Verified against ChatGPT · 2026-08-12
Write a Problem-Agitate-Solution ad that agitates a real consequence instead of manufacturing panic
Drafts a Problem-Agitate-Solution ad using a real, specific consequence of the stated problem to build the agitation section, explicitly avoiding fear-mongering or invented stakes that don't match the actual offer.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Write a Problem-Agitate-Solution (PAS) ad for An automated invoice-reconciliation tool for small bookkeeping firms., where the agitation section is built from a real, specific consequence the target audience actually experiences — not manufactured panic or a worst-case scenario disconnected from what this offer actually fixes. OFFER An automated invoice-reconciliation tool for small bookkeeping firms. REAL PROBLEM (in the audience's own words if possible) "I spend Friday afternoons manually matching bank statements to invoices instead of billing clients." ACTUAL, VERIFIABLE CONSEQUENCE OF NOT SOLVING IT Roughly 3-4 unbillable hours per week per bookkeeper, based on internal time-tracking data. AUDIENCE Solo and small-team bookkeeping firm owners handling 15-30 client accounts. TONE GUARDRAIL No language implying they're bad at their job or falling behind competitors — this audience responds badly to being made to feel incompetent. Structure: Problem (state it plainly in language the audience would use about themselves, not marketing language describing them), Agitate (expand on the ACTUAL consequence given — the cost of time, money, or reputation this problem creates if unaddressed — without escalating into a worst-case scenario or emotional manipulation not grounded in the stated consequence), Solution (introduce the offer as resolving specifically the consequence just described, not a broader pitch that reintroduces new claims). The agitation section is the part most likely to overreach — do not add a consequence more severe than what I gave you (e.g., if the real consequence is "you lose 3 hours a week to manual data entry," do not escalate to implying the business will fail); do not use fear-based language that implies physical, financial, or legal danger unless that danger is explicitly what Roughly 3-4 unbillable hours per week per bookkeeper, based on internal time-tracking data. describes. If a tone guardrail was given, treat it as a hard constraint on the agitation section specifically, since that's the section most likely to violate it. Do not fabricate a statistic to make the consequence sound worse than stated. OUTPUT: three labeled sections (Problem / Agitate / Solution) as ad copy ready to adapt to a specific channel, followed by one line confirming the agitation section's severity matches the stated consequence and did not escalate beyond it.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The Problem-Agitate-Solution framework's Agitate step is the section most prone to overreach in AI-generated copy, because the pattern is well-represented in aggressive direct-response marketing training data where agitation routinely escalates into manufactured worst-case scenarios disconnected from the actual stated problem, and a model asked to "agitate" without a ceiling will often reach for that more dramatic register by default since it reads as more persuasive on its own terms. Explicitly capping the agitation section at the real, given consequence — and giving a concrete example of what escalation beyond it looks like (three hours lost per week becoming implied business failure) — gives GPT-5.1 a specific boundary to check its own output against rather than a vague instruction like "don't exaggerate," which is unenforceable because exaggeration is a matter of degree the model can't self-assess without a stated ceiling. The tone guardrail being applied specifically to the agitate section (rather than the whole ad generically) matters because that's structurally where an audience-alienating tone is most likely to leak in — a Problem section stated plainly and a Solution section pitching the offer are lower-risk than an Agitate section that, done carelessly, can read as insulting or manipulative to the exact audience it's trying to move, which is a real risk with framing that implies the reader is incompetent or falling behind. Ending with an explicit self-check that the agitation didn't escalate beyond the stated consequence turns an easy-to-skip quality bar into something the output has to affirmatively demonstrate before being considered done.
What you get back
Agitate: "Every Friday, that's another 3-4 hours you're not billing a client for — time spent matching line items by hand instead of doing the work your clients actually pay you for." Solution: "[Tool] reconciles bank statements against invoices automatically, so Friday afternoon goes back to billable work." Confirmation: agitation stays within the stated 3-4 hour/week cost and does not imply business failure or incompetence.
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
