What should go on a battlecard for a team of three reps?

1. Blog
2. Demand & Channels

By Saroj Jha · November 2, 2026 · 14 min read

A three-rep team needs one card for each alternative its buyers actually weigh, and that list nearly always includes doing nothing and building it in-house as well as named rivals. Each card says, in the buyer’s words, where that alternative genuinely wins and where it loses, gives two or three questions that let the buyer discover the weakness for themselves, lists proof a rep can show on a call, and records the source and check date for every fact. Keep it to one page, check every claim against the standard the FTC applies to comparative advertising, and rewrite the cards each quarter from what your won and lost deals tell you.

Buyers now arrive with a comparison already in hand, and the rep’s job is to be the more reliable source in the room.

Running case: Example Co. Fictional

Example Co. is a fictional seed-stage company selling contract-review software to in-house legal teams at companies with 200 to 2,000 employees. It has three salespeople: the cofounder and two account executives. In the last two quarters it closed out 60 opportunities: 25 won, 21 lost to another option and 14 that ended with no decision. Its buyers compare it with an established legal suite, a cheaper point tool, building something in-house with the IT team, and carrying on with email, spreadsheets and outside counsel. Every number about Example Co. in this article is invented for illustration.

Why does a team of three need battlecards at all?

A small team does not need a competitive intelligence function, but it does need the three reps to say the same true thing about each alternative. Without a card, each rep builds a private version from memory and the last lost deal, and two of those versions are usually out of date. The card is a shared, dated record of what you know, not a sales script.

The Gartner figures in the key band explain why this matters more than it did. In a survey of 645 B2B buyers in August and September 2025, 45% had used generative AI in a recent purchase, mostly to gather information on vendors and products, and 69% said they prefer to check AI-generated insights with a sales rep^[1]. In the same survey, 49% said they are more likely to encounter misleading information from a sales rep, against 51% who said so of generative AI^[1]. A buyer who walks in with an AI-written comparison is testing whether you are more accurate than the chatbot, and a card with sourced, dated facts is how three reps pass that test.

The research behind the cards, who the alternatives are and how buyers choose between them, is a separate job covered in Competitive analysis for startups. This article starts where that one ends: you know the alternatives, and now you need something a rep can use in a live call.

Example Co., step 1 Fictional

The cofounder asks each rep to write down, without notes, the top three reasons buyers pick the legal suite. The three lists share one reason out of nine. One rep is still quoting the suite’s pricing from a year ago.

Do this

Ask each of your reps the same question about your most common rival and compare the answers before writing any card.

Which alternatives get a card?

Write one card for each alternative that appears as the main option in at least a handful of your recent deals, and no card for rivals you only read about. For most startups that is three to five cards. My rule of thumb, not a benchmark: if an alternative was the main option in fewer than 5% of the deals you closed in the last two quarters (won, lost or ended with no decision), it gets a line on a “minor rivals” sheet, not a card.

Two cards are missing from most startup sets. The first is doing nothing: the buyer keeps the spreadsheet, the inbox, the outside agency or the manual process. In my view it is often the alternative a startup loses to most, and it needs its own card because the arguments are different: the question is “why change at all”, not “why you rather than them”.

The second is building in-house: the buyer’s own engineers, or now a team with an AI model and a weekend, promise to build it. That card is about total cost, maintenance and time to a result, not features.

Pull the list from your own records, not from your assumptions. Your CRM’s closed-lost reason, the notes on deals that went quiet, and the question “what would you have done if we did not exist?” asked of won customers will give you the real set. The Win-Loss Analysis Playbook covers how to collect and code those answers properly.

Example Co., step 2 Fictional

Across the 60 closed opportunities, the main alternative was the legal suite in 22, the point tool in 15, carrying on as before in 18 and an in-house build in 5. The in-house build, at 5 of 60 (8.3%), clears the 5% line, so Example Co. writes four cards. Two other vendors that the founders worry about appear in none of the 60 deals and go on the minor-rivals sheet.

