Why a CFO, procurement and end users judge the same offer differently

Organisational buying has long been investigated as a process shaped by multiple participants and organisational circumstances. Webster and Wind (1972) presented a model that treats a purchase as an organisational decision rather than the action of an isolated buyer. Sheth (1973) addressed the differing expectations and conflicts that arise in industrial purchasing. Johnston and Lewin (1996) subsequently integrated a substantial body of research on organisational buying behaviour. These frameworks identify relevant dimensions; they do not tell you which individual has ultimate influence in a particular customer account.

A decision-making unit, or DMU, is useful precisely because the relevant people are not always obvious. Finance may hold the formal authority in one organisation and merely review the budget in another. An operator may lack purchasing authority while possessing knowledge that can stop an unsafe implementation. Procurement may design the process rather than select the strategic vendor. Influence can also change as the purchase moves from exploration to specification, evaluation, approval and rollout.

Two questions should precede every stakeholder profile: what collective choice must be made and who can affect it substantively, formally or operationally? Only then should we investigate the associations held by the people involved. A polished analysis of the wrong decision participants can be worse than a modest analysis of the actual buying centre.

A B2B proposal interacts with several systems. There is a technical system: does the equipment perform the necessary tasks? An economic system: what is the distribution of costs, benefits and uncertainty? An organisational system: who is accountable and who has to change how they work? Finally, there is a layer of meaning: what expectations, experiences and concerns does the proposal bring to mind?

We call that last layer associative. Words such as 'control', 'continuity', 'dependency', 'professional', 'more work' and 'future-ready' might emerge in interviews or association research. But their appearance is not evidence of a hidden emotional cause. 'More work' may describe a genuine implementation burden. 'Dependency' might reflect legitimate contractual restrictions. Association research describes how an option is understood; additional evidence is needed to show whether that interpretation affects behaviour.

An integrated DMU map therefore joins four distinct types of information: verified operational requirements, reported or indirectly measured associations, actual influence on the decision and evidence that remains missing. The dimensions must not be collapsed into a single psychological score. In the existing B2B series, the article on CFO versus procurement manager explores differences in evidence needs. This article adds the missing layer: how such differences interact within one purchase.

Consider 'standardisation'. The supplier may mean that all vehicles use common components and service procedures. Finance might interpret this as a possibility of more predictable expenditure. Procurement may see a simpler comparison between service packages but also a greater potential for supplier lock-in. Operators may welcome familiar controls while asking whether specialised tasks have been overlooked.

The same pattern can arise around 'automation', 'one partner', 'premium service' and 'flexible contracts'. The association does not reside permanently inside the word. It develops within a person's experience, responsibilities and immediate decision context. Gawronski and Bodenhausen (2006) distinguish associative and propositional processes in evaluation; that theory does not establish that a particular professional unconsciously rejects a specific contract.

Instead of asking only which benefits matter, ask what cost or risk can be activated by that same benefit. 'One supplier for everything' may evoke simplicity and dependence at once. A campaign promising simplicity cannot repair an exit clause that genuinely restricts freedom to change partners. The material condition and its perceived meaning must both be examined.

These are hypotheses, not survey findings. The table is useful because each appealing description is connected to something the organisation can actually check. A high average preference score cannot eliminate a binding operational or contractual obstacle.

Collective decision point	Possible positive meaning	Possible concern	What must be verified?
Multi-year capital commitment	Predictability and long-term readiness	Uncertain assumptions or restricted budgets	Sensitivity analysis, assumptions and approval route
One service partner	Clear accountability and responsive support	Supplier lock-in	Exit provisions, continuity and escalation
New controls and charging routines	Easier operation or less physical effort	Training demands and disrupted shifts	Representative workplace trial
Standardised fleet	Easier maintenance and consistent training	Poor fit for specialised jobs	Task analysis and exceptions
Phased implementation	Managed transition and learning	Operating old and new systems in parallel	Clear plan, ownership and acceptance criteria

Moriarty and Bateson (1982) examined ways to identify members of complex decision-making units and showed why starting from only one contact can give an incomplete picture. Silk and Kalwani (1982) reported limited agreement between informants about purchase influence in a particular industrial research setting. A persuasive internal advocate therefore cannot automatically speak on behalf of every person whose support is needed.

This is not an argument for a costly ten-person interview programme on every sales opportunity. It is an argument for checking whether the account team has identified the right people. Compare the supplier's stakeholder assumptions with the customer's own account of the process. Ask who specifies, who evaluates, who can veto, who approves and who is responsible after implementation. Do not assume that the person who speaks most in meetings also has the decisive vote.

