Verified against ChatGPT · 2026-08-05
Build a competitor battle card you can actually use mid-call
Turns what you know about a named competitor into a scannable, honest battle card — including where they beat you — built for a glance mid-call, not a slide nobody reads, with proof points flagged as illustrative or verified.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Build a battle card for when we're up against Ledgerline in a deal. What we know about them: cheaper entry tier, but no native integration with major ERPs. Our actual differentiation, as far as we can honestly claim it: native ERP integrations and same-day onboarding. Deals we've won or lost against them before: won 3 of the last 4 head-to-head deals where ERP integration mattered. Structure it for a rep to glance at mid-call, not read as a full document — every section needs to be scannable in the ten seconds before unmuting, not comprehensive: 1. WHEN THEY COME UP — the one-sentence version of how prospects typically describe Ledgerline when comparing us, based on won 3 of the last 4 head-to-head deals where ERP integration mattered if given; if no deal history exists yet, base this on cheaper entry tier, but no native integration with major ERPs instead and say plainly that it's an inference rather than a pattern observed across real deals. 2. WHERE THEY'RE ACTUALLY STRONGER — be honest here, at least one real point, using cheaper entry tier, but no native integration with major ERPs. A battle card that claims we win everywhere gets a rep caught flat-footed the first time a prospect raises the exact strength the card never mentioned, and the rep's credibility takes the hit, not the card's. 3. WHERE WE WIN — native ERP integrations and same-day onboarding, stated as buyer-facing outcomes a prospect would actually care about, not internal feature comparisons a prospect has no way to evaluate on their own. 4. LANDMINES TO AVOID SETTING OFF — specific questions or claims that would let Ledgerline counter easily if the rep raises them unprompted; a battle card should tell a rep what not to say just as much as what to say. 5. THREE PROOF POINTS — customer outcomes or specifics the rep could cite live on a call, each explicitly flagged as either a verified fact confirmed with a named account, or an illustrative example that shouldn't be cited as a specific customer result if pressed for the account name. 6. ONE QUESTION TO ASK — a question that surfaces whether the prospect's actual stated priorities favor us over Ledgerline, phrased so it never requires badmouthing the competitor to work. LENGTH DISCIPLINE Each section capped at two to three lines. If a section can't be made scannable at that length without losing the substance that actually matters, that's a signal the underlying information is too thin or too complicated for this battle card format — say so rather than quietly writing a longer section that breaks the scannability the whole card depends on. HONESTY CHECK If won 3 of the last 4 head-to-head deals where ERP integration mattered is empty or too thin to support a real pattern, say so in section 1 rather than writing a confident-sounding "how prospects typically describe them" line based on no actual deal evidence — an inferred pattern presented as an observed one is the fastest way for this card to mislead the exact rep relying on it to be accurate mid-call.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The explicit "where they're actually stronger" section targets the single biggest structural failure mode of internally-written battle cards: content that only lists advantages reads as marketing collateral rather than intelligence, and a rep who trusts a one-sided card gets genuinely blindsided the first time a prospect raises the exact weakness the card conveniently never mentioned, at which point the rep looks unprepared in front of the buyer through no fault of their own preparation. Flagging every proof point as illustrative versus verified stops the model from inventing a customer outcome that sounds precisely like a citable statistic but isn't something the rep can actually back up if a prospect asks for the account name on a live call — the difference between those two categories is invisible in the text of the proof point itself, which is exactly why it has to be labeled rather than left for the rep to discover the hard way mid-conversation. The line-capped, scannable format matches how battle cards actually get used in the field rather than how they get written in an enablement meeting: glanced at in the seconds before unmuting on a call, not read start to finish like a memo, and a card that fails the ten-second scan test in practice gets ignored entirely regardless of how accurate its content is, making length discipline a functional requirement, not a stylistic preference. The honesty check on {{past_deal_history}} matters because a pattern inferred from zero or one data point and a pattern actually observed across several real deals are different claims wearing the same confident sentence structure, and a rep repeating an inferred pattern to a prospect as if it were a track record is making a claim the data doesn't actually support — which is a small credibility risk in the moment but a compounding one if the prospect happens to reference a customer who had the opposite experience.
Verified against
ChatGPT GPT-5.1 · 2026-08-05
Claude Sonnet 4.5 · 2026-08-06
Changelog
- 2026-08-06 — Initial publish, verified against ChatGPT (GPT-5.1) and Claude (Sonnet 4.5).
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
