Enterprise Discovery Calls: How to Handle Five Stakeholders Who Want Different Things

Five people join your call. The functional lead wants the problem gone. The finance partner wants to know what it replaces. The security reviewer wants an answer about data residency. Somebody from the platform team was told to attend and is unclear on the reason. And your champion, who loved the demo, is now quietly managing all four of them.

That call cannot be run as a discovery conversation with extra attendees. It is a different job with a different objective: not learning what the company needs, but learning what each person needs and where those needs collide. This piece covers how big the room actually is, what the research says about the disagreement inside it, a question bank organised by function, and the call sequence that holds it together.

How Big the Room Really Is

Forrester published a blog in February 2026 drawing on its Buyers' Journey Survey from 2025. Per that post, written by Katie Fabiszak and Kaitlyn Bretagne, 73% of purchases involve three or more departments, and a buying decision touches an average of 13 people inside the buyer's organisation plus nine from outside it. Those are two separate counts and should not be added together. Forrester does not publish the survey's sample size on that page, so take the figures as directional.

Gartner's framing is narrower and more useful for planning a call. Delainey Kirkwood, a principal in Gartner's sales research practice, puts the size of a buying group at "five to 16 people" spread over as many as four business functions, each arriving with its own priorities.

Four functions is the number to design around. Not sixteen people, because you will never meet sixteen people, and not one champion, because a champion is a relay rather than a decision. Four functions means four different definitions of what a good outcome looks like, and your job in discovery is to collect all four before anybody writes a proposal.

Not Every Deal Is a Committee

Before you rebuild your process around a sixteen-person group, one correction that almost never appears in articles like this. Forrester's 2023 State of Business Buying study, based on responses from more than 18,000 global business buyers, found something inconvenient for the committee narrative. Per the press release, in North America 43% of business buyers make purchase decisions with no more than two people inside a single department, against 29% in Europe and 35% in Asia Pacific.

So a meaningful share of B2B purchases, including some wearing enterprise logos, are genuinely decided by one or two people in one team. Running a four-function discovery programme at those deals is not thoroughness. It is a way to turn a two-week decision into a two-month one, and the buyer will notice. The first job is therefore diagnostic: find out which kind of deal you have before choosing which process to run. That same study names the inability to build internal consensus on a vendor as a reason purchases stall, which is the problem the rest of this article addresses, for the deals that actually have it.

The 74% Figure, and What Gartner Actually Counted

The statistic everyone quotes rests on 632 B2B buyers whom Gartner polled in August and September 2024. Per the release published the following May, unhealthy conflict shows up in 74% of buyer teams while a decision is being made, and the groups that did reach agreement were 2.5x likelier to call the resulting purchase a good one.

Read Gartner's own definition before you build anything on it. Gartner's own wording covers three separate situations: members whose objectives clash, members who cannot agree which option is right, and members whose decision is reversed by someone above the group. Look hard at that third one. Being overruled from outside is not an argument between the people on your call at all. It is a structural fact about where authority sits, and no amount of skilful facilitation touches it.

Note too that deal quality here is self-reported satisfaction rather than a measured win rate. The consensus result tells you groups that agreed liked what they bought. It says nothing about whether you signed them.

Which Kind of Conflict Are You Actually Looking At?

That 74% headline, in Gartner's release, usually gets used to license one instinct: keep the room calm, do not poke at disagreement, stage the appearance of alignment, and present to everybody simultaneously so nobody feels left out. We think the instinct is backwards, and the evidence against it is substantial.

Frank de Wit, together with Lindred Greer and Karen Jehn, published a meta-analysis of intragroup conflict in the Journal of Applied Psychology in 2012, pooling 116 empirical studies covering 8,880 groups. Per the openly available version held by Leiden University, the analysis found stable negative relationships between group outcomes and two kinds of conflict, relationship conflict and process conflict, but did not find a strong negative association between task conflict and group performance. More pointedly, task conflict was more positively related to performance in top management teams than in other teams, and more positively related where performance was measured as financial performance or decision quality.

Translate that into a sales call. Arguing about which problem matters and what the right answer is, which is task conflict, is not the damaging kind, and in senior groups deciding something consequential it is associated with better decisions. Arguing about who owns this, who was not consulted and who is slowing things down is process and relationship conflict, and that is the kind that wrecks outcomes. Gartner's definition of unhealthy conflict blends all of it together, which is why the headline figure cannot tell you what to do next.

So our position is that a seller should actively surface task conflict and spend real effort defusing the other two. Ask who disagrees and about what. Put the disagreement in writing and send it back. A group that argues about the task in front of you is a group making a real decision, and silence on your call usually means the argument is happening without you, where you cannot answer it.

Someone showing that in a sales setting specifically, surfacing task disagreement slows a decision or costs win rate even when relationship conflict is held low, would end the argument. That test has not been published. Until it is, a quiet room is the signal we would worry about, not a loud one.

A Question Bank by Function

Different functions do not share a problem, so they should not share a question list. Keep each conversation short. Gong's study of 519,000 recordings carries an explicit warning that piling on questions hurts a sale at executive level, so senior time deserves a shorter and sharper list.

