Verified against ChatGPT · 2026-08-12
Build a retargeting ad sequence keyed to exactly where each segment dropped off, not one generic 'come back' ad
Produces a 3-touch retargeting ad sequence with distinct messaging for each named drop-off segment (viewed product, added to cart, started checkout), since each represents a different reason for not converting and needs a different argument.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Build a 3-touch retargeting ad sequence for A direct-to-consumer mattress brand., with separate copy for each of the drop-off segments listed below, since a visitor who abandoned checkout is not the same as one who only viewed a product page, and one generic "come back!" ad wastes the segmentation data most ad platforms already give you. PRODUCT / SERVICE A direct-to-consumer mattress brand. DROP-OFF SEGMENTS TO TARGET SEPARATELY Viewed product page only, added to cart, started checkout but abandoned at payment. (e.g., viewed a product page but didn't add to cart, added to cart but didn't start checkout, started checkout but didn't complete payment) MOST LIKELY REASON EACH SEGMENT STALLED Checkout abandoners frequently cite unexpected shipping costs shown only at the final step, per exit survey data. DISCOUNT OR INCENTIVE POLICY No discount for browse-only segment; 10% one-time code for cart-abandoners; free shipping (not percentage discount) for checkout-abandoners. For each segment, write a 3-touch sequence (Day 1, Day 3, Day 7 or similar spacing) where each touch's copy is built around the stated likely stall reason for that specific segment, not a generic reminder. A viewer who only browsed likely needs more information or reassurance, not a discount; a cart-abandoner likely needs urgency or a small nudge; a checkout-abandoner likely hit a specific friction point (payment method, shipping cost, trust at the final step) and needs that friction addressed directly, not a generic "you left something in your cart" reminder repeated three times with different subject lines. Respect the stated incentive policy exactly — do not introduce a discount if the policy says not to use one for a given segment, and do not escalate discount percentage across touches if the policy defines a fixed maximum. If a segment's real stall reason wasn't given, do not invent a specific psychological explanation as fact — write the sequence around addressing friction generically for that segment and flag that the stall reason should be confirmed with actual data (survey, session recordings) rather than assumed. OUTPUT: one section per segment (segment name, then Day 1 / Day 3 / Day 7 ad copy, each under 2 sentences plus a CTA), followed by a note on which segment is highest-value to fix first based on proximity to conversion.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
Retargeting platforms segment audiences by funnel stage specifically because each stage represents a different, mechanically distinct reason for non-conversion, but a prompt that just asks for "retargeting ad copy" tends to produce one templated "you left something behind" message applied uniformly across segments, wasting the exact behavioral signal the segmentation data provides — requiring separate copy keyed to each segment's likely stall reason forces the model to treat a browse-only visitor (who needs more information, since they haven't committed to intent yet) differently from a checkout-abandoner (who has already shown strong intent and likely hit a specific, addressable friction point like unexpected shipping cost), which is the actual determinant of what argument will move each group. The explicit instruction not to escalate discount percentage across touches or violate the stated incentive policy addresses a specific, real business risk: a model asked to write a persuasive 3-touch sequence will often default to escalating urgency and discount depth touch over touch because that reads as more compelling in isolation, but that pattern trains customers to wait out the sequence for the biggest discount, which is a documented and costly failure mode in real retargeting programs. Refusing to invent a specific psychological stall reason when none was given — and instead flagging that it should be confirmed with real data — matters because a plausible-sounding but fabricated explanation ("this segment is price-anxious" asserted with no evidence) can steer an actual ad budget toward the wrong lever, like adding a discount when the real issue was a confusing checkout form.
What you get back
Checkout abandoners, Day 1: "Your order's still saved — and shipping's on us if you finish today." No discount escalation Day 3 or 7 per stated policy; Day 3 copy instead re-surfaces the free-shipping offer with a product reassurance angle rather than a deeper discount, since the incentive policy caps this segment at free shipping only.
Verified against
ChatGPT GPT-5.1 · 2026-08-12
Changelog
- 2026-08-12 — 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
