पाठशाला Pathshala · ग्राहक Grāhak, The customer · Lesson 07 · Start

Who pays, who uses, who decides

Most stalled deals are not lost on price. They are lost to a person the founder never met. Map the five roles in every deal and write the one message each of them needs.

Pathshala, The Founder Library · 11 October 2026 · 8 min read

A bunch of house keys on a dark textured surface — Photograph: George Becker · Pexels

Most stalled deals are not lost on price. They are lost to a person the founder never met: the finance head who was not in the demo, the chartered accountant who said wait, the IT manager whose questionnaire arrived in week six. The fix is a map drawn in week one.

This lesson names the five roles in any sale, shows how they collapse in a small business and multiply in a large one, gives the message each role needs and ends with a weekly habit that keeps the map honest.

Five roles in every deal

The user touches the product every day and feels its benefits and its friction first. The payer owns the budget the money comes from. The decider says yes, and can say no even after everyone else has said yes. The champion wants the purchase to happen for reasons of their own and will sell it inside the building when you are not there. The blocker loses something if it happens, whether a budget, a vendor relationship, discretion over a process or simply a quiet life.

These are roles, not job titles. One person can hold three of them. One role can be held by a committee. The discipline is to write a name against each role for every live deal, and to treat a blank as the most important fact on the page.

The cost of not doing it is measurable. In a Gartner survey of 632 B2B buyers run in August and September 2024, 74% of buying teams showed what Gartner calls unhealthy conflict: members with conflicting objectives, disagreement about the best course or a decision overruled by someone outside the group. Buying groups that reached consensus were 2.5 times more likely to report a high-quality deal. A founder who sells only to the person in the demo is selling to one side of an argument they cannot see.

Why the user and the payer want different things

The user wants a better Tuesday: fewer clicks, fewer phone calls, less blame. The payer wants a number: a cost that fits a budget line and comes back within a year. The decider wants no surprises: nothing that fails in public, nothing that upsets parents or patients or the union. The same product, described in the same words to all three, persuades at most one of them.

This is the jobs-to-be-done idea applied inside a single customer. Christensen and his co-authors argue that customers hire a product to make progress in a circumstance. In a business sale each role has its own circumstance and therefore its own job. The fees clerk hires the tool to get Saturday back. The trustee hires it to bring fees in on time. The principal hires it to keep parents quiet. Write three job statements and the three messages follow from them. The [lesson on jobs to be done](/library/jobs-to-be-done-job-customer-hires-you-for) has the method.

Products that users can adopt on their own change the order but not the cast. Martin Casado and Peter Lauten at a16z describe how, when a user can sign up and expense a tool, all the complexity of the traditional sales cycle collapses into a single decision maker: the end user. It collapses only for a while. When the team wants the product for the whole department, the payer, IT and procurement return, and the founder who never mapped them is surprised by a contract that should have been routine.

The champion and the blocker

The champion is the most valuable person in the deal and the easiest to misread. A champion is not someone who likes you. It is someone who gains from the purchase, has standing with the decider and will spend that standing on your behalf. Test for all three. Ask what changes for them if this goes through. Ask who they would need to convince and whether they have convinced that person of anything before. Then give them something to carry. Mark Cranney’s line at a16z is that good products don’t sell themselves, but your champions can help sell them for you. What they need is a one-page case written for the decider in the champion’s voice, with numbers from your pilot and the risks already answered.

A bright empty corridor with a row of open doors on one side. — Inside the customer the decision travels down a corridor of offices. The champion walks it for you and the blocker sits behind one of the doors. Photograph: Pixabay · Pexels

The blocker is rarely hostile and almost never announced. In an Indian small business it is often the chartered accountant, whose question about GST invoices can end a deal by email, or the long-serving munim whose discretion the software removes. In a school it is the vendor of the old ERP. In a factory it is IT security and procurement, and Cranney again is blunt: even the best product can’t make a typical enterprise buyer change the way it does procurement. The rule is to find the blocker before they find you, ask what they would need to see to stay neutral, and give it to them in writing.

