Most bad decisions are not made badly. They are made without the context that would have changed them.
By the time a leadership team sits down to decide, most of the outcome is already set. Not by the quality of the discussion, but by what reached the room and what did not. The numbers that were prepared. The assumptions nobody wrote down. The objection that was filtered out two levels below because it was uncomfortable.
Research backs the intuition. McKinsey’s global decision survey finds a majority of executives rate most of their decision-making time as ineffectively used, and fragmented, filtered or missing context is one of the quiet drivers: the meeting spends its energy reconstructing reality instead of deciding[U]. Its 2026 operating-model research makes the same point for AI: the value sits where information flows and decisions are redesigned, not where another tool is added[O].
The context problem
In most organisations, context is fragmented by default. Finance holds one part of the picture, operations another, and the people closest to the customer hold a third that rarely makes it into the deck. What arrives in the meeting is a summary of summaries, shaped by whoever prepared it and by what they believed the room wanted to hear.
This is not a character flaw. It is a structural condition. Information loses fidelity every time it moves up a level, and it loses honesty every time it passes through someone with a stake in the outcome. The result is a decision made on clean slides and incomplete reality.
Why more meetings do not fix it
The usual response to a stalled decision is another meeting with more people. This makes the problem worse. More perspectives without a shared basis of facts produce more opinions, not more clarity. The discussion expands while the decision quality stays flat.
The missing ingredient is rarely another viewpoint. It is context readiness: the state where the relevant facts, constraints, assumptions and disagreements are visible before the decision is framed, not discovered after it fails.
What context readiness looks like
Context readiness is not a data lake or a dashboard. It is a discipline with a few identifiable marks:
- The real decision is named. Not the topic, the decision. Most stalled situations circle a topic because nobody has stated what is actually being decided.
- Assumptions are written down and separated from facts. Certainty, supposition and doubt are three different things, and treating them as one is how confident mistakes get made.
- Constraints are surfaced early. Budget, politics, capacity and regulation shape the option space. Pretending they arrive later only delays the collision.
- Dissent is collected before the meeting, not managed during it. The most useful objection is usually held by someone who will not raise it in front of the room.
| What existed in the organisation | What reached the meeting |
|---|---|
| Primary facts and raw numbers | A summary of summaries |
| Working assumptions | Undocumented, mixed with facts |
| Known constraints | Discovered during the discussion |
| Honest dissent | Filtered out two levels below |
The practical takeaway
Before your next consequential decision, ask one question: if this decision fails in twelve months, what will we say we knew but did not use? Then go and collect exactly that, before the meeting. The discipline sounds simple. It is also the difference between deciding on reality and deciding on presentation.
Context is also why AI stalls
The same readiness gap now decides AI outcomes. Stanford’s 2026 AI Index shows adoption at 88 percent while agent deployment stays in the single digits across most functions[ST], the tooling arrived, the organisational context did not. Deloitte finds 85 percent of companies expecting to customise agents to their specific business[DE], which is a polite way of saying generic AI without your context is generic output. Whether the decision-maker is a human in a meeting or a model in a workflow, the constraint is identical: fragmented context in, poor judgment out.
What I see in the field
The most reliable predictor of a wasted decision meeting is visible before it starts: three people arriving with three different versions of the numbers. Context readiness sounds unglamorous, but it is the cheapest performance upgrade a leadership team can buy, and it is also the foundation every serious AI initiative stands on.
Context is our starting point in both practices: what your organisation knows, where it lives, and whether it reaches the decision on time. If your meetings reconstruct reality before they can decide anything, begin here.
Personalised answer within 48 hours.
Sources
- McKinsey & Company, Decision Making in the Age of Urgency (global survey), 2019. mckinsey.com
- McKinsey & Company, The Operating Model Advantage: Why AI Winners Are Rewiring Their Organizations, July 2026. mckinsey.com
- Stanford HAI, The 2026 AI Index Report. hai.stanford.edu
- Deloitte, The State of AI in the Enterprise 2026 (survey of 3,235 leaders). deloitte.com