The functional owner, the person who feels it daily

• Which part of this costs your team the most time in a normal week?
• What have you already tried, and what happened?
• If this is still true in a year, what does that mean for your plans?
• Who on your team would notice first if it improved?

Finance, the person who must justify it

• How does a purchase of this size normally get approved here?
• What are you measuring that this would show up in?
• What does this replace, if anything?
• What would make you confident in the numbers rather than persuaded by them?

Security or IT, the person who can stop it

• What does your review process look like, in steps and documents?
• Which questions have killed a vendor for you before?
• Where does this need to sit relative to your existing systems?
• Who signs off at the end, and what do they need to see?

Your champion, the person managing the others

• Who is most likely to disagree with this, and what is their objection?
• What is the hardest question you expect to get internally?
• Is there anyone above this group who could overrule it?
• What would you need from me to make your case without me in the room?

That last set matters most, and the third question in it is the one almost nobody asks. Gartner counts being overruled from outside the group as a form of dysfunction, so finding out early whether that risk exists is worth more than another feature conversation. The preparatory work sits in our guide to charting a buying group early.

The Call Sequence That Fits Around It

1. Call one, with your first contact. Spend the last few minutes building a map rather than asking for introductions. Who feels this, who approves the money, who could block it, who might disagree. Write names and functions down.
2. Calls two to four, one per function. Twenty minutes each, built from the relevant list above. Three or four brief sessions normally reach each function holding a veto.
3. Draft the shared statement. A single paragraph naming the thing being fixed, the price of leaving it alone, and the points on which the four versions diverge. Spell the divergence out rather than sanding it off.
4. Send it back and ask what is wrong with it. Not whether it is right, which invites politeness. Asking what you got wrong surfaces more in one reply than three more calls will.
5. Only then, the group session. Bring the shared statement, not a deck. The meeting's job is to resolve the differences you already documented, which is a far better use of five calendars than hearing a pitch together.

Keep the number of questions per conversation low and the number of conversations adequate, which is the same trade we make in our three-problem discovery framework, applied across people instead of within one call.

Attendance Is a Signal You Can Count

If you want a cheap leading indicator, count heads. Gong's 2019 analysis of over 10 million sales conversations and 500,000 emails found that winning deals had at least three people from the buyer's side in meetings, while losing deals on average never got past a single point of contact. On the email side it recorded eight contacts on winning deals against three on losing ones, which Gong presents as a 243% difference in engagement.

Two cautions. That 243% is arithmetic on contact counts, not a lift in win rate, and the analysis is from 2019, so treat it as a pattern rather than a current benchmark. We are using it because no newer study with a comparable sample has been published on buyer-side attendance. What it supports is simple enough to act on tomorrow: a deal where you have met one person is a deal you cannot see, and the fix is covered in why a single champion cannot carry a decision.

Content Aimed at One Person Backfires

There is a specific finding in the Gartner work that changes what you send between calls. Per the release, tailoring content for buying-group relevance improved consensus by 20%, while content focused on individual-level relevance produced a 59% negative impact on consensus. Where buyers did experience group-level relevance, they were 3x likelier to rate the purchase highly.

That is a direct argument against the personalised one-pager. A document written for your champion alone is a document the other three cannot use, and it gives your champion something to defend rather than something to circulate. Write the material for the group, with each function's objection answered in it, and let your champion forward it unedited.

Objections You Will Meet Internally

"My champion asked me not to contact anyone else." That is information, not an instruction. Ask why. Most often they are worried about losing control of the story, or they are not confident the organisation is behind them. Offering to build something they can carry themselves answers the actual fear.

"Multi-threading drags the deal out." It lengthens the first month and shortens the last. The delay you are trying to avoid is almost always an unspoken objection maturing somewhere you are not.

"I do not have time for five calls." You need three or four twenty-minute conversations, not five hours. Pacing that across a quarter is the subject of navigating long enterprise cycles.

"They disagreed in front of me and it was awkward." That was the good call. Hold your neutrality, play each position back in simple language, then ask which piece of evidence would resolve it. Take nobody's side and do not steer away from the topic.

Five Things to Do This Week

• Close every opening call with the same extra question: who here is likely to object, and on what grounds?
• Keep one five-column grid per deal covering the person, their function, their requirement, their worry and the date you last spoke.
• After each conversation, redraft the single-paragraph statement of the shared problem and send it out to be corrected.
• Produce one document written for the whole group rather than for your champion, and stop sending individually tailored versions.
• Take your most stalled deal and list every function you have never spoken to. That list is your week.

Closing Read

Enterprise discovery is a sequence of short, different conversations whose output is a written disagreement you can work with. Diagnose whether the deal really has a committee, collect each function's version of the problem, surface the argument about the task and defuse the arguments about territory, and write for the group rather than the ally. None of this promises a signature, and results differ by deal and by market. What it does buy you is a group whose argument is about the correct subject, held while you can still hear it.