How Complex Buying Decisions Are Made

The Reditus B2B Buyer Model, condensed for sales leaders

1. The Deal That Died With No Cause of Death

The deal entered the forecast in March at 60 percent. By June it was in commit.

The rep had done everything the methodology asked. He was multi-threaded across six stakeholders. He had a champion in the VP of IT who returned calls on weekends. The pilot had finished with clean results. The buyer’s own finance team had validated the ROI model. Procurement was engaged. Legal had the paper.

In the July pipeline review, the rep called it a formality. The champion had said the words every seller wants to hear: everyone’s on board.

Then the meetings started moving. Not canceled, moved. A compliance director who had been quiet since spring asked for the security documentation a second time. Finance, previously supportive, suggested the start date might sit better in the next fiscal year. The champion stayed warm but began saying “we” less and “they” more.

Nothing went wrong. No competitor appeared. No objection surfaced that had not already been answered. The deal did not close lost. It simply stopped being a deal, one reasonable delay at a time, until the CRM note read: revisit next quarter.

If you run a sales organization, you have watched this happen. You have also sat in the postmortem where nobody could name the cause of death. The rep executed. The champion was real. The value was proven. The forecast was wrong anyway.

The most expensive outcome in complex sales is not rejection. It is sustained investment in deals that were never able to move. This brief explains why those deals exist, why your current tools cannot see them, and what your team would need to read differently to find them earlier.

2. The Explanation You Have Been Given Does Not Predict Anything

Most sales methodology rests on a false assumption: that a buying committee behaves like a single rational decision-maker. Under this assumption, the committee gathers information, weighs tradeoffs, and converges on the option that is best for the organization. Selling becomes the work of feeding that rational process. Better discovery. Sharper ROI. Stronger champions. More urgency.

If the assumption were true, three things would happen reliably. Decisions would converge as information increased. Late-stage objections would diminish rather than intensify. Decision speed would track the quality and clarity of the data.

Your pipeline reviews show the opposite. Objections appear late, after months of alignment. Additional information stalls momentum instead of building it. Deals slow down precisely when the data is strongest and consensus looks closest. These are not edge cases. In complex deals, they are the norm.

That pattern tells us something structural. A committee cannot be modeled as a person. The company does not decide. The org chart does not decide. A group does not decide. Individuals decide, inside a system that constrains what they can tolerate.

The Reditus B2B Buyer Model (RBBM) is a first-principles account of that system. It starts from a definitional point: a B2B purchase is complex the moment no single person can make the decision alone. Not deal size. Not segment. The disappearance of unilateral authority. From that point forward, the correct unit of analysis is not the account. It is the individual operating inside the buying system.

The buying system. The buying system is the set of individuals who must approve, can block, can shape, or will execute a complex purchase, each evaluating the decision through their own personal consequence.

Everything that follows describes how the buying system behaves, and why its behavior is predictable once you know what to look for.

3. Power Follows Consequence. Utilities Translate It.

Picture a steering meeting where a proposal with broad support gets amended, narrowed, and rescheduled because one person, nowhere near the top of the org chart, says a single true thing: if this goes wrong, my team lives with it. Every experienced seller has watched a voice like that carry a room. The RBBM explains why, in three claims.

The first claim: in stable organizational systems, power derives from consequence. An individual’s power in a decision is proportional to the personal consequence they bear if that decision succeeds or fails. This is a systems claim, not a moral one. Organizations that let decision power decouple from accountability accumulate errors faster than they correct them, and they destabilize. Systems that survive concentrate influence where downside is concentrated.

This is why qualification built on formal authority maps the wrong thing. The RBBM sets authority aside. The question is not who can sign. The question is who bears consequence, because that is where decision power actually lives.

The second claim: organizations bear outcomes; individuals experience consequence. An organization can be fined, miss targets, or take reputational damage. Those effects are real, but they exist at the organizational level, and at that level they are inert. A known risk can sit on a dashboard for quarters, understood by everyone and owned by no one, until an audit notice gives it a name. The risk does not change. Its ownership does. Once consequence lands on a specific person, behavior changes.

Personal consequence. Personal consequence is the impact a decision has on a specific individual through accountability, reputation, workload, or career, as distinct from the outcomes the organization bears.

The third claim: individuals evaluate consequence through exactly two dimensions. The first answers the question, what will this outcome do to me? It covers workload, exposure, career risk against career leverage, stress, and reversibility. The second answers the question, what does taking this action say about me? It operates through three lenses: the story the individual tells about the kind of professional they are, how the decision positions them in the eyes of others, and whether it moves them toward the role they want next.