For a small buying team, an accurately documented joint conversation may be an appropriate first step. For repeated lost bids, major investments and risky implementations, more systematic investigation can be justified. The methodological principle is the same: investigate the buying centre as a decision system, not as an assortment of titles.

A distribution centre wants to replace an ageing forklift fleet. A vendor proposes electrically powered vehicles, charging equipment, training, maintenance and a phased rollout. The initial reactions are positive. Management likes the strategic direction, procurement appreciates a single service arrangement and several operators find the controls comfortable during a short demonstration. Nevertheless, approval stalls.

This is a fictional teaching case, not a Neurofactor project and not measured customer research. There are no measured associations, verified environmental benefits, actual operating savings or conversion outcomes in the example. Every possible objection that follows is hypothetical.

A closer discussion reveals three different sets of unresolved conditions. Finance has a baseline calculation, but uncertainties around charging infrastructure and residual values are not yet represented adequately. Procurement appreciates one accountable service provider but sees ambiguity around termination and replacement parts. Operators only used the equipment in a quiet demonstration area, whereas the actual concern is charging, manoeuvring and workflow during busy shifts.

The vendor does not need a universal persuasion trick. It needs a joint validation plan. Finance receives a transparent scenario model with assumptions that can be changed. Procurement reviews contractual continuity, escalation and exit terms. Operators complete a representative trial with predefined safety and usability criteria. The participants then meet to assess one shared list of open conditions rather than receiving inconsistent presentations from different sales teams.

The result may still be no purchase. The actual infrastructure could be too costly, contractual safeguards may be insufficient or the equipment may be wrong for specific tasks. That would be a useful outcome of a sound evaluation, not an objection that must be bypassed. The initially positive association - 'modern and efficient' - never represented the entire decision.

Different priorities are not inherently a conflict. Conflict appears when satisfying one condition makes another harder to satisfy, or when nobody has authority to resolve the trade-off. A longer service contract may offer financial predictability while limiting switching options. A faster rollout may conserve short-term resources but put training under pressure. A more sophisticated machine may appeal to leadership while complicating work for some operators.

A factual trade-off must not be mistaken for an associative one. A genuine shortage of charging capacity will not be solved by better messaging. But if different participants interpret the same transition period as either 'careful preparation' or 'loss of control', that divergence is worth investigating. It might affect willingness to cooperate; whether it does so in this case must be tested.

Samuelson and Zeckhauser (1988) demonstrated status-quo preferences in several experimental and applied settings. Their research provides a reason to evaluate the incumbent option seriously. It does not prove that an organisation rejecting a new supplier is irrational. Sometimes retaining the existing system is the more defensible decision.

Reconstruct the actual purchase process. Who first identified the problem? Who set the technical specification? Who approved the shortlist? Who reviews safety, compliance and training? Who can commit the budget? Who will be accountable after delivery? Different individuals may hold several roles, and influence may move from one group to another during the process.

Ask separate informants to draw the path to approval, then compare their descriptions. If finance believes operations owns the decision while operations waits for procurement and procurement waits for the board, the first problem is governance rather than an implicit association. Silk and Kalwani's (1982) evidence of disagreement about influence supports the need to check these assumptions instead of choosing one respondent as the truth.

Do not equate speaking time with authority. A quiet specialist may introduce one legitimate safety condition that blocks the project. Conversely, the most vocal sponsor may have no power to approve the budget. Useful DMU mapping records responsibility, veto conditions, critical evidence and the stage at which each person can influence the decision.

A useful analysis does not produce three disconnected persona slides. It builds one map showing how the decision conditions fit together. The workflow below is an editorial application framework, not a separately validated psychometric scale.

The goal is not to calculate an average emotional index for the organisation. It is to understand how the same proposal enters a connected network of decisions. A blank field labelled 'not established' may be more useful than a confident but unsupported score.

1. Define the exact decision. Specify the offer, alternatives, timing and non-negotiable conditions.
2. Identify the actual participants and their influence. Verify formal and practical decision rights using more than one perspective.
3. Gather objective requirements separately from meanings. Do not merge a hard cost limit with an association such as 'expensive'.
4. Identify shared terms and divergent interpretations. Where do words like reliability or flexibility connect the discussion or cause tension?
5. Map conflicting requirements and evidence gaps. Which assumption remains unsupported, and who needs it tested?
6. Design a joint validation process. Combine documentary evidence, contract clarification and representative trials where appropriate.
7. Evaluate the decision and implementation. Record actual outcomes and alternative explanations rather than attributing everything to the intervention.

The solution is not three contradictory sales brochures. The organisation is considering one product, service or system. Its central proposition must therefore remain consistent, while the evidence is accessible from several perspectives. For the forklift example, a possible shared proposition is an operationally fit fleet with transparent costs, testable performance and clear service accountability. These are claims to verify, not established results.

