a reminder if I leave without paying".
• Keep the first email a plain reminder of the items and a link, without promotional content.
• Treat any email with a discount or recommendations as marketing, and send it only to people who agreed to marketing.
• Put an unsubscribe link in every email and honour it straight away, and make sure the plugin's list is respected by your newsletter tool.
• Say in your privacy policy that you send cart reminders, what data you hold, and for how long.
• Stop the sequence at three emails. A fourth is where complaints begin.

An illustrative consent line: "Email me about my basket if I leave before paying. You can unsubscribe at any time." Keep the box unticked by default. If you serve customers in several markets, ask a qualified adviser to look at the wording once, and see the consent and unsubscribe checklist for the general pattern in an email tool.

Plan the sequence before the AI writes a word

Three emails with different jobs work better than one email sent three times.

The timings are starting points. Fifteen-minute emails reach people who are still choosing; waiting a week means the intent has gone. Test a longer first delay if your customers are likely to compare prices in another tab.

Worked example: a store with 420 abandoned carts a month

This is an illustration with invented figures. Say a small store selling home textiles runs on WooCommerce with an average order of $64 and 1,100 checkouts started in a month. Of those, 680 finish and 420 abandon. The plugin captured an email address on 130 of the 420 abandoned carts (31%), either because the shopper typed it before leaving or because they were logged in. The other 290 are unreachable by email.

The owner turns on the three-email sequence. At the end of the month the plugin's report shows:

On its face that is $1,664 for a free plugin and an evening's work. Do not believe it yet. The plugin counts any order from an emailed shopper within its window, including shoppers who would have come back anyway, and who the first email merely preceded. The next section is how to find out how much is real.

Measure what the emails really recovered

The clean way is a holdout: leave a random slice of abandoned carts without emails and compare. Many plugins cannot do this natively, but you can approximate it:

1. For a month, send the sequence to only 80% of captured carts. Pick them by alternating in the plugin's list, or by pausing the sequence on every fifth cart, or by using the "exclude" option on a test tag if your plugin has one.
2. Count how many carts in each group turned into an order within seven days.
3. Compare the two rates.

Illustrative result for the following month, with similar traffic: 104 emailed carts gave 21 orders (20%) and 26 held-back carts gave 3 orders (12%). The emails lifted recovery by about 8 percentage points. Applied to the first month's 130 emailed carts, that is roughly 10 orders the sequence earned, not 26: at $64 each, about $640 rather than $1,664, less the $57.60 in discounts. That is still worth having for a free tool, and now you know what the number is. Small samples are noisy: 26 held-back carts is thin evidence, so repeat the holdout for a second month before making a decision.

Also look at what the discount did. Of the 9 discounted orders, how many had already visited the basket link in email 1 or 2 without buying? If most of them had, the discount was a gift to people who were returning anyway, and you might run the last email without one. Check the orders list for those codes, and compare average order value with and without the code.

Writing the three emails with AI

Give the assistant your tone, your real policies, and a rule against invented claims. Do not let it make up delivery times, stock levels, reviews or urgency.

I run a small WooCommerce store selling home textiles. Write a three-email abandoned cart sequence.

Facts you may use (nothing else):
- Delivery 3 to 5 working days, free over $60, otherwise $5.
- Returns within 30 days if unused; the customer pays return postage.
- Support: reply to this email, answered within one working day.
- Brand voice: warm, plain, short sentences, no exclamation marks.

Merge tags: [paste the exact tags your plugin's email editor lists for first name, cart contents, restore link, coupon code, coupon expiry and unsubscribe link].

Email 1 (1 hour): reminder only, no discount, no promotional content.
Email 2 (24 hours): answer likely doubts about delivery, returns and support. No pressure, no fake scarcity.
Email 3 (72 hours): last note, mention the 10% code using the coupon code and expiry tags.
For each: three subject lines under 50 characters and a preview line under 90, plus the body under 110 words with one button labelled with a verb.
Do not claim items are selling fast or low in stock. Do not invent reviews.

A realistic first draft of email 1 (illustrative):

Subject: Your basket is waiting
Preview: The items you chose are still saved.

Hi [first-name tag],

You left a few things in your basket. They are saved, so you can pick up where you stopped.

[cart contents tag]

Delivery takes 3 to 5 working days, and it is free over $60. If you have a question first, reply to this email and we will answer within one working day.

[Button: Return to your basket, linked to the restore-link tag]

[unsubscribe link tag]

Now the edits you will want to make. A typical first draft also gives subject lines such as "Don't miss out on your items!", which breaks your no-exclamation rule and your no-pressure rule, so replace it. The AI may also write "Hurry, stock is limited" in email 2 despite the instruction, so search for "hurry", "limited", "last chance" and "selling fast" in every draft. Check each fact against your own policies page: delivery times, free shipping threshold, return window. Fill the merge tags exactly as the plugin writes them, because the tag syntax varies between plugins and a mismatched tag prints as literal braces in a customer's inbox. Send yourself all three and read them on a phone.

A before and after for the second email

The second email is the one that most often turns into filler. Here is a weak draft and its improved version (illustrative):

The second version has facts the shopper can use and nothing that cannot be proved. "Our customers love these products" is a claim; if you have genuine reviews, quote one, with the customer's permission, in its own words.

Two more carts, two different treatments

A single sequence suits most carts, but two cases deserve their own handling, and neither needs extra software.

