The Pre-Decision: Why What Happens Before the Choice Matters Most

Executive summary

Most decisions feel like they are made at the conference table, in the hiring panel, or in the moment of purchase. In reality, a large share of decision quality is locked in before formal deliberation begins, through the way options are framed, contrasted, and made salient. Research on framing effects, context effects, defaults, and decision scaffolding shows that the “entry” to a decision quietly shapes what people see, how they feel about risk, and which evidence they treat as relevant.[^1][^2][^3]

This report argues that there is no neutral frame, that pre-decision conditions routinely matter more than analysis quality, and that organizations need explicit pre-decision design. It synthesizes work on framing (Tversky and Kahneman), context effects (attraction and compromise), choice architecture and defaults (Thaler and Sunstein), the Mediating Assessments Protocol (MAP), implementation intentions, the outside view and reference class forecasting, pre-mortems, counterfactual thinking, and decision scaffolding. It then translates these ideas into practical tools for leaders who want better decisions in high-stakes environments.[^2][^4][^5][^6][^3]

A hook: decisions that were over before they began

Consider a leadership team “deciding” whether to acquire a competitor. The deck they receive includes two options: proceed with Acquisition A or stay independent. The modeling assumes aggressive synergies and cites several “comparable” deals that created value. Alternatives such as partnering, minority investment, or acquiring a smaller firm just below the radar never make it into the pack.

By the time the team meets, the problem has already been framed as a binary choice, the baseline is a rosy synergy scenario, and dissenters are implicitly arguing against the momentum embedded in the documents. What looks like free deliberation is actually a late-stage negotiation inside a structure that was fixed weeks earlier.

Behavioral research suggests this pattern is not an exception but the norm: how a decision is posed, which options are present or absent, and what is made salient create predictable shifts in preference before any conscious weighing of pros and cons begins. The “pre-decision” phase, where problems are framed and choice sets are constructed, is where much of the real action happens.[^1][^2]

Framing as the unavoidable starting condition

The Asian Disease problem

The classic demonstration of framing effects is Tversky and Kahneman’s Asian Disease problem. Participants are told that a disease is expected to kill 600 people and must choose between two programs:[^7][^2]

• In a gain frame, options are described in terms of lives saved. A typical version is: “If Program A is adopted, 200 people will be saved. If Program B is adopted, there is a one-third probability that 600 people will be saved and a two-thirds probability that no one will be saved.” Most respondents prefer the certain gain.
• In a loss frame, extensionally equivalent options are described in terms of deaths: “If Program C is adopted, 400 people will die. If Program D is adopted, there is a one-third probability that nobody will die and a two-thirds probability that 600 people will die.” Here, most respondents prefer the risky option.[^7][^2]

Logically, the options are the same: saving 200 is equivalent to losing 400 when 600 are at risk. Yet the shift from “lives saved” to “lives lost” reverses preferences, consistent with prospect theory’s finding that people are risk averse in gains and risk seeking in losses.[^2]

Replications using careful wording controls confirm that this framing effect is robust, despite debates about linguistic nuances. Changing a few words before deliberation begins is enough to flip the majority view on a life-or-death public policy.[^8][^9][^7]

No neutral frame

The Asian Disease study illustrates a broader point: there is no view from nowhere. Any description emphasizes some reference point, outcome metric, or time horizon rather than others. Framing determines whether a decision feels like avoiding a loss, capturing a gain, or maintaining a status quo.[^2]

Kahneman’s broader work on System 1 and System 2 thinking shows that people rely heavily on fast, automatic interpretations of problems, with slow analytic reasoning engaged only occasionally. Once an initial frame is set, confirmation processes and “what you see is all there is” dynamics mean that later deliberation often rationalizes that frame rather than rethinking it.[^10][^11][^12]

In that sense, the relationship between framing and free choice is similar to the relationship between a user interface and “free” navigation. People can, in principle, click anywhere, but the way options are laid out, labeled, and highlighted shapes what they are likely to do. Framing is the cognitive UI of decisions, and neutrality is impossible; the only question is whether the frame is designed deliberately or by accident.

