Somewhere in the middle of a call this week, a buyer is going to say a competitor’s name, and one of your reps is going to open a battlecard.
There will be a line on it. It will be a good line, written by somebody who thought hard about it.
What happens next is not decided by the line. It is decided by whether the rep says it like a person who knows it is true, because the buyer on the other end has already sat that competitor’s demo and is holding a version of the story your card has never heard.
Mystery Demo goes and sits that demo. We book your competitors’ sales processes as a real buyer, run each one end to end, and hand back the recordings, the transcripts, the email threads and our read on what all of it means, so the lines on your card can come from the competitor’s own mouth with a timestamp underneath.
That is the difference between a card a rep opens in a live call and one they skim once and quietly stop trusting. A battlecard is only as strong as the worst-sourced line on it, and most of them are sourced from the one place the competitor controls completely.
Every Template Hands You the Fields and Leaves You the Sourcing
Run the search you would run. What comes back is field lists, and they are good field lists.
Product Marketing Alliance publishes eight components, from the company overview through to pricing and testimonials. TinyCommand publishes a seven-row table: overview, why we win, their strengths, their weaknesses, objection handling, trap-setting questions, proof points. Customer Marketing Alliance publishes five blocks, built around the competitor’s claims and how to address each one.
Nothing is wrong with any of that. Those are the rows a rep needs.
Above the rows, both of them tell you to do the hard part. TinyCommand says to “fill each section with real, sourced detail,” and its template has no row for the source. Customer Marketing Alliance says to “fill in the factual sections first so every rep is working from the same accurate picture of the competitor,” and publishes no field for where the picture came from.
One of them does ask, and it earns the credit. Visualping’s template puts a Last verified column on every table and closes with an update log whose columns are Date, What changed, Source and Verified by. The page says in its own words that those two columns are what make the template different, and it is right.
Then look at what its own worked example puts in the Source cells. “Their pricing page.” “Their homepage.”
Dock, whose guide collects two dozen real cards, gets closest to the bone: “The best source is sales calls: the objections that keep repeating and the competitors that keep showing up uninvited.” The calls in that sentence are yours, which puts the competitor inside them secondhand, described by a buyer who is also trying to get a discount out of you.
So the fields are right, and the sourcing chain is one link long. Every honest template in the results points its evidence at pages the competitor wrote about itself.
The Rows That Decide Deals Are the Ones Nobody Publishes
We wanted to know how much of a card that chain can carry, so we went and counted.
Ten contract lifecycle management vendors, named before we measured and picked by a rule: standalone products listed by at least two independent category roundups on the same day. Agiloft, Conga, ContractPodAi, Docusign CLM, Icertis, Ironclad, Juro, Malbek, Sirion, SpotDraft. Three checks each, made on their own live sites on 29 August 2026.
Ten out of ten put a demo request or a sales conversation between you and the product. Seven out of ten publish a page naming a specific rival. Not one of the ten publishes a price.
The zero deserves its detail, because a zero always does. Docusign will sell you eSignature at a published per-seat rate, and its CLM product page says “Talk to an Expert.” Icertis, Sirion and ContractPodAi have no pricing page at all.
Ironclad’s is headed with pricing “shaped by your products, users, workflows, and growth,” under a button reading Get a Custom Quote. Conga’s lists Transparency among the six things its pricing prioritizes, above no figure of any kind.
Vendors in this category will, on the other hand, publish a great deal about each other.
Ironclad runs a page on Icertis with a full feature comparison, and the implementation row reads “Service-led with added fees” on one side and “Self-start or choose the package that fits you best” on the other. It is specific, it is confident, and there is no price anywhere in the table. Sirion’s page on Ironclad backs its claim about the rival’s document repository with a quote lifted from a G2 review.
So the one row your competitor is delighted to fill in for you is the row they wrote in order to beat you, and their evidence for it is a review site. Everything a buyer decides on sits on the far side of a call that only a buyer gets into.
So Somebody Has to Go Through the Funnel as the Buyer
Mystery Demo is mystery shopping for B2B SaaS, and that sentence is the whole method.
We agree the competitive set and the question list with you at kickoff, and you demo your own product to us so we can ask what a real evaluator in your market would ask. Then we build the fábula, which is the buying scenario we carry into every one of those funnels.
The fábula is rooted in truth and that is why it holds. We never impersonate a real person and we never lie about our own background, and a firm that evaluates software on behalf of the companies it works with is an ordinary thing for a vendor to hear from. The only invented element is the company we say we are evaluating for.
Six things build it: the industry, the headquarters, the team size, the funding stage, the stack they run today and what is straining it, and a decision window tight enough to earn attention. Each one is tuned so the vendor’s qualification questions get answers that sound like their own best-fit buyer. A latency-first vendor hears about real-time workloads, a compliance-first vendor hears about the regulated market we are moving into, and both are hearing something true about the scenario.
Then we go through each vendor’s full sales motion. Discovery, the demo, the technical deep dive, the pricing conversation, the follow-ups. Everything is recorded, every thread and every document is kept, and every claim a rep makes is checked afterwards against that vendor’s own public material.
A competitor costs a flat 499 euros, everything included: the scenario, the booking, every meeting, the recording, the transcript, the analysis and the delivery. One competitor takes one to two weeks. A landscape of six to fifteen takes four to eight, because the funnels run in parallel rather than in a queue.
