Playbook
Customer discovery questions for a new consulting business
The cheapest market research a founder can run: ten questions, ten conversations, and a ranked list of what actually hurts.
The methodThe Wince Test
This playbook is a set of customer discovery questions for a new consulting business: ten pain statements wrapped in context questions, run across ten conversations and scored twice, producing a ranked list of what actually hurts the people you want to sell to. It is for founders who have expertise and an idea, no customers yet, and a habit of describing their answer instead of asking about the problem. The method is called the Wince Test, because the signal you are collecting is not what people say; it is which sentence makes them flinch. It works before you have a productised (productized) service to describe, and it costs ten conversations rather than a research budget.
The problem is that a founder in a discovery conversation is selling, even when they believe they are listening. You describe the service, the prospect is polite, you leave with a list of things they agreed with, and none of it tells you what they would pay to make go away. The other failure is the open question. Ask someone what their biggest problem is and you get the answer they gave their last consultant. Pain that people have lived with for years has stopped feeling like pain; it is just how things are. You have to put the sentence in front of them and watch.
The method starts with the pain list: ten short statements, each describing one specific thing going wrong, written in the buyer's language rather than yours. Around the ten sit context questions about the size of the team, what has already been tried, who owns the problem and who pays for it, so the same pain can be sized differently in two different companies. You run the list in ten conversations, reading each statement aloud and scoring the reaction on the spot: recognition, indifference, or the wince. Afterwards you score again, cold, against what they actually said rather than how it felt in the room. The two scores are compared, and the gap between them is itself information about your own bias.
The output is a ranked list. At the top is the pain that made the most people wince and that they had already spent money or time failing to fix. That is the one you build the first offer around, and the playbook ends by turning it into a rewritten first line for your proposition, so that the thing you say to the next ten people is the thing the first ten flinched at.
What it deliberately leaves out: survey design, statistical significance, persona templates and anything that needs a sample larger than ten. Ten is not a scientific number. It is the number of conversations a founder can have in a fortnight without stopping the day job, and it is enough to separate a pain everyone shares from a pain one loud person has. There is also no script for closing. The Wince Test finds the problem; selling the answer is a different playbook.
Before we had a company, in early 2013, Steve drafted a list of the things we suspected hurt the people we wanted to sell to, and we asked them to rank it. The list was wrong in useful ways. The ranking told us which service to describe first, and which of our favourite ideas nobody would pay to fix.
The playbook is in development, with the worksheet and the interactive scorer alongside it. Joining the waitlist means you hear when it is ready, and it gives me the chance to ask what your ten conversations would need to cover. Nothing is for sale yet.
What you will get
Who it is for
Questions
- What is the Wince Test?
- A way of running customer discovery that watches reactions instead of collecting opinions. Ten short pain statements are read aloud, one at a time, and the reaction to each is scored on the spot, then scored again cold afterwards. The sentence that makes the most people flinch, and that they have already spent money or time failing to fix, is the one the first offer is built around.
- How many customer discovery interviews does a new consulting business need?
- Ten is the working number here. It is not statistically significant and is not meant to be. It is what a founder can do in a fortnight around the day job, and it is enough to separate a pain everyone flinches at from one loud person's complaint.
- What questions should you ask in customer discovery for a consulting business?
- Not open ones. Put ten specific statements about things going wrong in front of the buyer and watch the reaction, with context questions around them so the same pain can be sized in different companies. The playbook supplies the ten, the context set and the scoring.
- Why score each conversation twice?
- The first score is taken in the room, on the reaction. The second is taken cold, afterwards, against what was actually said. The gap between them measures your own bias, which is the thing most likely to send you building the wrong service.