Roughly seven in ten online baskets are abandoned. That statistic is usually presented as lost revenue, which overstates it — most of those people were browsing, comparing prices or using the basket as a wish list.

The opportunity is the smaller group who intended to buy and were interrupted or put off. Reaching them is worthwhile. Bombarding everyone else is not.

Fix the causes before building the recovery

Recovery emails treat the symptom. Before investing in them, address the reasons people leave, because that improves the whole funnel rather than chasing the leak.

The main ones: delivery cost appearing at the last step, forced account creation, a checkout that feels long, missing payment methods, and doubt about returns. These are covered in why your shop gets traffic but not orders.

A shop that fixes those has fewer abandonments to recover and better economics on the ones it does.

The sequence that works

First message: one to two hours

Sent while the intent is still fresh. Treat it as helpful rather than commercial.

Show what they left, with images. Make returning to the basket a single click that restores everything. Include a line about delivery and returns, because uncertainty about those is a common cause. Offer a way to ask a question.

No discount. No urgency. The tone is "you left this, here it is again".

Second message: about twenty-four hours

Address the hesitation rather than repeating the basket.

This is where reassurance belongs: your returns policy stated plainly, delivery timescales, a relevant review of the item, an answer to the question people most often ask about it.

Still no discount, in most cases.

Third message: three to five days, optional

The last one. If you use an incentive at all, it belongs here — and only for baskets above a value that justifies the margin.

Then stop. A fourth message earns unsubscribes, not orders.

Discounting in the first email teaches your best customers to abandon on purpose. It is the fastest way to convert a recovery programme into a permanent price cut.

The discount question

It is tempting because it works in the short term and the measurement flatters it. The problem is that people learn.

Once customers discover that abandoning produces a ten per cent code, the sophisticated ones will do it every time, and you are discounting orders you would have received at full price.

Safer approaches: free delivery rather than a percentage off, which costs less and removes a known objection; a discount only above a basket value; or no discount at all, with the effort spent on reassurance instead.

The rules that apply in the UK

Marketing email to individuals requires consent or the soft opt-in — which applies where you obtained the address in the course of a sale or negotiations for a sale of similar products, and gave a clear chance to opt out at that point and in every message.

Someone who typed an email address into your checkout and did not complete may be covered by soft opt-in, and the position is not identical for every business. Practical steps that keep you safe:

  • Make the opt-out obvious in every message and honour it immediately.
  • Keep a record of where and when the address was captured.
  • Do not carry abandoned-basket addresses into your general marketing list without separate consent.
  • Set a limit — do not email someone about a basket they left eight months ago.

Beyond email

On-site messages when someone moves to leave a page can work, particularly for offering help. On mobile they are intrusive and should generally be avoided.

Retargeted advertising reaches people who never gave you an address. It requires proper consent for the tracking involved, and the frequency needs capping — following someone around the internet for a fortnight with an item they decided against damages the brand more than it recovers.

Saved baskets are the quiet winner. Simply keeping the basket intact when someone returns days later removes the friction entirely, and it requires no message at all.

Measuring it honestly

The trap is counting every recovered order as incremental. Some of those people would have returned anyway — the email merely happened to arrive first.

To get a real figure, hold back a small proportion of abandoners at random and send them nothing. The difference in completion between that group and the emailed group is your actual uplift. It is usually smaller than the headline number and it is the one worth optimising.

Watch unsubscribes and complaints alongside recovered revenue. A programme that recovers a modest amount while eroding your list is a poor trade.

Our automation team builds recovery sequences for UK retailers, with the measurement set up so you can see the real effect. Start a conversation.