Below is the objectives and methodology page from our public example board, where the client and all five vendors on it are invented. It is the page that fixes what the calls are for before anybody books one, and every engagement gets its own version, written from your competitive set at kickoff.
The recordings are not the product, though, and a folder of them would be raw material and somebody else’s weekend.
What Lands on the Board, and Which Layer Is Card Material
Every project is delivered as a Notion board, and each competitor gets a page of their own. On it: every recorded meeting, the cleaned transcript of each one, the full email thread from first outreach to last follow-up, and every deck, pricing sheet and scoping document they sent.
Underneath the raw layer sits the read on that vendor, and this is the part a card is built from.
Across the vendors, the same evidence is cut the other way. Comparative dimensions put every vendor in one table, a sales process scorecard grades the motion step by step, and one page says where each competitor genuinely wins.
Holding it all up is the Database page, where every recording, transcript and document resolves to a verified link. That is what makes a claim chaseable by somebody who was not in the room, which is the whole difference between a card and an opinion.
None of that is a battlecard. Every layer of it is a row on one.
The Template: Eight Rows, and What Fills Each One
One card per competitor. Eight rows, and a right-hand column that most templates do not have, because most templates have nothing to put in it.
| The row | What it has to answer | What fills it, from the board |
|---|---|---|
| Their opening | The pitch a buyer hears before you get a word in | The discovery call, where the rep chose which story to lead with |
| What they charge | The number this buyer is being quoted, not the tier list | The verbatim pricing quote, and the minute in the call it moved |
| Shipped or slide | Which capabilities exist on a screen and which exist in a sentence | The demo recording, at the moment they showed it or moved past it |
| How they come at you | The comparison a rep makes when your name comes up | The transcript passage where your name was raised, in their words |
| What they never raise | The subject that decides deals and never appears in the pitch | The agreed question list, ticked off against every call |
| Where they genuinely win | The buyer you should stop fighting for | The comparative dimension across the whole set, and the scorecard |
| The question that lands | The prompt that pulls a rep off the rehearsed path | The moment a rep retreated to a case study or a roadmap date |
| The receipt | Where this line came from, and when | The recording, the timestamp, the thread, the file |
The eighth row changes how the other seven get used. A line with a recording behind it can be checked by the rep who has to say it, by the manager who signs off the card, and by whoever inherits both next quarter.
It also enforces the only editorial rule the card needs. If a row cannot resolve to something a competitor said or showed, it does not go on the card, which is the same test an evidence grade in competitive research applies to everything else we hand over.
The card gets shorter that way, and it gets opened more, because the places it gets opened are unforgiving.
Four Moments When the Card Gets Opened
.png)
A card is not read at the start of a deal. It is opened at four points, and each one wants a different thing from it.
| When the card opens | What the buyer says | What your rep has, because somebody sat the call |
|---|---|---|
| Discovery, the name drop | “We are also looking at them.” | Their opening pitch, so the rep can ask which part of it the buyer bought rather than guess |
| The demo, the parity claim | “They showed us the same thing.” | The recording of what that vendor showed, including the step they skipped and the data they used |
| The technical review | “Can you handle a migration like ours?” | What their engineer promised on a live call, and what the follow-up email quietly narrowed |
| The finalist round | “They came in lower.” | The figure their rep quoted, the objection that moved it, and how far it moved |
Look down the middle column. Not one of those four moments is answered by a fact about the competitor’s company. All four are answered by something a person at that company said out loud to a buyer.
The fourth is where a sourced card earns its whole cost. A rep who knows the shape of the rival’s discount behavior, because somebody sat through it, stops treating “they came in lower” as a number to beat and starts treating it as a position to test.
What a Timestamp Changes in the Room
A sourced line survives the second question, and an unsourced one does not.
Unsourced, the rep has one sentence and no next one, so a buyer who pushes gets the first sentence again in a slightly different order. Sourced, there is always a next sentence, because there was a next minute of the call.
Then there is the buyer, who has also sat the competitor’s demo. A card assembled from the competitor’s website will eventually contradict what that buyer watched with their own eyes, and once that happens in front of them, nothing else on the card gets believed.
Sourcing also tells you when a card has expired, which watching a website cannot. A monitored pricing page tells you the page changed. It says nothing about the week a vendor’s reps started opening on security instead of speed, or quietly dropped one of the things competitors almost never do in a demo, and those are the changes that move deals.
Partners re-run projects one to three years later for exactly that reason, and the second board is usually the more interesting one, because it shows what moved.
The page below is where the whole set lands: one line per competitor on where they genuinely win, drawn from the calls rather than from anyone’s opinion of them. It is the row your reps need before they decide which deals to fight for, and it is the layer battlecard research built on recorded demos exists to produce.
Build the card from what you can find and you will have eight rows, all of them true, most of them true about the version of your competitor that lives on the internet. Build it from recorded demos and the same eight rows carry a sentence a rep can say, a reason to believe it, and the minute it was said.
So tell us which competitor keeps turning up in your losses. We will go through their whole sales process as the buyer they most want, then hand you the recordings, the pricing, the pitch and the eight rows already filled in. Start with one competitor and watch what your reps do with a card they can check.
.png)