Anderson, Narus and Van Rossum (2006) argued in a practitioner article that business-market value propositions should document rather than merely assert their claimed value. Apply this by building an evidence architecture: one promise, several proof requirements. Finance can inspect the scenario assumptions; procurement can inspect risk allocation and alternatives; users can inspect trial results. Crucially, the underlying numbers must not change with the audience.

Association research can check whether the proof is credible and coherent. Does 'less hassle' make sense if operators anticipate extra implementation work? Is 'flexibility' an accurate label for an exclusive, long-term agreement? The aim is not to optimise the prettiest word but to align promise, lived experience and contractual reality. The knowledge-base entry on core conflict in associations explores this tension in more depth.

This is not a collection of tactics designed to make every buyer say yes. Sometimes the right test shows that the offering is unsuitable. Discovering that before signature can prevent a disappointing rollout and a damaged relationship.

Decision tension	What must not be assumed?	Suitable test or evidence	Potential adjustment
Attractive investment but uncertain scenarios	That brand confidence replaces financial scrutiny	Transparent model and sensitivity testing	Scope, phasing or financing structure
One service partner also means dependency	That one contact point is enough for everyone	Exit terms, escalation rules and supply continuity	Stronger termination and continuity provisions
Positive demonstration but uncertain daily use	That a short trial predicts real adoption	Representative shift-based trial with predefined criteria	Training, configuration or workflow redesign
Apparent agreement without approval	That the most vocal participant holds the mandate	Independent mapping of roles and decision records	Explicit ownership and approval milestones
High overall rating but one critical failure	That averages override non-negotiable conditions	Test of safety and mandatory requirements	Correct the problem or reject the solution

Homburg, Workman and Jensen (2002) examined how key account management combines activities, actors, resources and formalisation. Their work is not a universal blueprint for sales teams, but it reinforces that cross-functional coordination is an organisational task, not a property of an attractive pitch deck.

An account manager can apply this approach before commissioning a formal research project. Rather than starting with 'What does your CFO think?', ask: 'Which conditions must all hold before this decision can be approved?' Then establish who needs to defend the risks, whose daily work is affected and which assumptions remain untested. Use the customer's own language and do not impose a psychological label.

Maintain one decision log with the open question, evidence required, responsible owner, method of verification and agreed review point. Encourage disciplines to challenge each other directly. Technical concerns deserve technical responses. Unsupported financial projections need better assumptions. An unclear decision right requires governance clarity. A language or association problem deserves research into meaning, not automatic price discounting.

If a word such as 'safe', 'easy to use' or 'sustainable' is repeatedly used, ask what concrete experience or condition it refers to. When the condition is objectively unsatisfied, change the offer first. Respect confidentiality and commercial boundaries when the buyer cannot share every internal consideration. Researching associations is not a licence to manipulate decision makers.

It is easy to ask participants whether they felt heard after a workshop. The answer is useful but does not demonstrate that the organisation made a better decision. Track whether material requirements were surfaced, assumptions were tested, the relevant people took part and the outcome was feasible. For recurring opportunities, a comparison of similar decision processes may help evaluate whether a shared evidence dossier or a joint review improves the quality of decisions.

Davis (1989) and Venkatesh and colleagues (2003) studied determinants of technology use and acceptance. Their research supports the relevance of perceived usefulness, effort and related constructs in the contexts examined, not an automatic prediction for the purchase of physical handling equipment. Adoption and implementation should therefore be evaluated separately from purchase approval. A DMU can approve a contract while operators later struggle with day-to-day use.

Finally, consider alternative explanations. Did approval change because the evidence improved, the price fell, the market changed or a decision maker was replaced? Without a suitable evaluation design, it is usually not possible to isolate the effect of an associative intervention. Clear reporting of this uncertainty increases the practical value of the findings.

An associative DMU analysis begins with one shared buying decision, not three job-title stereotypes. It identifies the decision participants, investigates the meanings attached to the proposal and connects those meanings to operational evidence and real decision rights. Its value lies in showing the relationships: where one promise is interpreted differently, where interests reinforce each other and where communication alone cannot repair a weak product, contract or process.

This is also the boundary of association research. An association map cannot, by itself, reveal what someone will buy. It can generate better hypotheses about why an option feels attractive, risky or incompatible and what evidence the group is missing. Combine such insight with interviews, documents, observed behaviour, trials and other appropriate measures. The principles of research triangulation and the introduction to associations in sales provide useful next steps.

The central point is straightforward: do not sell the same benefit three times to three job titles. Build one decision that can withstand three different realities. If the outcome is better evidenced and more implementable, the analysis has been worthwhile - even when the correct answer is no.