Selfish utility. Selfish utility is the dimension of personal consequence that answers one question: what does this decision do to me if it goes right or wrong?

Identity utility. Identity utility is the dimension of personal consequence that answers one question: what does supporting or opposing this decision say about who I am?

Two dimensions are sufficient because the drivers sellers usually cite collapse into them. Risk aversion is narrative identity (”I exercise sound judgment”) plus selfish utility (”failure here is visible and asymmetric”). Budget objections are identity (”I am fiscally responsible”) plus selfish utility (”I am blamed for overruns, not for missed opportunities”). Even ROI resolves this way. ROI does not drive decisions on its own. It functions as permission. It lowers perceived consequence and provides cover if the decision is questioned.

One more constraint matters. These evaluations run on perceived consequence, not objective outcome. A workload reduction that is real but not yet believed moves nothing. A risk that is manageable but does not feel manageable still blocks.

Return to the deal that opened this brief. The compliance director who asked for the security documentation twice was not being thorough. She was weeks from an audit, and if anything went wrong during implementation, the finding would carry her name. Her second request was not a question about the product. It was a question about herself.

4. The Tolerance Curve

Now put the two dimensions together. For any individual, a decision occupies a position in a two-dimensional space: identity utility on one axis, selfish utility on the other. Within that space, each individual has a boundary. Combinations that fall on one side are survivable. Combinations that fall on the other are not. The RBBM calls that boundary the tolerance curve, and it is the central mechanism of the model.

Tolerance curve. A tolerance curve is the boundary, in the space defined by identity utility and selfish utility, that separates decisions an individual can survive from decisions they cannot, at a specific point in time.

Two properties of the curve do most of the explanatory work.

First, tolerance precedes preference. An action is selectable only if it is tolerable. Actions outside the curve are not candidates, regardless of organizational merit. Most buying models assume every option on the table is viable and spend all their energy on preference. This model explains why preference often never becomes operative: no option falls within tolerance in the first place.

Second, the curve is individual and time-specific. It is shaped by current workload, recent successes and failures, available political capital, and proximity to audits, reviews, or promotions. None of those appear in your CRM. All of them move the curve. The same proposal can pass in one quarter and fail in the next without any change to its merits, because the buying system is no longer the same system.

Viability is also collective. A decision becomes viable only when it lands inside the tolerance curve of every individual who holds power, and power, as established above, follows consequence. Enthusiasm elsewhere does not compensate. One binding tolerance curve is sufficient to stop a decision, which is why complex deals so often die without anyone ever saying no.

In the deal that opened this brief, the curves moved twice. An expanded audit consumed the compliance director’s capacity to absorb anything new. Forecast pressure narrowed what finance could carry inside the fiscal year. Neither event had anything to do with the product. The deal’s position never changed. The space it needed to fit inside did.

5. One Purchase, Three Decisions

From the outside, a purchase looks like one decision. Inside the buying system, it is up to three, and each is evaluated against the same tolerance curves at the same time.

The three embedded decisions. Every complex purchase contains up to three embedded decisions, whether to act at all, whether to build or buy, and which vendor to choose, each evaluated simultaneously against the same tolerance curves.

These are not sequential gates. A committee does not settle the question of acting, then the question of building or buying, then the question of vendors. All three sit in the tolerance space at once, and each occupies its own position.

The decision to act or not act typically carries high personal risk early. Acting introduces exposure, visibility, and accountability. Doing nothing preserves existing narratives, roles, and options. This decision often sits outside tolerance even when the problem is acknowledged and the value of a solution is accepted, and no-act can appear two ways: as a position someone actively holds, or as the default the system produces when nothing else clears.

The decision to build or buy is a separate position, not a subset of vendor selection. Building scores higher on narratives of control and self-reliance while raising execution burden. Buying reduces workload while concentrating responsibility for the selection on whoever made it. This is why “we could just build it” persists even where it is plainly inefficient.

The decision of which vendor rarely maximizes upside. It minimizes regret within tolerance. That single sentence explains why safer vendors outperform better ones, why familiarity outweighs differentiation, why references matter more than features, and why ROI parity fails to break ties.

Here is what this does to your forecast. Your pipeline stages track exactly one of these decisions: vendor selection. Demos, evaluations, proposals, and redlines are all selection activity. Meanwhile, the act versus no-act decision may still sit outside tolerance for someone with power. A deal can show 75 percent on the selection decision while the decision that governs everything else has never cleared. That deal is not late-stage. It is unviable, dressed in late-stage activity.

