ingridleiria Competitive Battlecard
A battlecard is read in the thirty seconds before a call, or under the table during one. That single fact determines everything about how it should be built, and it is the fact most battlecards ignore. What gets produced instead is a twelve-page competitive analysis that is accurate, current on the day it was written, and never opened again, because nobody reads twelve pages while a prospect is talking.
The failure that costs money is different and worse. A battlecard built from the competitor's own marketing and the sales team's folklore will contain at least one claim that is out of date or was never true. The first informed prospect who hears it corrects the salesperson in the meeting, and from that moment the salesperson is the less-informed person in the conversation about their own market. That is a very hard position to recover, and it costs more than saying nothing about the competitor at all. A false statement about a named competitor made to a prospect also carries legal exposure in most jurisdictions, which is a separate reason the evidence discipline is not optional.
The third failure is the card that claims a win on every row. Every experienced buyer has seen a comparison table where one vendor wins everything, and every one of them discounts it entirely. A card that concedes three things and is right about the fourth is believed; a card that concedes nothing is not read.
When to use this, and when not to
Use it when a named competitor appears repeatedly in the pipeline, when a salesperson asks what to say about a rival, when win rates against one company are visibly worse than the rest, when a new entrant starts appearing in evaluations, or when a competitor makes a move that changes the conversation such as a funding round, an acquisition, a repricing or a public outage.
Do not use it to size a market or map a category, which is market-research. Do not use it for a one-off, deal-specific refresh two days before a decision, which is external-insights and should take hours rather than the week this takes. Do not use it to choose a supplier for your own organisation, which is vendor-evaluation and asks a different set of questions with different weights.
Do not build one for a competitor that has appeared in two deals. The evidence base will not support it, the card will be built from marketing material, and it will teach the sales team wrong things confidently. Below roughly eight to ten encounters, write a short watch note instead: who they are, where they showed up, what happened, and what to collect next time.
What you need before starting
Win and loss records from real deals where this competitor was present. The last ten to fifteen, from the customer system and from actual conversations with the people who ran them. This is the most valuable input and the one most often skipped, because it takes days and the marketing material takes an afternoon. Missing: interview the three salespeople with the most encounters for forty minutes each before writing anything, and mark the card as folklore-weighted until the records exist. Never substitute the sales team's collective impression for the record; the impression over-weights the most recent loss and the most articulate loser.
The competitor's own material, captured with dates. Positioning statements, pricing and packaging pages, documentation, release notes, case studies, job postings, and investor material where they are public. Missing: this is rarely missing, but the discipline is to save a dated copy rather than a link, because vendors change pricing pages and a claim you cannot evidence is a claim you must remove.
Third-party evidence. Review sites read for patterns rather than for the extremes, analyst coverage, community forums where their users actually talk, and the trade press. Missing: rely more heavily on win and loss interviews, and say in the card that the weakness claims rest on internal evidence only.
What the buyer actually decides on. The three or four criteria that genuinely move a decision in your segment, which are usually not the criteria either vendor's marketing emphasises. Missing: extract them from the win and loss interviews, since the reason given for a decision is the closest thing available to the criteria.
Access to the competitor's product, where legitimately available. A public trial, a sandbox, a recorded demonstration, or a customer who will show you. Missing: rely on documentation and release notes, and label capability claims as documented rather than observed. Do not misrepresent yourself to obtain access; the reputational cost when it surfaces exceeds anything the card could be worth.
A named owner and a review cadence. Missing: propose one person and a quarterly review before writing. An unowned battlecard is worse than none within two quarters, because it will still look authoritative while being wrong.
The method
Set the scope before gathering. One competitor, one segment, one buyer type. A card covering three competitors is a document. A card covering all segments will be wrong in each, because the same competitor is strong in one and weak in another and the averaged claim is true nowhere.
Pull the deal records first, not the marketing. Order matters. Reading the competitor's positioning first anchors you on their framing and you will spend the rest of the work arguing with it. Start with what happened in real deals: outcome, deal size, segment, why the buyer said they chose, price seen, terms seen, and what the competitor said about you.
Interview the losses harder than the wins. Wins are pleasant and uninformative; people attribute their own wins to their own strengths. Losses carry the information. Where a lost buyer will speak, ask what nearly changed their mind, which surfaces the criterion that was closest to decisive. The rule for what counts as a reason: what the buyer said, not what the salesperson concluded.