Context effects: how option sets shape evaluation

Even when the description of each option is fixed, the context created by other options strongly affects choice. Context effects violate the independence assumptions of classical rational choice and are central to understanding pre-decision architecture.[^13][^1]

Attraction and compromise effects

In the attraction effect, adding a dominated “decoy” option makes a similar but superior option more attractive. For example, if option X is better than Y on both price and quality, introducing a clearly inferior version of X (worse on both dimensions) increases the share of choices for X, even though Y has not changed.[^4][^13][^1]

In the compromise effect, introducing a more extreme option makes a middle option more appealing as a “safe” compromise. When consumers face a low-priced, low-quality option and a high-priced, high-quality option, adding a very expensive, premium alternative shifts choice toward the mid-tier product, which now feels like a reasonable middle ground.[^14][^13][^1]

Trueblood and colleagues show that attraction, similarity, and compromise effects all emerge from basic decision processes, not just from marketing tricks, highlighting that preference is inherently context sensitive. These effects appear in multiattribute choices across domains including consumer products and perceptual tasks, not just hypothetical questionnaires.[^13][^1]

Implications for decision environment design

Context effects mean that decision quality depends partly on which options are allowed into the room and how extreme they are. When leadership is given a false trichotomy of “do nothing, overinvest, or moderately overinvest,” the presence of an obviously reckless overinvestment can make a still-aggressive middle option seem prudent.

From a design perspective, this has two sides:

• Risks: Poorly curated choice sets can steer teams toward systematically biased compromises or decoy-driven choices that favor one stakeholder’s preferred option.
• Opportunities: Carefully including “principled extremes,” such as a low-cost, low-risk path and a high-return, high-risk path, can help highlight trade-offs and make genuine middle-ground strategies more meaningful.

In either case, what happens before deliberation, when the option set is defined, has a powerful influence on where people land.[^4][^1]

Defaults, anchors, and salience: the invisible architecture

Defaults and libertarian paternalism

Thaler and Sunstein popularized the idea of choice architecture, arguing that the environment in which people choose is unavoidably structured and can be designed to nudge people toward better outcomes while preserving freedom of choice. A central tool is the default: what happens if the decision maker does nothing.[^15][^3]

Empirical work summarized in Nudge shows that default settings for pension enrollment, organ donation, or savings rates have enormous effects on behavior, even when opting out is easy and costless. Because many people stick with defaults out of inertia, implied endorsement, or simple inattention, whoever sets the default effectively makes the decision for a large fraction of the population.[^3]

Later work in political and legal theory frames this as “libertarian paternalism,” emphasizing that environments always influence choices and that making this influence pro-welfare, transparent, and reversible is preferable to pretending neutrality. The inevitability of choice architecture means that the pre-decision state, including defaults, is already a policy.[^16][^17][^18]

Salience, availability, and anchors

Beyond defaults, the salience of information and anchors presented before a decision shape what enters deliberation. Kahneman’s account of the availability heuristic shows that people judge frequencies and risks based on how easily examples come to mind, which is itself influenced by recent headlines, vivid stories, and what the environment highlights.[^12][^10]

Anchoring effects show that even arbitrary numbers presented before a judgment can pull estimates in their direction, from real estate valuations to legal damage awards. Although these classic anchoring studies are not pre-decision design in organizations, they highlight a broader principle: early exposure creates a reference point that later “analysis” rarely fully corrects.[^12]

In practice, this means that the first forecast, the initial vendor quote, or the opening scenario in a strategy offsite anchors subsequent discussion. If the first model assumes 20 percent growth, counterproposals tend to be framed as “slightly higher” or “slightly lower,” not as a wholesale reframing of the opportunity.

What good pre-decision design looks like

A well-structured pre-decision phase treats framing, context, and salience as design parameters rather than accidents. Several tools from decision science can be combined into a repeatable pattern.

The Mediating Assessments Protocol (MAP)

Kahneman and coauthors propose the Mediating Assessments Protocol as a way to reduce noise and bias in strategic decisions, especially about people and projects. Instead of jumping straight to an overall judgment (for example, “Hire this candidate” or “Approve this acquisition”), MAP asks decision makers to:[^5]

1. Define a small set of mediating assessments that are predictive of success but can be judged independently (for example, problem solving, stakeholder management, cultural fit for a hire; market attractiveness, competitive position, execution capability for a strategy).
2. Gather data and make structured ratings on each dimension before any global verdict is discussed.
3. Only after individual assessments are recorded, bring the group together to discuss patterns and make a final call.

