Coverage and the central idea
A decision is a choice among alternatives to address a problem or opportunity. Decisions appear throughout planning, organizing, leading and controlling. This chapter explains four steps: identify the need for a decision, develop alternative responses, choose an appropriate alternative, and implement and monitor it. These steps form a cycle because implementation produces evidence, new problems and opportunities. A management approach influences what gets noticed, whose interests count, which alternatives seem acceptable, and whether managers learn from mistakes.
The learning goals include using scripts to explain differences among FBL, TBL and SET; recognizing programmed and non-programmed choices; selecting approaches appropriate to knowledge and agreement; understanding resistance and escalation of commitment; and applying transaction cost reasoning to entrepreneurs' make-or-buy decisions. Learning the sequence without its judgment and ethical dimensions gives only a partial understanding.
7.0 · The Ford Pinto case: rational according to what purpose?
The chapter's opening case concerns Ford's Pinto recall analysis and recall coordinator Dennis Gioia. A financially framed comparison favored paying expected claims rather than recalling vehicles. Public reaction showed how an apparently rational calculation can omit morally important effects. Gioia's experience illustrates how organizational scripts and professional routines can displace an individual's earlier commitments. The case is a textbook account of historical events, not a recommendation to place a monetary price on human dignity.
The central issue is that a calculation is only as complete as the outcomes and values it includes. FBL scripts emphasize company finances; TBL can add reputation and financially relevant stakeholder harms; SET treats affected people's well-being as significant in its own right. Legality, accepted professional standards and inner moral judgment need not give the same answer. Managers should deliberately examine their assumptions rather than assume a familiar formula settles every ethical question.
7.2 · Step 1: identify the need for a decision
Cognitive scripts are learned patterns for recognizing situations and responding. They help managers manage many issues but can hide problems that fall outside established categories. A manager filters issues to decide what requires action now, what can wait and what can be ignored. Viability threats and familiar triggers often receive priority, but this is not a neutral process.
Script | What is most likely to trigger attention? |
|---|---|
FBL | A financial threat or opportunity for the individual firm |
TBL | Financial effects plus profitable ways to reduce negative externalities, often highlighted by external stakeholders |
SET | Financial viability, prevention of harm, positive externalities and stakeholder needs even without a profit improvement |
The AT&T example shows how Robert Greenleaf recognized inequitable career opportunities that existing scripts had not treated as problems. New recruitment and development practices could then become new routines. The lesson is to ask whether a situation affects people or ecosystems beyond the firm's immediate financial reports. Scripts can be revised; they need not permanently determine what a manager sees.
7.3 · Step 2: develop alternatives
Managers can do nothing, use an existing routine, or create a new response. Doing nothing can be appropriate when intervention is unnecessary, but convenience alone is not a sound justification. A programmed decision chooses from standard alternatives for a recurring situation, such as a reorder policy. A non-programmed decision develops a new way of addressing a novel or significant situation. A new policy can later support programmed decisions.
Repeated lateness illustrates the need to choose the appropriate response instead of the easiest one: illness, conflict or irresponsible behavior may need different actions. The chapter's Parlour Coffee case contrasts an alarm system with a new response addressing local homelessness and inequality after a break-in. The latter considered an underlying cause instead of only moving a symptom elsewhere.
More important and unfamiliar decisions usually justify more investigation. Urgency limits available time. Because SET concerns include less-established socio-ecological problems, SET managers may need new routines more often. That is a tendency, not a rule forbidding SET organizations from having standard procedures.
7.4 · Step 3: assess knowledge and agreement
Choosing an alternative requires evaluating both available knowledge and agreement on aims and means. Aims are desired outcomes; means are the plans or actions used to reach them. Managers may agree on the ultimate purpose but disagree about subgoals or the methods that best serve it. Neither disagreement nor an impressive quantity of data alone determines decision quality.
Condition | What decision-makers know |
|---|---|
Certainty | Outcomes, payoffs and their likelihoods are known |
Risk | There is some knowledge allowing estimates of alternatives, payoffs and probabilities |
Uncertainty | Outcomes, payoffs or their probabilities are unknown |
Most managerial decisions lie in a broad region of risk. Knowing that several outcomes are possible is not the same as knowing which will occur. Explicit knowledge can be articulated or codified in a manual, report or plan. Tacit knowledge is difficult-to-articulate insight developed through experience; intuition often draws on it. All three management approaches can use both forms. SET may rely more on tacit insight because complex well-being outcomes are difficult to quantify and its routines are less established.
Available information still needs interpretation. The Edsel example warns that managers can use research selectively to confirm prior worldviews. AI can process large datasets but may reproduce biased data or assumptions. Better processing does not eliminate incomplete knowledge, limited judgment or the need for human responsibility. A dilemma involves complex positive and negative trade-offs without an identifiable optimal choice. In a partial dilemma, managers know enough to distinguish better from worse alternatives despite being unable to identify an optimum. In a total dilemma, available information is insufficient to predict which choice will be better.
