How to write a competitor battlecard salespeople actually use
A one page format, the five sections that matter, what to leave out, and how to stop it going stale.
Most battlecards are written by marketing, admired once, and never opened again. The ones that get used share a shape, and it is smaller than what most teams produce.
The format
One screen. Five sections. Date at the top.
Competitor · Updated [date]
1. When you will hear this name. The deal shape where they show up. "Companies with 20 to 100 staff who already have an in house analyst." Sellers need to recognise the situation before they need the arguments.
2. Why we win. Three reasons, each one sentence, each with proof. Not adjectives. "Our setup is same day; their own documentation quotes a one week implementation."
3. Why we lose. Two or three, honestly. Sellers already know these. Writing them down means nobody is surprised on a call, and it tells the seller when to walk away, which saves more money than winning the marginal deal.
4. Their three lines, and your answers. What their sellers say about you, and the response. Word for word, because a seller under pressure will use exactly what is written.
5. Questions to ask. Three questions that surface the difference without mentioning the competitor at all. "How quickly do you need this live?" is worth more than any claim about your setup speed.
That is the card. Anything else belongs in a linked document that nobody will open, which is fine, as long as the card stays one screen.
Section 4 is the one that matters
Everything else is context. Section 4 is what a seller uses live.
Getting it right needs input you cannot generate internally. Two sources.
Ask your sellers what they hear. They know, and nobody has asked them.
Ask lost prospects. In win-loss interviews, ask what the competitor said about you. People repeat it happily, and their phrasing is more precise than your guess.
Write the answer as a real sentence a person would say out loud, not a bullet point of positioning language. Read it aloud before you publish it. If it sounds like a brochure, rewrite it.
What to leave out
Feature matrices. You will tick your boxes generously and theirs meanly, and the buyer will know. If they check one row and you were wrong, everything else on the card is discredited.
Anything you cannot support. Attacks fail the first time a buyer checks. Only write facts you can point at.
Estimated revenue, headcount, funding. Interesting to executives, useless in a call.
Long context. Their history, their investors, their strategy. Not on the card.
Keeping it from going stale
This is where battlecards actually fail, and the failure is worse than not having one, because sellers quote stale information to buyers who know better.
Date it visibly. Top of the card. A seller who sees a date eight months old will treat it appropriately.
Name an owner. One person, not a team.
Set a trigger, not just a schedule. Quarterly review, plus immediate update when: they change pricing, they launch something, or the same new objection comes up twice.
That last trigger is the useful one. When two sellers report the same unfamiliar objection in a fortnight, something changed at the competitor and the card is already out of date.
Where the facts come from
Section 1 and 3 come from win-loss conversations. Sections 2 and 4 come from sellers and lost prospects. The factual details, their price, their limits, their claims, come from watching them.
That last part is the bit that decays. Their pricing page changes, they launch a feature, they change their headline claim, and nobody notices for a quarter.
The manual version is a monthly check: screenshot their pricing, diff their sitemap, read their new reviews, look at what their ads are saying. Thirty minutes per competitor.
The automated version is a monitoring tool. Enterprise platforms like Klue, Kompyte and Crayon build battlecard maintenance into the product, which is much of what you are paying for and the reason they are priced for companies with large sales teams. At the other end, our own Figo watches pages, pricing, ads, content, rankings, reviews and AI answers weekly for $49 a month and writes a Monday summary, which keeps the facts current, but you write the card yourself.
Either way, the writing is a human job. A generated battlecard reads like one, and sellers can tell.
Questions people ask
How long should a battlecard be?
One screen. A seller reads it in the ninety seconds before a call, and anything requiring a scroll does not get read.
Who should write it?
Marketing writes it, sales corrects it. A battlecard written without the people who use it is a document, not a tool.
How often should it be updated?
Quarterly at minimum, and immediately when a competitor changes pricing or launches something. Put the date on it so a stale one is visibly stale.
Should I name the competitor's weaknesses?
Yes, factually and only where you can support it. Unsupported attacks lose credibility the first time a buyer checks, and buyers check.
See it on your own competitors
Figo checks their ads, pages, rankings, reviews and AI answers every week, then tells you what to do in plain words. Set up in two minutes.
Keep reading