By structuring the pre-decision phase around independent assessments, MAP reduces halo effects, anchoring on overall impressions, and dominance of the loudest voice in the room. It effectively scaffolds the entry into the decision so that System 2 has real inputs to work with instead of post hoc rationalizations.[^5]

The outside view and reference class forecasting

Kahneman and Lovallo argue that managers are systematically overoptimistic, treating each project as unique and focusing on inside details rather than the distribution of outcomes for similar past efforts. They distinguish between the inside view, which builds narratives from case-specific information, and the outside view, which starts from base rates of comparable cases.[^19]

Reference class forecasting operationalizes the outside view in three steps: identify a relevant reference class of past projects, establish the probability distribution of outcomes in that class, and position the current project within that distribution. Research on megaprojects and infrastructure investments shows that using reference class forecasting can reduce average cost overruns dramatically, for example from around 38 percent to about 5 percent in some UK transportation projects.[^20][^21][^6][^22]

Embedding an outside-view step into pre-decision routines acts as a structural debiasing mechanism. Before debating the fine points of a business case, teams are required to confront how similar efforts have actually performed.

Implementation intentions and pre-commitment

Implementation intentions, as defined by Gollwitzer, are “if–then” plans that specify the situational cue and the intended response (for example, “If it is Monday at 9am, then I will review the risk register”). They go beyond general goal intentions like “We intend to manage risks carefully” by tying action to concrete triggers.[^23][^24]

Research shows that forming implementation intentions improves goal attainment across health, academic, and organizational behaviors, in part by “passing the control of one’s behavior on to the environment.” In decision contexts, implementation intentions can be used to pre-commit to good process before incentives to cut corners kick in, for example:[^24][^23]

• “If a capital project exceeds a given size, then we will run a reference class forecast.”
• “If a strategic decision is above a certain threshold of irreversibility, then we will conduct a pre-mortem.”

These if–then rules live squarely in the pre-decision phase: they determine which tools are triggered and which evidence is required before a decision can be tabled.

Decision scaffolding as environment design

The broader concept of scaffolding in educational psychology refers to structures that support learners’ cognition and motivation, often by guiding attention, breaking problems into manageable parts, and making system decisions transparent. When applied to organizational choice, decision scaffolding refers to environment design that shapes what people attend to and in what order before they decide.[^25][^26]

Examples of decision scaffolding include:

• Templates that force articulation of alternative options and explicit “do nothing” baselines.
• Dashboards that juxtapose base-rate data and case-specific forecasts.
• Prompts in digital tools that ask for mediating assessments separately before recording an overall judgment.

Evidence from scaffolding research shows that such structures can improve performance and self-efficacy by clarifying goals, highlighting important information, and supporting self-regulation. In organizations, analogous scaffolds can ensure that pre-decision time is spent constructing a high-quality representation of the problem rather than simply polishing a preferred narrative.[^26][^25]

The pre-mortem as pre-decision hygiene

Gary Klein’s pre-mortem technique is a paradigmatic example of pre-decision design. Instead of asking, “What might go wrong with this project?”, teams are asked to imagine that it is some months in the future and the project has already failed catastrophically. Their task is to write down all the reasons why it failed.[^27][^28]

This use of “prospective hindsight” increases the number of identified risks compared with standard risk analysis, in some studies by 30 percent or more. By assuming failure, it legitimizes negative thinking and reduces social pressures toward optimism and agreement, which often suppress dissenting information.[^29][^30][^31][^27]

Kahneman has endorsed the pre-mortem as a particularly effective way to counter groupthink and optimism bias, precisely because it alters the frame of the discussion before commitment. Rather than evaluating an advocated plan, the team is temporarily placed in a world where the plan has failed and is asked to treat that as a fact.[^10]

In practice, a pre-mortem is most powerful when institutionalized as a gate: for major decisions, it is not an optional workshop but a required step before approval. This turns a single exercise into a standing improvement in the organization’s decision entry architecture.

Counterfactual thinking and learning before the next decision

Counterfactual thinking, the mental simulation of “what could have been,” serves both emotional and preparatory functions. Upward counterfactuals (imagining better outcomes) can increase regret but also promote learning by highlighting controllable causes of failure and triggering plans for improvement.[^32][^33]

Research by Roese, Markman, and others shows that upward counterfactuals can enhance persistence, preparation, and performance on subsequent tasks, particularly when people feel dissatisfied with their prior performance and see future opportunities to act differently. Counterfactual analysis can inform the design of future pre-decision processes, for example:[^33][^34][^32]

• “If we had required an outside view, we would have recognized that our schedule was wildly optimistic.”
• “If dissenting opinions had been surfaced earlier, we might have avoided this escalation.”

However, counterfactuals can also lead to dysfunctional patterns, such as frequent tactical changes without strategic adjustment or rigidity in long-term strategy. The key is to channel counterfactual thinking into structural improvements to decision entries rather than into blame or cosmetic tweaks.[^34]

How much is decided before deliberation?