Five approaches to choosing
Approach | Knowledge and agreement | Characteristic response |
|---|---|---|
Classical rational | High knowledge, high agreement | List options, evaluate costs/benefits, choose the best |
Political | High knowledge, low agreement | Negotiate competing aims or means and reach a compromise |
Trial and error | Low knowledge, high agreement | Use modest experiments and incremental improvements |
Random | Low knowledge, low agreement | Make a chaotic choice without clear problem/solution fit |
Administrative | Partial knowledge and agreement | Satisfice within practical limits |
The classical rational approach assumes enough information and consensus for optimization. Decision trees, break-even analysis and other tools can support it, but calculations cannot decide which moral outcomes ought to count. The political approach involves bargaining and coalitions. Networking, compromise, image building, selective information, scapegoating, favor trading and alliances can be used. In SET, political negotiation can advocate for stakeholders rather than only the manager's personal gain.
Trial and error suits shared goals with inadequate knowledge of the best method. Small pilots and continuous improvement, or kaizen, allow learning without immediately committing to a large change. The random approach may select a fashionable solution before identifying the problem. This is different from a deliberately designed experiment. The administrative approach recognizes bounded rationality: people lack complete information and unlimited processing capacity. Satisficing selects an adequate alternative rather than proving it is optimal. It can conserve time and support learning when treated as revisable, but managers still need clear criteria for adequacy.
Constructed example: expected cost is useful but incomplete
River Workshop compares two delivery options. Option A costs 40,000 pesos if demand is normal, with probability 0.75, and 80,000 pesos if demand is high, with probability 0.25. Option B costs 55,000 pesos in either condition. Assume those two demand outcomes exhaust all possibilities.
Worked answer: expected cost of A = 0.75 × 40,000 + 0.25 × 80,000 = 50,000 pesos. Expected cost of B = 55,000 pesos. A has the lower expected financial cost by 5,000 pesos. This does not prove A is best overall: assess reliability, worker conditions, environmental effects and stakeholder purposes. The example is constructed to teach the chapter's probability-weighted reasoning; it is not a copied Ford problem or an official quiz.
7.5 · Step 4: implement, monitor and learn
Implementation has two distinct challenges: resistance to carrying out the decision, and resistance to admitting it was mistaken. Participation can improve the first by bringing knowledge and commitment into earlier steps. It is especially useful when the manager lacks information, members share relevant goals, their support matters, and sufficient time is available. FBL/TBL often justify participation instrumentally; SET also values participation and dignity in themselves. Participation does not guarantee unanimity, and experimental or invitational implementation can respect disagreement.
Monitoring asks whether the choice solved the original problem or created unintended effects. Escalation of commitment means persevering with a poor decision despite evidence it is failing. Its sources include information distortion, excessive persistence and administrative inertia. Information distortion highlights favorable feedback while downplaying bad news. Persistence becomes harmful when determination overrides relevant evidence. Administrative inertia keeps structures and systems because they already exist, with sunk costs reinforcing attachment.
Ways to reduce escalation include shared responsibility, explicit unacceptable-outcome criteria, performance milestones, openness to independent review and a willingness to reverse mistakes. A past investment cannot by itself justify another harmful investment. Treat decisions as learning experiments rather than personal legacies. Monitoring renews the first step in the process rather than merely proving the manager was right.
7.6 · Entrepreneurs and make-or-buy decisions
A make-or-buy decision asks whether to produce an input internally or purchase it. Transaction costs are costs that do not directly produce the output but protect against opportunism and uncertainty, such as contracts, certification, monitoring and inspections. Opportunism means misleading or cheating others for self-interest. Limited cognitive capacity means contracts cannot anticipate every possible condition.
Specialization can make buying efficient, but high purchasing safeguards and uncertainty can make internal production preferable. Compare the complete relevant costs, including the new management and equipment requirements of making. The Tall Grass Prairie Bread Company initially bought wheat from local farmers and milled flour itself, supporting quality and community relationships. As the organic market developed, purchasing flour became an option. This shows a revisable decision rather than a permanent rule that firms should always integrate.
FBL transaction reasoning emphasizes financial efficiency. TBL may add harm reduction where it improves finances. SET can value fair local relationships and social/ecological outcomes even where another input has a lower purchase price. Entrepreneurs also apply the theory from customers' perspectives: transparency, trustworthy service, personal relationships and generous return policies can reduce customers' transaction concerns. Outsourcing by established organizations can create opportunities for new suppliers.
Quick recall with answers
Is doing nothing always wrong? No; it can be appropriate when intervention is not warranted.
Does satisficing mean random guessing? No; it searches for an adequate option using available knowledge.
Which approach fits agreement on goals but little knowledge of methods? Trial and error.
Does admitting a mistake end decision-making? No; it creates knowledge for another cycle.
Are transaction costs identical to production costs? No; they safeguard exchange against uncertainty and opportunism.
Sources and attribution
Read the source chapter. Adapted study material from Bruno Dyck, Arran Caza, Patrick Adriel H. Aure, James Leaman, Joel Marcus, Shieradel Jimenez, and Anne Kato (2026), Management: Financial, Social, and Ecological Well-Being, 2nd edition, Sapajo Publishing and University of Manitoba Chiu Centre for Business Serving Community. Original: https://pressbooks.openedmb.ca/management/. Licensed CC BY-SA 4.0: https://creativecommons.org/licenses/by-sa/4.0/. This reviewer paraphrases and organizes the chapter; practice cases and questions are original study exercises. This adaptation is shared under CC BY-SA 4.0. It is not an official or predicted course exam.