The large basket. Say the store sells made-to-measure curtains and an abandoned cart is worth $380. A three-email sequence with a 10% code gives away $38 on a sale that needs a human touch. Exclude carts above a set value from the automatic sequence (most plugins let you exclude by value or by product) and have the owner send a short personal note instead: "I saw you were looking at the linen curtains. If it helps, I can check measurements with you before you order." Ask the AI to draft three variants of that note for you to edit, and keep the one that sounds like you. This takes minutes and tends to suit high-value, considered purchases.

The returning customer. A customer with three previous orders who abandons a basket does not need the first email to explain delivery. If your plugin or email tool can tell new from returning shoppers, shorten the sequence to the reminder and the last note, and leave out the discount. If it cannot, check the first week's emails to see how many went to repeat buyers and decide whether a manual exclusion is worth it.

How to read the report each week

Once the emails run, ten minutes a week is enough. Paste the plugin's weekly figures into the assistant (carts captured, emails sent, orders recovered, revenue) with the previous four weeks and ask for the changes, not the totals:

Here are five weeks of cart recovery figures: week, abandoned carts with an email, email 1 sent, email 2 sent, email 3 sent, orders recovered, revenue recovered, discount given. Tell me (1) which week changed most against the four-week average and by how much, (2) the share of recovered orders that used the discount, (3) whether the recovery rate moved with the number of carts captured. Do not explain causes unless the numbers show them; list questions I should check instead.

An illustrative answer: "Week 5 recovered 7 orders from 36 captured carts (19%), in line with the average of 20%. Discount use rose from 30% to 44% of recovered orders. Questions: did a sitewide sale start in week 5? Were new customers added to the list through a competition?" The second question matters, because recovery figures change when the audience does.

What goes wrong in the first month

• Emails to people who already bought. A customer completes the order on their phone after abandoning on the laptop. If the plugin matches on the email address it stops the sequence; if the shopper used two addresses it does not. Check the first week's sent emails against the orders list.
• Coupon stacking. The cart-recovery code combines with a sitewide sale, and margins vanish. In WooCommerce's coupon settings, tick Individual use only so it cannot combine with other coupons and Exclude sale items so it does not discount stock already on sale, and set a minimum spend.
• Broken restore links. Caching plugins or a changed checkout page can break the link that rebuilds the basket. Test it logged out, on a phone.
• Odd entries in the cart list. A burst of carts with no names or an unusual country can be bots. Exclude them in the plugin's settings, or ignore them in the recovery rate.
• The shopper changed their mind for a reason you could fix. The cart report shows shipping cost as a frequent reason to abandon. Read how to set a free shipping threshold from your average order value: a fix at checkout often beats any email.

Where this fits with the rest of the store

Abandoned cart emails are one automated flow among several, and a store that builds them in a tool such as Klaviyo will find the shape familiar; the Klaviyo and Shopify Email comparison shows what a dedicated email platform adds. There is also a more general piece on setting up AI abandoned cart emails for a small online store that is not tied to WooCommerce. Keep the product pages that carts start from in good order with AI descriptions and support replies for WooCommerce, and check the sequence before any busy period using the peak season checklist.

Budget one evening for setup, an hour for the copy with edits, and ten minutes a week for the first month to read the report, check for oddities and run the holdout. The plugin is free; email sending costs depend on your host or email provider. If you want someone to go through your checkout, consent line and sequence with you, an AI implementation consultation is where that happens.

Questions store owners ask before turning this on

Do I need a discount in the recovery emails?

Not in the first one. Many carts are abandoned because the shopper was interrupted, so a reminder with the basket and a clear checkout link is often enough. Hold any incentive for the last email in the sequence, limit it in value and time, and test against a group that gets no discount to see whether it earns its cost.

Is a cart reminder email marketing in law?

It depends on the content and on where your customers are. A bare reminder of an unfinished order has the best case for being treated as a service message, but anything with promotional content or a discount is likely to count as marketing. Collect consent at checkout, include an unsubscribe link in every email and ask a qualified adviser about your own case.

How soon should the first email go?

Around one hour after abandonment is a common starting point, with a second at about a day and a last at about three days. Fifteen-minute emails can feel pushy and can reach someone who is still deciding. Treat the timings as your first test, not as fixed facts.

Why do some carts never get an email?

The plugin can only email shoppers whose address it captured. That happens when a visitor types an email at checkout or is logged in. Visitors who add to cart and leave without entering an address are not reachable by this method, so expect to email only part of the abandoned carts.

Further reads

• AI Chatbot for WooCommerce: Costs, Order Lookup and a Test Script — Answer the doubts that cause abandonment before checkout.
• WooCommerce SEO With AI: Meta Descriptions and Schema Checks — Bring better-matched visitors to the carts in the first place.
• Klaviyo Back in Stock Emails for Shopify: Setup and AI-Drafted Copy — Another triggered email flow, written with AI.
• Dropshipping Returns and Refunds: Policy Wording and AI Replies — Draft the returns policy your second email points to.
• AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.

Sources: the WordPress.org plugin pages for Cart Abandonment Recovery for WooCommerce (Brainstorm Force) and Abandoned Cart Lite for WooCommerce (Tyche Softwares), read in November 2026; WooCommerce's coupon management documentation; published guidance on PECR's soft opt-in. The worked example is illustrative.