Synthesizing these strands suggests a rough answer to the question of how much decision quality is determined before formal deliberation:

• Framing studies show that simply changing wording or reference points can reverse majority preferences without altering the underlying payoffs.[^7][^2]
• Context effects show that adding or removing options can systematically shift choices, independent of people’s stated preferences over attributes.[^1][^13]
• Defaults and anchors demonstrate that starting points and implied recommendations have large behavioral effects even when opt-out is easy.[^15][^3]

Combined with evidence that most everyday judgments rely on fast, automatic processing rather than extended analysis, it is reasonable to treat the pre-decision phase as setting strong priors that deliberation only partially updates. While it is impossible to put an exact percentage on “how much” is determined up front, the empirical pattern is clear: if organizations want better decisions, they must invest disproportionate effort in designing how decisions start, not just how they end.[^10][^12]

How organizations sabotage themselves before they start deciding

Many organizations unintentionally undermine decision quality in the pre-decision phase through recurring patterns.

1. Single-option advocacy: Strategy or product teams present one polished recommendation instead of a genuine choice set, turning the decision into a yes/no vote on a favored plan. This encourages motivated reasoning and makes it socially costly to raise alternatives.
2. Unexamined defaults: Renewal of vendors, continuation of projects, or automatic rehiring are treated as passive baselines, even when circumstances have changed.
3. Lack of outside view: Business cases rely exclusively on inside narratives and tailored scenarios, ignoring base rates from comparable efforts and thereby falling prey to the planning fallacy.[^35][^6][^19]
4. Unstructured judgment: Hiring, promotion, and investment decisions lean on global impressions formed early in the process, with limited decomposition into predictive factors.[^5]
5. Token risk analysis: Risk registers and RACI charts are created after the decision is effectively made, serving to document rather than challenge the plan.

Each of these failure modes operates before or at the very moment that formal deliberation begins. By the time a committee convenes, the space of possibilities has been narrowed, reference points have been set, and social and cognitive momentum favor a particular path.

Structural fixes: designing better decision entries

Improving pre-decision design requires structural interventions that change how problems arrive at the table.

A practical starting toolkit includes:

• Explicit framing guidelines: Require decision documents to state the reference point (status quo, current forecast), specify whether outcomes are framed as gains or losses, and include symmetric descriptions where possible. For high-stakes public or ethical decisions, test alternative frames with pilot groups to detect reversals like those in the Asian Disease problem.[^9][^2]
• Curated option sets: For major decisions, insist on at least three options: status quo, a conservative move, and an ambitious move. Ask proponents to articulate the case against their preferred option and to specify what would have to be true for each alternative to win.
• Default transparency: Make defaults explicit and justify them. In policy and product contexts, follow Thaler and Sunstein’s criteria for good nudges: transparent, easy to opt out, and aligned with people’s long-term interests.[^17][^3][^15]
• Embedded outside view: Standardize the use of reference class forecasting above certain thresholds of budget or risk, drawing on project databases to construct relevant reference classes and distributions.[^21][^6][^20]
• MAP-based templates: Redesign hiring, promotion, and investment processes around mediating assessments, separating fact gathering from overall evaluation and sequencing discussion to avoid early verdicts dominating the process.[^5]
• Institutionalized pre-mortems: Make pre-mortems a mandatory step for strategic bets, using prospective hindsight to surface hidden risks and failure modes before commitment.[^28][^27][^10]
• Counterfactual reviews: After major decisions, run structured counterfactual analyses focused on improving future pre-decision design, not assigning blame. Capture lessons about missing options, ignored base rates, or mis-specified frames and encode them into future templates.[^32][^33]

These interventions serve as decision scaffolds: they shape what information is produced, in what order, and under which frames, so that when deliberation begins, the decision space is richer and less distorted.[^25][^26]

Conclusion: if you want better decisions, design better decision entries

The evidence from framing, context effects, defaults, outside views, pre-mortems, and decision scaffolding converges on a simple claim: the quality of most decisions is largely determined before deliberation begins.[^6][^3][^1][^2][^5]

There is no neutral frame and no neutral choice architecture. The question is whether organizations will continue to let pre-decision conditions be set by happenstance, habit, and internal politics, or whether they will treat decision entries as a first-class design problem.

For leaders who care about decision quality, this means shifting effort and attention upstream: designing how problems are posed, how options are generated, what data is required, and which scaffolds support judgment. Done well, this work is largely invisible, but its effects show up downstream as fewer surprises, fewer reversals, and more decisions that look wise in hindsight.