6. Nobody Killed the Deal. It Emerged Dead.

No single actor selects the outcome of a complex purchase. A purchase occurs only when all required embedded decisions fall within tolerance, for every individual with power, at the same moment in time. When that condition is not met, the buying system has not failed to decide. It has resolved into the outcome it can sustain, which is often nothing. The RBBM calls this emergence, and it is why postmortems come back empty. There is no single objection, no decisive moment, no person who changed their mind. The buying system simply moved where it could, and where it could not, it did not.

Emergence reframes friction. Meetings without decisions, analysis without commitment, engagement without closure: this is motion without progress, and it is structural. Friction is not resistance. It is the buying system signaling constraint.

It also explains why the standard rescue moves disappoint. Urgency increases stress, and stress narrows tolerance curves, making convergence less likely rather than more. Executive sponsorship can genuinely help, because it changes who bears downside, but it cannot force convergence while identity or selfish utility constraints remain unaddressed for the people who must live with the outcome after the executive’s attention moves on.

Constrained systems do not only stall, though. They adjust. When an initial configuration prevents convergence, the buying system redistributes what it is carrying until an outcome becomes possible. The RBBM calls this horse-trading, and it works through three mechanisms of equal weight. Consequence reassignment changes who bears downside: implementation ownership shifts, accountability diffuses, an outside party absorbs delivery risk. Tolerance expansion grows an individual’s capacity to carry strain: political capital strengthens through visible endorsement, workload lifts, formal authority confers legitimacy, and the whole curve moves. Utility recalibration changes how a specific option is evaluated, moving where the decision sits relative to the curve without changing the curve itself.

Horse-trading. Horse-trading is the redistribution process through which a constrained buying system adjusts, by reassigning consequence, expanding tolerance, or recalibrating perceived utility, until an outcome the system can sustain becomes possible.

None of this requires coordination. Individuals act locally, within their own limits, and the configuration that emerges was assembled by no one. This explains the pattern every sales leader has seen and none can account for: the stalled deal that suddenly closes with no forcing function and no change to the solution. From the CRM, it looks like perseverance. Inside the buying system, it is redistribution.

The deal that opened this brief found no redistribution. Nobody reassigned the compliance director’s exposure. Nothing expanded what finance could carry inside the year. Nothing recalibrated how the decision would be lived by the people who would live it. The buying system resolved into the only outcome it could sustain. “Revisit next quarter” was not a delay. It was the outcome.

7. What This Changes on Monday

The RBBM is a way of seeing, and a sales organization that sees this way runs differently in four places.

Deal reviews. Stop asking only what the next step is and start asking who carries the risk right now. For each individual with power: where does each embedded decision sit relative to their tolerance, and what has moved since the last review, in their world rather than in the deal? A rep who cannot answer those questions is reporting activity, not viability.

Forecasting. Stage-based probability fails in complex deals because stages measure activity, not viability. The corrective is a judgment overlay on every material deal: has each embedded decision cleared tolerance for each person with power? Until the answer is yes across the board, the deal carries a structural probability problem that no volume of activity resolves.

Coaching. Emergence is not an argument for passivity. It is an argument for expertise. The skill to build in your team is reading the buying system: who bears consequence, how tolerance is shifting, and which interventions the buying system can sustain. Reps who read the system shape conditions. Reps who do not, generate motion.

Opportunity selection. Some opportunities are aligned. Some are shapable. Some are structurally non-viable, and in those, no legitimate lever exists. Walking away from the third category is a performance decision, not a concession, because capacity invested in systems that cannot move is capacity unavailable to systems that can. The Seller Corollary develops this into practice.

8. Where This Goes

This brief compresses a larger body of work. The RBBM Foundational Paper establishes the full argument, including the boundary conditions and exceptions this brief omits. The Seller Corollary translates the model into diagnostic practice for individual sellers: when familiar tactics work, when they fail, and when no approach will succeed because the buying system cannot tolerate the decision. The Seller’s Toolkit operationalizes that practice. Additional corollaries for marketing, customer success, and buying organizations are in development. Each translates the same foundational model into role-specific judgment without altering the underlying theory.

Traditional buying models work where a single decision-maker holds authority. They fail in complex environments because they cannot explain why acknowledged value does not produce action, why identical proposals succeed in one quarter and fail in the next, or why committees fragment without disagreement. The RBBM explains all of these, not as anomalies but as predictable outputs of constrained systems. It does not offer a playbook. It offers a way to see.

What if the deals you lost this year were never winnable, and the only real question is how early your team could have known?