In short
- All three routes work. They fail in different places, and the failure is predictable enough to plan around.
- Building it yourself is the cheapest and the hardest, because you cannot hear your own offer from outside.
- What you pay an outsider for is the extraction. Words and design both get laid over the logic without pulling it out.
- Software organises a playbook without knowing what belongs in one, so it comes after the content rather than before it.
- The test is the same whichever route you take. Somebody who was not there makes the same call you would have made.
Building it yourself
The cheapest route, and the one that asks the most of you.
You already have everything the work runs on. The deals are yours, the people who ran them still answer the phone, and if you have just sold the company or stepped back, you have the thing this work actually needs, which is unhurried time. Nobody understands the offer better than you do, and that is also the problem. The person who understands it best is the worst placed to hear how it sounds to somebody who does not, so the in-house version tends to record the process you believe you run rather than the one your team runs on a Thursday afternoon.
But the correction is cheap and uncomfortable. Put somebody in the room who was not in the deal, give them permission to ask why after every answer, and write down the version you give on the fourth why. The first answer is the one you give on stage. And the fourth is usually the rule.
The signal that you are drifting.If a rule you have written does not describe what you actually did in the last ten deals, it is a preference rather than a rule. Test each one against closed deals, won and lost, before it goes in.
Bringing somebody in
You are buying distance. What matters is what they do with it.
So the useful question is what they start with, because that decides what you get. Start with words and you get sharper words, which still need their author in the room. Start with design and you get a better-looking deck that decorates the logic without extracting it. Both are real skills and both leave the dependence exactly where it was, because the explanation still cannot travel without the person who wrote it.
But what is worth paying for sits underneath both of those. Structured interviews across your own won, lost and awkward deals, with the rules pulled out and named before anything gets written or drawn. That is slower to start and it is the part that transfers, because a rule with its exception attached survives being handed to somebody who was not there. A sentence does not.
And the same logic applies to a workshop. A day in a room is good at producing agreement and bad at making the agreement outlast the week, unless something leaves the room that people can point at afterwards.
Buying software
It organises a playbook without deciding what belongs in one.
Templates, version history, somewhere for the team to comment, a record of what gets opened, permissions so the pricing guardrails are not editable by everybody. All of that is genuinely useful, and all of it sits downstream of the hard part. So buy the tool before the extraction and you will fill it with stage names and a metrics grid, and the thing will still not tell anybody which rule applies when.
So the moment software earns its place is when more than a handful of people use the playbook and somebody has to answer who changed what, and when. Before that, a shared document does the job and costs nothing.
Scroll the table sideways to see every column.
| Approach | What it is good at | Where it breaks | Worth it when |
|---|---|---|---|
| Build it in-house | You already own the deals, the people and the context | You cannot hear your own offer from outside, so it records the process you meant to run | You have time, recent deals, and somebody willing to interrogate you |
| Bring somebody in | Distance, and a method for getting judgment out of a head | If it starts with words or design instead of extraction, the dependence survives | The offer is technical, the cycle is long, or the argument has only ever come out of one mouth |
| Buy software | Storage, versions, permissions, and a record of what gets opened | It organises a playbook without knowing what belongs in one | The content exists and more than a handful of people need to use it |
What are you actually buying from an outside build?
The part of the sale that has never been written down.
A person who had just sold their company described the gap precisely, writing about the first task of the new job: there was no shared way of selling, and the two sales groups were working from different slides. Both sets of slides were correct. That is what makes it hard to catch, because a disagreement about how the thing is sold does not look like a disagreement until the logic sits in one picture and the two versions can be held next to each other.
So judge it by the decision rules rather than by the document count: rules with their exceptions attached, in a form somebody can point at, plus the evidence that a person who was not in the room can use them. And the honest limit belongs in the conversation too. Roughly eighty to ninety percent of an expert area can be structured this way. The rest stays judgment, and naming it as judgment is more useful than pretending a rule exists.
What should you ask before you hire anybody?
Six questions, and the answers separate the routes faster than any proposal does.
- What do you start with? If the answer is words or design, ask what happens to the logic underneath.
- Which of my deals will you go through, and who do you need in the room? Real cases, won and lost and awkward, or it is a template being filled in.
- Show me what a decision rule looks like when you hand it over. Ask for one with its exception attached.
- How will we know it worked? The pass condition should be written down before the work starts, not judged at the end.
- What happens to the price if it runs long, and what do I pay if we stop halfway?
- Who owns the models, the rules and the documents afterwards?
If the offer is technical or the cycle is long, add one more. Ask how they will test the thing in a live conversation before anybody calls it finished, because a rule that has not survived a real buyer is still a hypothesis.
What does it cost to have it done with you?
This is the shape of it at Strategem, so you have one concrete set of numbers to compare against.
Two workshops. The first is diagnosis: the problem, the customer, the stakeholders and how each of them has to be spoken to. Between the workshops we draft the messaging and build candidate models. The second fuses the model into how you actually sell. You then use it in real conversations, and we meet again about three weeks later to correct whatever the field exposed. The extraction runs as structured interviews on your own real cases, won, lost and difficult, using grounded qualitative method from an organizational psychology PhD. We draw while we listen, and the bar is the moment somebody says, that is exactly what I do, I just could never name it.
The price depends on who is buying, because a committee sale and an expert-led sale are not the same job. For a multi-person B2B company with committee buying, the opening workshop is €5.000 and a full engagement starts at €15.000. For a smaller or expert-led company, the workshop is €1.500 and the full engagement is €5.000. The price is fixed before we start, iteration until the agreed acceptance test passes is included, and if we stop early you pay only for the phases that finished.
Strategem was founded by Dr. Fritz Hermann, who holds a PhD in organizational psychology and spent a decade as a CMO, with work across banks, insurers and Premier League clubs. If you are still deciding what the document itself should contain, the companion piece covers what a commercial playbook is and how to build one after an exit. The same work runs on expert judgment that has never been written down, on a founder the team keeps pulling into the same sales moment, and on an argument a champion has to carry without you.