Do this

Count the main alternative in each of your last 30 to 60 closed deals and write cards only for the ones that clear your threshold.

What goes on each card?

Every card has the same six parts, so a rep can find the right line in seconds during a call. I keep each part to a few lines, because a card that runs past one page will not be read mid-call.

Where they win. The alternative’s real strengths, stated in terms the buyer would use. “Already approved by procurement and bundled with tools you pay for” is a buyer’s reason; “broad feature set” is not. Being honest here is what makes the rest of the card credible to your own reps.

Where they lose. The specific situations where the alternative is a worse choice, again in the buyer’s terms and tied to an outcome the buyer cares about: time, cost, risk or effort. “Your team will still be reading every clause by hand” is a loss in buyer terms; “no AI” is a feature gap.

Landmines and the questions that reveal them. A landmine is a weakness the buyer can discover for themselves when they ask the alternative a fair question. The rep’s job is to supply the question, not the accusation. More on this in the next section.

Proof you can show. Evidence a rep can put on screen or send after the call: a customer result with permission to use it, a public document, a live demo of a specific task, a security report. Claims without proof go to the bottom of the card or come off it.

Facts with sources and dates. Prices, integrations, contract terms, certifications: each with where it came from and the date someone last checked it.

When to walk away. The conditions under which the alternative really is the better fit for this buyer. A rep who knows when to stop saves the team weeks on deals it should not be chasing.

Example Co., step 3 Fictional

The legal-suite card opens with where the suite wins: one vendor for the whole legal department, already in procurement, and a deep clause library. Where it loses: a long implementation before the first contract is reviewed, and per-seat pricing that makes occasional users expensive. Walk away when the buyer is replacing the whole legal stack this year, because the suite’s bundle wins that deal.

Do this

Draft the “where they win” section of each card first, and get a rep who lost to that alternative to approve it.

Which questions reveal the landmines?

A landmine question is one a buyer can reasonably ask every option, whose honest answer favours you. It works because the buyer hears the weakness from the rival or from their own team, not from you. A question that only makes sense as an attack on one rival will be recognised as one.

Good landmine questions share three traits, in my view. They are about the buyer’s outcome (“how long until your first contract is reviewed in the tool?”), not about a feature; they have a checkable answer; and you would be happy for the buyer to ask them of you. Two or three per card is enough, and each one sits next to the answer the rep expects the alternative to give and the proof that shows your answer.

The doing-nothing card needs different questions, because there is no rival to ask. These are questions about the cost of staying put: how many hours the current process takes, what happened the last time it failed, and who carries the risk if it fails again. The discovery call framework covers how to ask those questions without turning the call into a pitch, so I will not repeat it here.

The build-in-house card’s questions are about everything after the first version: who maintains it, who answers when it breaks during a deal, and what the engineers would otherwise be building. A buyer who answers those honestly often finds the build is cheaper only for the first month.

Example Co., step 4 Fictional

Landmine questions on the legal-suite card: “How many weeks from signature to the first contract reviewed?”, “What does a reviewer who looks at five contracts a month pay per year?” and “Can we see it working on one of our own contract types before we sign?” On the do-nothing card: “How many outside-counsel hours did contract review cost last quarter?” and “What happened with the last contract that slipped through?”

Do this

For each landmine question, write the answer you would give yourself; drop any question you would not want a buyer to ask you.

How do you keep every claim true and current?

Treat every comparative line on a card as if it were in an advertisement, because to the buyer it is one. The FTC’s guidance for businesses is that advertising must be truthful and non-deceptive, that advertisers must have evidence to back up their claims before the ad runs, and that “comparative advertising is legal as long as it is truthful”^[2]. The Commission’s policy statement on comparative advertising says it evaluates comparative claims in the same way as other advertising, encourages naming competitors, and requires clarity and, where needed, disclosure so buyers are not misled^[3].

