Leadership question: Why does speed often reduce confidence rather than improve it?
Answer: Speed without explicit judgement standards creates fragility, not performance.
Your organisation didn’t just get more informed. It got faster. That’s what happens when AI enters decision workflows. Analysis that once took weeks now takes days. Options surface in hours. Scenarios arrive pre-modelled. On paper, this looks like progress.
In many ways, it is. But there’s a quieter pattern emerging alongside the speed: decisions are being reversed more often, and confidence in direction is thinning rather than strengthening.
The discomfort rarely sounds like a statement. It sounds more like this:
- “Are we moving fast because we’re clear, or because it’s now easy to move?”
- “What standard did this actually pass?”
- And “would I make the same call if the outcome were uncertain rather than favourable?”
That intuition is pointing at something real and this isn’t just a subjective feeling.
Research from Boston Consulting Group shows that while organisations are rapidly accelerating decision workflows through AI, only a minority are realising sustained value from those decisions. (Boston Consulting Group 2024 “Where’s the value in AI?”). Adoption and speed are outpacing the leadership disciplines required to turn faster decisions into better ones. In other words, acceleration is happening. Confidence and durability are not guaranteed.
Not because leaders stop caring about quality, but because the standards that would normally be surfaced through discussion, challenge, and friction become implicit. Assumed. Unspoken.
And implicit judgement is fragile judgement.
You might have made the same decision in a slower process: a week of debate, documented assumptions, visible trade-offs. The process itself forced clarity. You knew what you believed, what you were uncertain about, and what risks you were consciously taking.
Now the same decision is made in two days. Often with better data. But with less shared clarity about what “good” actually meant.
AI compresses decision cycles brilliantly.
What it does not do is clarify decision quality thresholds.
Judgement still requires human clarity:
- What matters most here?
- What level of evidence is sufficient?
- What risks are we prepared to own?
- What trade-offs are we explicitly accepting?
Speed doesn’t answer those questions. Pressure doesn’t either.
Slow is steady, steady is smooth, smooth is fast.
This is where organisations can get themselves into trouble, by celebrating speed as a proxy for effectiveness. Speed is valuable, but only when it is in service of an explicit standard. Otherwise, you’re just moving quickly towards decisions you’ll need to revisit.
There’s another pattern that shows up as speed increases. Leaders often assume judgement is shared. “Everyone knows what good looks like here.”
That assumption tends to hold, right up until decisions accelerate.
Suddenly, two leaders are making calls at the same pace but with different internal bars. One is comfortable acting on partial evidence. Another assumes the data has already screened for downside. Same speed. Different standards. Neither realises the mismatch until outcomes diverge.
That’s where confidence erodes. Not because decisions are wrong, but because no one is quite sure what game is being played.
In a faster system, good judgement becomes less about instinct and more about explicit clarity:
- What counts as “enough” evidence for this type of decision?
- How reversible does it need to be before we treat it as final?
- When does speed stop being an advantage and start being a liability?
- What makes the difference between a decision that moves today and one that waits?
These aren’t soft questions. They’re the core discipline of leadership. And they become more important as systems get faster.
The guardrails are simple:
- Define “good enough” before you accelerate. Not in hindsight. Before the decision is in motion.
- Be explicit about reversibility. Some decisions can be undone cheaply. Others can’t. Speed should follow that distinction.
- Pause where judgement is unclear. Not every decision deserves to move at the speed the model allows.
The standard this sets is straightforward: Speed is only a virtue when judgement is explicit.
When standards are implicit, speed becomes risk. Decisions move fast, reverse often, and quietly weaken organisational trust.
The companies navigating AI adoption well aren’t the ones who moved slowest. They’re the ones who made the trade-off between speed and clarity explicit. They decided what could accelerate without weakening reasoning and held the line on the rest.
That isn’t resisting AI. It’s using AI as what it should be: an accelerant for good judgement, not a substitute for it.
Leadership Instrument: The “Good Enough” Rule
When to use it: Before accelerating a decision cycle or delegating a consequential decision.
Ask three questions:
- “What evidence would be sufficient for this decision?”
- “What evidence would be insufficient?”
- “What is the reversibility level; reversible, partially reversible, irreversible?”
What it changes: Speed becomes disciplined rather than impulsive.
What to listen for: “We can decide now and figure it out later.”
Leadership standard: Speed is only a virtue when judgement is explicit.