Buying Center: Five Roles, Two You Can Afford

What is a buying center in B2B sales?

A buying center is the group of people inside a company who decide on a purchase together. In B2B software that group is rarely one person and never quite stable. Five roles recur across deals: champion, user, decision maker, financial buyer and technical influencer, each with a different reason to say yes or to stall.

The term itself is old and the practice around it is thin. Most teams can name their buying center in a workshop and then run outbound as if a single title bought the product. The gap between the two is where deals go quiet after a good first call.

What makes the concept operational is not the list of roles. It is deciding, before a single message goes out, which of those roles you will actually contact, in what order, and what each of them has to be able to say internally for the deal to move.

Which roles belong in a buying center?

Five, and the three main role models agree more than they differ. Champion and decision maker appear in all of them. The variation sits at the edges: whether procurement counts as its own role, whether the day-to-day user is tracked separately, and whether the technical evaluator gets a seat of their own.

Our ICP template asks for one thing the public models leave out: a win condition per role, next to the title, the pain and what they care about. Not what the role wants in general, but what has to be true for this specific person to push the deal forward one step.

That column is the useful one. A role without a win condition tends to be a placeholder someone added because the framework had five rows.

When do you map the buying center?

During the ICP definition, before positioning and long before any sequence is written. The buying committee is one of the fields the ICP work has to produce, alongside the primary use case, the problems and the firmographics. Everything downstream reads from it, so a late answer is an expensive one.

The order is not cosmetic. Segment first, because a buying center that holds for a 40-person company does not hold for a group with a central IT function. Roles second, because the messaging work reads directly from that table. Capacity third, because it is the step that decides how much of the model you can afford to act on.

Skip step three and you get a five-role map with two roles ever contacted, which is worse than a two-role map honestly drawn.

What does each role actually need to hear?

The same capability, framed against a different problem. Core positioning does not change per role, but the entry point does: the user hears about their workflow, the champion gets ammunition for an internal argument, the financial buyer gets a business case, and the technical influencer gets the integration answer before they have to ask.

Four fields per role are enough to write from. What they care about, in one sentence and from their perspective. The relevant challenge, in their words rather than yours. The new capability, framed for their responsibility. The promised value, specific to them.

The classic example is one root problem with two faces. A user says the tool wastes three hours a day. A decision maker says productivity sits below benchmark. Same issue, and a message that mixes the two framings lands with neither.

Where this goes wrong is not the writing. It is skipping the ICP work and inventing the pains at draft time, which produces five variants of the same paragraph with the job title swapped.

How many roles can you realistically contact per account?

Fewer than the model suggests, and the arithmetic settles it rather than the ambition. Work backwards from the target: weekly touchpoints equal the weekly positive-reply target divided by the expected reply rate. One LinkedIn sender or one inbox carries roughly 100 touchpoints a week, follow-ups included.

Take a real target. Ten positive replies a week at a fair reply rate of one in four hundred means four thousand touchpoints a week. Split evenly between LinkedIn and email with one follow-up on each step, the LinkedIn share alone comes to four thousand touches, which is forty LinkedIn senders. Not feasible. That is the conversation to have with the client before launch, not in month two.

The rate you pick decides everything above it. Around one in two hundred for strong ICP fit in the primary region with sharp messaging, one in four hundred for average, one in a thousand for cold, weak fit or off-region work. Add a fifth to the inbox and domain count for spares and reply bandwidth, and remember warm-up takes three weeks before the first send.

Now put the buying center back into that equation. Contacting three roles per account instead of one triples the touchpoints for the same account list, or divides your account coverage by three at the same sender count. Both are legitimate choices. Making them by accident is not.

The published research is what makes this a real constraint rather than a scheduling detail. Forrester's State of Business Buying, 2026, drawn from a survey of nearly 18,000 global business buyers, finds that on average 13 internal stakeholders and nine external participants influence a buying decision, and that procurement acts as a decision maker in 53% of cycles. 6sense, working from close to 10,000 buyer interviews over three years, puts the typical group at around 10 members and reports that in 80% of journeys the requirements and the winning vendor are settled before sellers enter the picture. Its 2025 Buyer Experience Report puts the point of first contact at 61% of the journey, moved forward from roughly 69% a year earlier. Nobody sequences 13 people per account. The honest reading is that you pick two or three roles, reach them early enough to shape requirements, and accept that the rest of the group will be reached by your champion rather than by you.

Reaching the right two beats reaching more of the wrong ones. Selling to decision makers instead of analysts within less than 4 months is what that shift looked like for Jua, a SalesPlaybook client.

Why keep one list per role instead of one combined list?

Because a merged list cannot answer the only question that matters after four weeks: which role replied. Roles get sourced separately, sequenced separately and reported separately, with primary and secondary tiers kept apart as well. The account view is rebuilt on top of those lists, never in place of them.

The structure is dull and it holds. One list per role per tier, one row per contact, and a separate account-level view that carries the outreach priority, the contact counts per role and the top primary-role contact for that company. Priority is then a rule rather than a feeling: an account only reaches the top band if it sits in the primary tier, clears the fit score and has a primary-role contact attached. No primary-role contact, no top priority, however attractive the logo.

One reporting habit belongs with it. Never report a total without the split. "Four hundred contacts" tells a client nothing; four hundred across three roles and two tiers, with the counts named, tells them where the coverage is thin and where the next sourcing round has to go.

What goes wrong when the buying center stays a slide?

It quietly becomes decoration. The roles get defined in the strategy document, the sourcing runs on one title anyway, and six weeks later nobody can explain why the meetings stall after the first call. The map was right. It just never reached the list.

The fix is not a longer document. It is making the buying center a column in the list rather than a section in the deck: every contact carries the role it was sourced for, every sequence is written against that role, and every report splits by it. From there the discipline maintains itself, because a missing role shows up as an empty column instead of as a vague sense that outreach is not landing.

Upstream of all of it sits the segment itself: the buying committee is one field in a B2B market segment definition, and the touchpoint budget it implies is the same one a pipeline generation plan has to carry. The same logic runs through our work on B2B pipeline generation, on the sender side in LinkedIn GTM and AI outbound, and in fractional sales leadership, where the buying center usually turns out to be the reason a strong product keeps losing to a worse one with a better internal advocate.