Code the encounters into patterns. Cluster by segment, deal size, buying trigger and the stated reason. A pattern needs three or more instances before it goes on the card. Two instances is an anecdote, and an anecdote on a battlecard is repeated by twelve salespeople until it becomes a belief.
Write the strengths before the weaknesses. This ordering is deliberate. Forcing the honest strengths onto the page first prevents the card becoming an argument, and it is what makes the weaknesses believed. Include at least one place where the competitor genuinely beats you and the buyer is right to prefer them. If you cannot find one, the research is not finished.
Evidence every weakness, or cut it. Each weakness gets a source and a date: a documented limitation, a pattern in reviews with the count, an architectural consequence, a support model, a pricing structure with a knock-on effect, or a gap visible in what they are hiring for. The judgement call is how much evidence is enough; the rule is that a salesperson must be able to answer "how do you know that" in one sentence without leaving the call. Where you believe something and cannot evidence it, either move it to a discovery question, which is safe and often more effective, or leave it out.
Write the losing patterns as qualification guidance, not as excuses. Name the deal shapes where they win and say what to do: qualify out early, bring in a specialist, change the entry point, or compete on a different criterion. A salesperson who recognises an unwinnable deal in week one and redeploys the time has been given something more valuable than a talking point.
Turn their attack lines into short factual answers. Collect what the competitor actually says about you in deals, from the loss interviews. For each, one short answer and one proof point that can be shown, not described. The rule: answer the substance in the first sentence, then the proof. A response that opens by characterising the competitor as unfair loses the room.
Write discovery questions rather than attacks. Five to eight questions that surface the competitor's real limits without naming them, phrased so a neutral buyer would ask them anyway. "Who administers this day to day, and what happens when they are on leave" is a question a buyer should ask. It is also the question that exposes an administration-heavy product. Questions are more effective than claims because the buyer reaches the conclusion themselves and therefore owns it.
Record pricing as observed, with dates and deal context. List prices where public, the discounts actually seen with the deal size and date, contract length, and the structural consequences of their model such as per-seat costs at scale or charges for a capability you include. Never present a competitor price as certain when it came from one deal; write "seen at" with the date.
Compress to one page and put the appendix behind it. The one page is the deliverable. If a section will not fit, it is either not decisive or it belongs in the appendix. The comparison matrix, the raw deal records and the source list live behind the card.
Date it, name the owner, and set the review. Version and date visibly at the top, because a salesperson must be able to tell in one glance how old the information is. Review quarterly, and immediately after any notable competitor event or any quarter's win and loss review.
What goes on the one page
Eleven blocks, in this order, because it is the order a salesperson needs them under pressure.
Snapshot: founded, size, funding, geography, target customer, and their positioning in one sentence, in their own words. Where they are strong, three to five honest points. Where they are weak, three to five evidenced points with the source. How we win, two or three differentiators stated as buyer outcomes with proof. Where we lose, the deal patterns and what to do. Their objections and our answers. Discovery questions. Traps, meaning the evaluation criteria worth planting early because they are real requirements this buyer will have and this competitor struggles with. Pricing and packaging. Recent moves, the last three to five dated events and what each signals. Sources, owner and date.
A trap is only legitimate when the criterion is a genuine requirement for that buyer. Planting a criterion the buyer does not need in order to disadvantage a rival is manipulation, it is usually detected, and it produces a customer who bought the wrong thing.
Worked example
Situation. A workforce scheduling software company selling to mid-market healthcare operators, roughly 60 to 400 sites. Over three quarters, win rate against one competitor had fallen from 55 percent to 38 percent across 29 competitive deals, eleven won and eighteen lost. The sales team's explanation, held with confidence, was that the competitor was buying deals with heavy discounts. The existing battlecard was fourteen months old and four pages long.
Task. Produce a card the team would actually use in calls, and establish whether the discount theory was true, within three weeks.
Action. The deal records came first. Twenty-nine encounters, of which pricing data was recoverable for 21. The discount theory did not survive contact: median discount from list was 18 percent for the competitor and 22 percent for the company itself. In the eight losses where a price comparison existed, the competitor was more expensive in five.