The FTC’s substantiation policy adds the part that catches sales teams: firms must have substantiation before disseminating a claim, and a claim that implies a level of proof must be backed by that level of proof^[4]. A rep who says “they take three months to implement” is implying they know; if the source is a lost buyer’s grumble from a year ago, the card should say so or drop the line.

Three habits keep a small team safe and credible. Every fact about a rival carries a source and the date last checked; anything older than my rule of thumb of 90 days (not a benchmark) is re-checked before it is used. Claims are about things that can be verified, such as published prices, documented integrations or a timed demo, not about a rival’s intentions or quality in general. And reps never present a rival’s private information, such as a leaked price sheet, as fact.

This is orientation, not legal advice.

Example Co., step 5 Fictional

Example Co. audits its four cards. “Suite pricing starts at $45 per seat per month” is checked against the suite’s public pricing page and dated. “The point tool loses data” came from one lost deal and cannot be supported, so it comes off. “Their implementation takes months” is rewritten as the landmine question “how many weeks from signature to first contract reviewed?”, because Example Co. cannot prove the rival’s timeline but can let the buyer ask.

Do this

Run every comparative claim on your cards through the Claims & Proof Checker and delete any you cannot back today.

How should win-loss change the cards?

A card is a hypothesis until your deal record supports it. Each quarter, record the main alternative and the outcome for every closed opportunity, calculate your win rate against each alternative, and read the reasons in the deals you lost. Then change the cards: strengthen what worked, cut what did not, and add the objection that kept coming up.

Small numbers deserve humility. With a few dozen deals per alternative, a win rate can move by 20 points by chance, so show it with a range. I use a 90% interval, which is wide on purpose at this sample size; a card should say “hypothesis” until the range is narrow enough to act on. The Win-Loss Analyzer codes the reasons and calculates those ranges for you.

Keep the update cheap. For three reps, a monthly 30-minute review of the cards, with each rep bringing one deal where the card helped or failed, is enough between quarterly rewrites. In my view, the best single test of a card is whether a rep reached for it in the last month; a card nobody used is either wrong or in the wrong place.

Example Co., step 6 Fictional

Win rates by main alternative: legal suite 8 of 22 = 36.4% (90% range 22% to 54%); point tool 11 of 15 = 73.3% (52% to 87%); in-house build 2 of 5 = 40.0% (14% to 73%); doing nothing 4 of 18 = 22.2% (10% to 41%). Overall 25 of 60 = 41.7%. Of the 14 losses to the suite, 9 named the bundle with existing tools: 9 / 14 = 64.3%. The team rewrites the suite card around occasional-user pricing and the landmine questions, and wins 7 of 12 against the suite the next quarter, 58.3% (36% to 78%), a rise whose range still overlaps the old one, so the card stays labelled a hypothesis.

Do this

Add a required “main alternative” field to closed deals in your CRM this week, so next quarter’s card review has data.

What changes by stage?

• Pre-seed — The founder is the only seller and does not need cards yet. Keep a one-page note per alternative with dated facts, so the first hire inherits something accurate.
• Seed — Three reps is the point where cards pay for themselves. Write three to five, including doing nothing and building in-house, and add the “main alternative” field to every closed deal.
• Series A–B — Win-loss volumes are large enough to trust the ranges. Give one person ownership of the cards, add cards for segment-specific rivals, and connect the card review to product and pricing decisions.

Free tool

Build a card for each alternative

The Battlecard Builder turns your research into one-page cards with where you win, where they win, dated competitor facts, objections, reframing questions, walk-away conditions and a head-to-head record with a 90% range, labelling the card a hypothesis until you have deals to back it.

If you have not mapped your alternatives yet, start with Competitive analysis for startups and the Competitor Research Grid. For the questions that sit behind the doing-nothing card, read the discovery call framework, and for a structured loss-review process use the Win-Loss Analysis Playbook.

If you want me to look at your own funnel, positioning and channels, my Growth Audit ends with three prioritized fixes and a 30-day action plan.