Cart Abandonment Recovery
The number everyone misreads
The commonly cited ~70% cart abandonment rate gets treated as a single problem with a single fix: send an email. In reality that number is an average across two very different behaviors. A meaningful share of "abandoned carts" are people who added an item to compare price or shipping cost, never intended to buy that session, and are better described as browse abandonment with a cart artifact left behind. A smaller, higher-intent share genuinely got to checkout and stopped — payment friction, surprise shipping cost, a stalled trust decision. Recovery programs that treat both groups identically waste discount budget on the first group and under-serve the second.
Browse abandonment vs. checkout abandonment
| Signal | Browse abandonment | Checkout abandonment |
|---|---|---|
| Where they stopped | Product page or cart, pre-shipping info | Shipping/payment step of checkout |
| Likely cause | Comparison shopping, low intent this session | Price shock, friction, trust hesitation |
| Best first message | Product reminder, social proof, no discount | Direct checkout link, address the specific friction |
| Discount sensitivity | Low — discount trains behavior, doesn't convert intent | Higher — can tip a near-decision |
| Ideal channel | Email, retargeting ads | SMS or email with urgency, same-day |
Timing the sequence
The single highest-converting message in most recovery programs is the first one, sent within 1 hour of abandonment while intent is still warm — open and click rates on this message typically run well above the rest of the sequence combined. A workable default cadence is: message 1 at ~1 hour (reminder, no discount), message 2 at ~24 hours (address objections — shipping cost, reviews, FAQ), message 3 at ~48-72 hours (final nudge, discount reserved for checkout-stage abandoners only). Sending faster than an hour feels surveillance-adjacent to some customers; waiting past 72 hours mostly fails because purchase intent has moved on to a competitor or simply decayed.
Trigger: cart created, no purchase within window
T+1h Reminder email — cart contents, no discount, 1 CTA
T+24h Objection-handling email — reviews, shipping/returns info,
segment by abandonment stage (browse vs. checkout)
T+48h SMS (if opted in) — short, direct checkout link
T+72h Final email — discount ONLY if checkout-stage abandoner
and no prior discount used in last 30 days
Suppress sequence immediately on purchase or cart clear.
Cap discount exposure: track % of recovered revenue that
used a discount vs. converted at full price.Discount-first is a self-inflicted wound
Testing discount vs. no discount properly
A valid test holds the message timing and copy constant and varies only the offer, run on a large enough segment to detect a real effect, over at least one full purchase cycle. The metric that matters is not recovery-sequence conversion rate in isolation — it's net revenue per abandoned cart after discount cost, and ideally a read on whether the discount pulled forward a purchase that would have happened anyway at full price within the same week. Many teams discover the discount variant wins on conversion rate but loses on net margin once that cannibalization is priced in.
What to measure beyond recovery rate
| Metric | What it tells you | Common mistake |
|---|---|---|
| Recovery rate | % of abandoned carts that convert | Treated as the only success metric |
| Net revenue per cart | Recovery rate minus discount cost and cannibalization | Rarely calculated at all |
| Time-to-recovery | How much of the sequence was needed | Ignored — informs sequence length |
| Repeat-abandon rate | Whether discounts are training repeat behavior | Almost never tracked |
Segmentation that actually changes outcomes
First-time abandoners and repeat abandoners should not get the same sequence. A first-time visitor abandoning at checkout is plausibly hitting friction (unexpected shipping cost, a required account creation step); a returning customer who has abandoned three times in the past has probably already decided discount-waiting is the winning strategy, and continuing to offer discounts simply confirms it. High-value cart segments (above your average order value) deserve a human-reviewable escalation path — a support outreach rather than an automated discount — since the margin at risk justifies the attention.
Shipping cost is the single biggest checkout-stage killer
What's next
Recovery is the first touch after a near-purchase; what happens after the purchase actually completes is just as underused a marketing surface. Next: Post-Purchase Marketing →
I build these systems professionally.
Whether it's a RAG pipeline, analytics migration, or AI workflow — let's talk.