Interviews with eleven losses found the real pattern, and it was not price. The competitor had shipped a rota-swap approval workflow for shift-level staff nine months earlier. In organisations where scheduling was devolved to site managers rather than run centrally, that feature moved from convenient to decisive, because it removed a daily task from about forty people rather than from one. Twelve of the eighteen losses were devolved-scheduling operators. Of the eleven wins with recoverable detail, nine were centralised.
The wrong turn: the first draft treated this as a feature gap and wrote a competitive response section arguing that the company's bulk-edit tools achieved the same outcome. Two salespeople tested that line in live calls and both reported the same result, that the buyer had already seen the competitor's version and the comparison invited a demonstration the company lost. The line was cut. What replaced it was a qualification rule and a discovery question. The qualification rule: where scheduling is devolved to site managers and rota-swap volume is high, expect to lose on the current product and either compete on total cost across the contract or qualify out in week one. The discovery question, which worked: "How many people currently need to approve a shift swap, and how long does one take end to end?" In centralised operators the answer made the company's central controls look strong. In devolved operators it surfaced the gap early enough for the salesperson to redeploy rather than spend eight weeks losing.
The strengths section conceded the rota-swap workflow explicitly, and named two more places the competitor was better: mobile experience and a faster implementation for single-site pilots. Three weaknesses were evidenced and one was cut. The cut one was a widely repeated claim inside the sales team that the competitor's reporting could not handle multi-entity structures. Two hours in their public documentation showed it could, and had for over a year. That claim had been in the old battlecard and had certainly been said to prospects.
Pricing was recorded as observed, from 21 deals, with the note that it derived from the company's own deal records and might be unrepresentative of deals it never saw.
Result. The card went to one page with a four-page appendix. Over the following two quarters, win rate against the competitor recovered to 44 percent, but the more useful number was that time spent in eventually-lost deals with devolved operators fell by about a third, because the discovery question was surfacing the pattern in the first meeting rather than the sixth.
The honest limits: the win rate recovery cannot be attributed cleanly to the card, since the product team also shipped a partial swap workflow in that period. What can be attributed is the removal of a false claim that had been in circulation for over a year, and the qualification rule, which salespeople reported using by name.
A second scenario, where it goes differently
A competitor with no meaningful evidence base: a well-funded entrant that has appeared in four deals in two months, with no public pricing, no review history and no lost deals to interview because none have closed either way.
The method cannot run. Building a card here would mean building it from their marketing site and a funding announcement, which is precisely the folklore this skill exists to prevent, and the card would carry authority it has not earned.
What replaces it is a watch note, half a page, kept by the same owner: who they are, the dated facts that are actually known such as funding, headcount trajectory from job postings and named hires, where they have appeared and against which deal shapes, what they said in the room as reported by the salespeople present, and a specific collection list of the three things to capture in the next encounter. The note explicitly says it is not a battlecard and carries no positioning guidance.
What changed: the evidence threshold, not the standard. The card is written when the evidence exists, and the interim artefact is honest about being an interim artefact. The most damaging version of this work is a confident card built on four data points, because the sales team cannot tell the difference between a card built on 29 encounters and one built on four.
Output
One page, plus appendix.
[COMPETITOR NAME] Segment: [x] Buyer: [y]
v[n] Updated [date] Owner [name] Next review [date]
SNAPSHOT [founded, size, funding, geography, target customer]
Their positioning, their words: "[one sentence]"
WHERE THEY ARE STRONG [3 to 5, including where they beat us]
WHERE THEY ARE WEAK [3 to 5, each with source and date]
HOW WE WIN [2 to 3 buyer outcomes, each with a shown proof]
WHERE WE LOSE [deal patterns, and what to do about each]
Their objections and our answers:
| What they say about us | Our answer, first sentence | Proof to show |
Discovery questions:
| Question | What a revealing answer sounds like | What it opens |
Pricing, as observed:
| Tier or model | List, if public | Seen at | Deal size | Date | Structural consequence |
Recent moves:
| Date | Event | Source | What it signals |
Appendix. The comparison matrix, scored with evidence, covering only capabilities that move a decision in this segment, and showing the rows lost. The deal record table behind the patterns. The dated source list.
Failure modes
Built from marketing and folklore. Recognise it when no claim on the card has a date next to it. Fix by pulling deal records before anything else, and by deleting any claim nobody can source in one sentence.