---

References

1. Theories of Context Effects in Multialternative, Multiattribute Choice - Jennifer S. Trueblood, 2022 - Over the past several decades, researchers in psychology, neuroscience, marketing, and economics hav...

2. [PDF] The Framing of Decisions and the Psychology of Choice

3. Nudge (book) - Wikipedia

4. A bargaining rationale for the attraction and compromise ...

5. A Structured Approach to Strategic Decisions - A practical, broadly applicable approach to reducing errors in strategic decision-making. We call it...

6. Reference class forecasting - Wikipedia

7. Evaluating framing effects - This paper examines Tversky and Kahneman's well-known Asian disease framing problem (A. Tversky, D. ...

8. The Most Famous Framing Effect Is Robust To Precise Wording - In an intriguing new paper, David Mandel suggests that the most famous demonstration of framing effe...

9. Judgment and Decision Making, Vol. 13, No. 6, November 2018, pp. 529–546

10. Thinking Fast and Slow: Kahneman explained - Kahneman estimates that around 96% of our thinking runs through System 1. Four percent through Syste...

11. Of 2 Minds: How Fast and Slow Thinking Shape Perception and ... - In psychologist Daniel Kahneman's recent book, he reveals the dual systems of your brain, their pitf...

12. Thinking, Fast and Slow - Wikipedia - Its main thesis is a differentiation between two modes of thought: "System 1" is fast, instinctive a...

13. [PDF] Context effects are fundamental to decision-making

14. The Research Review and Prospect of Compromise Effect - Discover the rational choice theory and context effect in consumer decision-making. Explore the comp...

15. Choice architecture - Wikipedia

16. Libertarian Paternalism and the Problem of Preference Architecture - Libertarian Paternalism and the Problem of Preference Architecture - Volume 52 Issue 2

17. ‘Good Nudge Lullaby’: Choice Architecture and Default

18. Nudge, Choice Architecture, and Libertarian Paternalism - By all external appearances, Nudge is a single book-two covers, a single spine, one title. But put t...

19. [PDF] Timid Choices and Bold Forecasts: A Cognitive Perspective on Risk ... - The cognitive analysis accepts choice between gam- bles as a model of decision making, but does not ...

20. Reference class forecasting: promises, problems, and a research agenda moving forward - Reference Class Forecasting (RCF) has emerged as a prominent tool to counter chronic optimism in pro...

21. [PDF] Top-Ten Behavioral Biases in Project Management: An Overview - I developed this into a practical tool called. "reference class forecasting" (Flyvbjerg 2006) and in...

22. 1

23. [PDF] Implementation Intentions - Prospective Psychology - How good intentions can be implemented effectively has been analyzed in recent research on goal stri...

24. [PDF] Implementation Intentions Peter M. Gollwitzer New York University ... - Whereas goal intentions specify what one wants to do or achieve (i.e., “I intend to achieve. X!”), i...

25. Scaffolding - A problematic, yet common, assumption among educational researchers is that when teachers provide au...

26. Who regulates? Effects of scaffolding in system- and self ... - Frontiers - Scaffolding then serves as a tool to reveal system decisions and thus make external regulation trans...

27. Pre-mortem: how to anticipate failure with prospective hindsight - Most people are familiar with post-mortem documentation, where team members come together at the con...

28. Pre-Mortem - Evaluate potential risks and possibilities of negative outcomes by evaluating a decision from the pe...

29. 6 Qs to Conducting a Premortem: Who, What, When, Where, Why ... - Just because the outcome of a business project or strategic initiative is acceptable doesn’t mean th...

30. Premortem Analysis: Anticipate Failure to Achieve Success - A premortem analysis helps you get things right before they go wrong! Learn how to use this tool to ...

31. Prospective Hindsight and the Pre-Mortem - Whiteark - Mark Easdown writes about Prospective Hindsight and the Pre-Mortem. Given the importance and value o...

32. [PDF] Counterfactual Thinking About Actions and Failures to Act - Counterfactual thinking may serve a preparatory function in helping people to establish the causes o...

33. [PDF] counterfactual thinking - WyoScholar - When holding a “do your best” goal (i.e., a task of unfixed difficulty), experiencing positive affec...

34. Counterfactual thinking and decision making - PubMed - NIH - Counterfactuals may cause decision makers to become liberally biased (i.e., capricious) in terms of ...

35. How Cognitive Biases Shape Decisions & How to Overcome Them - A cognitive bias is a systematic error in thinking that affects the decisions and judgements that pe...