Verified against ChatGPT · 2026-08-09
Pitch an adjacent product to an existing customer without sounding like a cold pitch
Writes a cross-sell pitch grounded in a specific problem the existing customer already has, connecting it to what they already trust about the relationship, instead of a generic new-product announcement.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Write a cross-sell pitch to Riverton Health Group, an existing customer, for Our analytics add-on module — it needs to read like it's grounded in a real problem they have, not a generic 'check out our other product' announcement. WHAT WE KNOW ABOUT A PROBLEM THIS PRODUCT SOLVES FOR THEM Their team has mentioned manually exporting data to build reports for their board meetings each month. HOW LONG THEY'VE BEEN A CUSTOMER AND OF WHAT Customer for 18 months, on the core scheduling platform, generally satisfied based on last QBR. WHY THIS ISN'T JUST A COINCIDENTAL PITCH Their ops lead specifically asked in our last call whether there was a way to automate the board report exports. Ground the pitch in the specific known problem, stated in their own likely language, before mentioning the adjacent product by name — the problem has to feel recognized, not invented to justify a pitch. Reference the existing relationship specifically enough that it's clear this is coming from someone who actually knows their account, not a mail-merged cross-sell blast — cite something concrete about their tenure or current usage, not a vague 'as a valued customer' line. If the connection between the known problem and the adjacent product requires two logical steps rather than one, spell out the middle step explicitly rather than assuming the reader will make the jump themselves — a cross-sell pitch that requires the customer to do inference work to see the relevance loses them before they get to the ask. Keep the tone as a colleague flagging something useful, not a rep hitting quota, and make the next step something easy to say yes to. OUTPUT FORMAT One outreach message, 100-150 words, plus a one-line internal note on the strongest single reason this customer specifically is a good fit for the adjacent product, for my own reference before I send it.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
Requiring the problem to be stated in the customer's own likely language before the product is named addresses a structural weakness in AI-generated cross-sell copy: models will readily produce grammatically fluent product-benefit sentences, but a pitch that leads with the product name and works backward to a justification reads as exactly what it is — a pitch looking for a reason, rather than a reason that happens to have a pitch attached — and buyers who've seen enough vendor outreach can tell the difference on a skim. Requiring a specific, concrete detail about tenure or usage rather than a generic 'valued customer' line matters because that generic phrase is a well-known marker of mail-merge automation, and using it (or letting a model default to it, which it will if not told otherwise) signals to the reader that this wasn't actually written with their account in mind, undermining the entire premise of relationship-based cross-selling. The instruction to spell out an implied two-step logical connection rather than leaving it implicit exists because models tend to compress reasoning chains when writing persuasive copy — they'll happily state a conclusion ('this add-on would help you') without showing the connecting logic, but a reader who has to reconstruct that connection themselves is a reader who disengages before reaching the ask, since making an unstated inferential leap is real cognitive work that most recipients skimming an email won't do. The internal-note request at the end is a deliberate quality check: forcing the model to state its single strongest reasoning for fit, separate from the customer-facing copy, surfaces whether the pitch is actually grounded in something specific or whether it's dressed-up generic reasoning, before the message goes out.
What you get back
Hi Priya, I remember your ops lead asking on our last call whether there was a way to automate the manual data exports your team pulls together for board reporting each month. Our analytics add-on does exactly that — it would plug directly into the scheduling data you're already using and generate that board-ready report automatically, no manual export needed. Want me to put together a quick walkthrough of what it would look like with your actual data? Internal note: strongest fit reason — this directly answers a specific, recent, unprompted question from their own team, not an inferred need.
Verified against
ChatGPT GPT-5.1 · 2026-08-09
Changelog
- 2026-08-09 — 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