Wins every row. Recognise it in the comparison matrix. Buyers discount it entirely and so do experienced salespeople. Fix by finding and stating the places the competitor is genuinely better; if none can be found, the research is incomplete rather than the competitor weak.
A claim that was true eighteen months ago. The most expensive single failure, because it is delivered with confidence and corrected by the prospect. Fix with dated sources and a quarterly review, and re-verify any capability claim about a competitor that ships frequently.
Four pages. Recognise it because nobody in the field can quote anything from it. Fix by moving everything except the eleven blocks into the appendix.
The anecdote promoted to a pattern. Recognise it when a claim traces to one memorable deal, usually recent and usually lost by a senior person. Apply the three-instance rule.
Price theory substituting for analysis. Recognise it when the team's explanation for losses is discounting and nobody has compared the actual figures. Check it early; it is frequently wrong and it conceals the real cause for months.
A trap that is not a real requirement. Recognise it when the planted criterion is one the buyer had not raised and does not need. Buyers detect this, and the salesperson loses standing as an adviser. Cut it.
Unowned and undated. Recognise it because nobody can say when it was last checked. An out-of-date card that looks current is more dangerous than no card.
Edge cases
Fewer than eight encounters. Write a watch note, not a card, and say so on the note.
The competitor is a customer, a partner or a potential acquirer. All three happen. Keep the card factual, keep it internal, and avoid characterisation entirely. Anything you would not want read aloud in a partnership meeting does not go on the page. partnership-assessment handles the relationship side.
A public competitor with regulated disclosures. Their filings are the best evidence available for scale, growth and unit economics. Cite them with the filing date. Do not infer product direction from a financial statement.
They win on something you will never match. Say so plainly on the card and give the qualification rule. Sales teams respect a card that tells them where not to go far more than one that pretends everywhere is winnable.
An incumbent rather than a vendor, or a decision to build in-house. The competitor is inertia or an internal team. The strengths section is switching cost and political ownership. The discovery questions turn to the total cost of the current approach and who maintains it.
A public incident at the competitor. Record it dated, and give the guidance that nobody uses it as a talking point. Buyers respond badly to a rival's outage being sold against, and the same could happen to you. Where a buyer raises it, the honest answer is a factual statement of your own record.
Quality bar
- Every weakness claim carries a source and a date, and a salesperson can answer "how do you know that" in one sentence.
- At least one place where the competitor genuinely beats us appears on the page.
- The win and loss patterns rest on three or more real deals each, drawn from records and interviews rather than recollection.
- Where we lose is stated as qualification guidance with an action, not as an excuse.
- Pricing is labelled as observed, with the date and deal context, never asserted as certain.
- The card is one page and every trap is a criterion the buyer genuinely needs.
- The version, date, owner and next review are visible at the top.
Adapting this to your context
The thresholds here come from business software sold to mid-market operators, with a customer system that records which competitor was in the deal. They are defaults.
- Eight to ten encounters. The line below which you write a watch note instead of a card. In enterprise sales running six deals a year the line is lower and the loss interviews carry more weight; in high-volume transactional sales, thirty encounters is a better floor.
- The three-instance pattern rule. Sound where deals are broadly comparable. Where deal shapes vary widely, require three instances inside one segment rather than three across the whole pipeline, or the pattern is an artefact of mixing.
- Quarterly review. Set for a competitor shipping every few weeks. A rival with an annual release and a published price list can be reviewed twice a year; one in a funding or acquisition cycle is reviewed on the event.
- Naming a competitor at all. Comparative claims are restricted, or require substantiation on request, in several jurisdictions, and some regulated sectors prohibit them outright. Check what your market allows before the card leaves the building.
- What not to change. Every weakness carries a dated source a salesperson can defend in one sentence, and at least one place the competitor genuinely beats you appears on the page.
Related skills
external-insights gathers and dates the external evidence this card rests on, and does the fast deal-specific refresh when a competitor turns up in a live opportunity. pipeline-deep-dive supplies the win and loss records that make sections four, five and six real, and is the natural quarterly trigger to review the card. customer-interview-synthesis is the method for coding the loss interviews into patterns rather than anecdotes. market-research sizes and maps the category this card sits inside. outreach-email, tailored-client-deck and proposal-writer consume the differentiators and the proof points. vendor-evaluation is the mirror image, applied when your own organisation is the buyer. partnership-assessment takes over where the competitor is also a partner.