How the map changes with the size of the customer

The owner-run business. In a trader, a clinic or a small manufacturer, the payer and the decider are usually the owner, and often the champion too. The roles still matter because the owner holds each one for a different reason and changes hats in the middle of a meeting. Sell the user benefit to the owner as an observer of the staff, then the payback as the owner of the money, then the risk plan as the person who will be blamed. The CA and sometimes a family member sit just outside the room.

The mid-size company. Three to five people across two or three departments. The user and the decider rarely meet. The champion is usually a manager one level above the user. The payer is finance. Expect a procurement step, a vendor-registration form, and a question about where the data is stored.

The enterprise, the hospital, the government body. Roles become committees. Budget cycles decide timing more than enthusiasm does. A pilot with a success number agreed in writing before it starts is the only reliable way across. A later lesson in this track, on selling to Indian enterprises, covers the procurement half.

The household. Families buy like small committees. The child uses the maths app. One parent pays. The other may run the child’s week and decide. A cousin recommends, and the home tutor whose hour it might replace is the quiet blocker. Consumer founders who think they have one customer usually have three.

Writing the message each of them needs

For each role write one sentence that answers the question they are actually asking, then one piece of evidence that makes the sentence believable. The user is asking whether their day gets easier; the evidence is a before-and-after of a real shift, measured in minutes. The payer is asking whether the money comes back; the evidence is an arithmetic they can check in their own spreadsheet, in rupees. The decider is asking what could go wrong; the evidence is a rollout plan with an exit. The champion is asking what they get to present; the evidence is the page you write for them. The blocker is asking what they lose; the evidence is the export, the integration or the role you have kept for them.

Gartner’s finding is worth taking literally here: content built around what the buying group shares raised consensus, and content tailored only to individual priorities lowered it. So the five messages need a sixth sentence that all five people can repeat, the shared outcome. For the school it is fees in on time without a fight. For the plant it is fewer unplanned stops. Put that sentence at the top of every message.

Every blank on the deal map is a person who will meet your product for the first time when it is already too late to change their mind.

A worked example: the school in Lucknow

A founder sells a fee-collection and attendance tool. The administrative officer of an 1,800-student school in Lucknow loves the demo; he lost two Saturdays last term reconciling fee slips. Three weeks later nothing has moved. The founder had met one person and filled one row.

The map, drawn properly, has five rows. Users: the class teachers and the fees clerk. Payer: the trust that owns the school, represented by a trustee who visits on Thursdays. Decider: the principal, who will not change anything mid-term. Champion: the administrative officer, who is real but has never taken a vendor to the trust before. Blocker: the vendor of the existing school ERP, who also maintains the computer lab and has the principal’s ear.

The founder writes five messages. To the clerk, a timed comparison of the Saturday reconciliation. To the trustee, the overdue-fee arithmetic in rupees for a school of this size. To the principal, a go-live in the summer break with one section first. To the administrative officer, a one-page case in his own name for the Thursday meeting. To the ERP vendor, a promise in writing that every record exports to the system he runs. The deal closes in the break. The vendor did not help, but he did not object, and that was the job.

The deal map, every Monday

Keep one row per live deal with five columns, one for each role, and a sixth for the shared outcome. Every Monday review it for twenty minutes. Any deal with a blank column gets one action this week whose only aim is to fill it: a question to the champion, a request to meet finance, a call to the CA. Any deal where the champion has not introduced you to a second person in two weeks is a deal with a fan, not a champion; mark it down. Any blocker whose concern is not answered in writing gets that answer before the next meeting.

After every won or lost deal add one line: which role decided it, and whether you had met that person before the decision. After ten deals the pattern will tell you which role your sales process keeps missing. That is the role your next hire or your next piece of collateral is for.

The five roles are a map, not a script. Real buyers will hold more than one and change their minds in between; redraw it whenever they do.