Verified against ChatGPT · 2026-08-10
Draft a three-email abandoned-cart sequence that escalates reasons, not just reminders
Generates three distinct abandoned-cart emails — nudge, objection-handling, and final incentive — each doing a different job instead of repeating the same 'you left something' message three times.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Write a three-email abandoned-cart recovery sequence for the cart below. Each email must do a genuinely different job — this is not one email resent three times with a bigger discount each round. ABANDONED CART CONTENTS One pair of leather ankle boots, size 8, in cognac — $189. LIKELY REASON FOR ABANDONMENT Shipping cost — cart abandonment rate spikes right after the shipping cost is shown at $12.99, based on funnel analytics. SEND TIMING 1 hour after abandonment, 24 hours after, 72 hours after. DISCOUNT POLICY Up to 10% off on the third email only, never on the first two, per brand pricing policy. BRAND VOICE Warm and understated, like a small independent boutique — never exclamation-heavy or hypey. EMAIL 1 — THE NUDGE (send per the first timing given) A short, low-pressure reminder. State exactly what's in the cart (not vague "your items") and nothing else — no discount, no urgency. Assume the most likely explanation is simply distraction, not price hesitation, so don't undercut the price before you know that's the issue. EMAIL 2 — THE OBJECTION HANDLER (send per the second timing) Address the likely reason for abandonment directly and by name — if it's price, show value or a payment-plan option before a discount; if it's shipping cost or delivery time, state the actual policy; if it's simply uncertainty about fit or quality, include something that reduces that specific uncertainty (a return policy line, a review excerpt). Do not default to a discount here unless the likely reason given is actually price — matching the wrong lever to the actual objection wastes the strongest email in the sequence. EMAIL 3 — THE FINAL ONE (send per the third timing) Only at this point introduce a discount, and only if the discount policy allows one — if it doesn't, use a real closing signal instead (limited stock if true, or simply state this is the last reminder) rather than inventing urgency. State the discount plainly with an expiry, no stacked exclamation points or manufactured scarcity language beyond what's actually true. WHAT NOT TO DO Do not open any email with "we noticed you left something in your cart" verbatim — vary the opening across all three so the sequence doesn't read as templated. Do not repeat the same product description language across all three emails. OUTPUT FORMAT Three complete emails, each labeled with its send timing, subject line, and body — ready to load into an ESP.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
Assigning each email in the sequence a distinct job counters the single most common failure in AI-generated cart-recovery sequences: without an explicit constraint, GPT-5.1 will produce three emails that are structurally identical — reminder plus escalating discount — because that's the dominant pattern in the training data for 'abandoned cart email sequence,' and it's also the laziest way to satisfy a three-email request without doing the harder work of differentiating intent per email. Requiring the second email to match its lever to the actual stated abandonment reason, rather than defaulting to a discount, addresses a specific and costly mistake: offering a discount to someone who abandoned over shipping cost or fit uncertainty (not price) trains that segment to expect discounts on future purchases while never actually addressing why they left, which depresses full-price conversion over time even as the immediate recovery email 'works.' Delaying the discount to only the final email, and gating it on the stated discount policy rather than assuming one exists, prevents the model from inventing a markdown lever that may not reflect real margin constraints — a common failure when a model is asked to write persuasive copy without an explicit boundary on what levers are actually available. The instruction against a templated opening line addresses a surface-level tell that undermines trust in the whole sequence: if all three emails visibly share the same generator, the brand-voice work in the rest of the copy is wasted, since recipients pattern-match templated bulk email at the subject-line and opening-line level before reading further.
What you get back
Email 1 subject: "Still deciding on the cognac boots?" — no discount, no urgency, just the fact of what's saved. Email 2 subject: "About that $12.99 shipping" — addresses the actual likely objection with the real shipping policy, no discount. Email 3 subject: "Last call — 10% off before these go back to stock" — states the discount plainly with a real expiry.
Verified against
ChatGPT GPT-5.1 · 2026-08-10
Changelog
- 2026-08-10 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — Google Ads management, built and maintained for you — that's Scult's day job.
EXPLORE GOOGLE ADS MANAGEMENT
