Why Improvisation Fails When Stakes Are High

We’ve all been there. Or close enough.

I certainly have.

Years ago, I delivered an online training for a UK-based company on material I could teach in my sleep. I had other priorities at the time and assumed that any misalignment with the audience, who were different from my usual clients, could be fixed live. I improvised.

I also felt close to the source material. They wanted a training that blended public speaking with internal visibility. A topic I know well. Too well, as it turned out.

I spoke from experience. But I spoke from my experience, not from theirs.

I relied on familiarity rather than on audience alignment. There was vagueness where there should have been structure. I assumed relevance instead of checking it. I trusted my instinct to carry work that should have been done upstream.

The feedback was lukewarm. One participant called it wishy-washy.

Ouch. Fair. I had rested on my laurels, and it showed.

Here’s the underlying issue.

Many capable people rely on improvisation to compensate for weak alignment before the moment arrives. Not because they’re careless. Often because they’ve been successful before.

They fill gaps. They smooth uncertainty. They “bring value” using instincts that have worked in other rooms, with other audiences, under different stakes.

There are reasons for this. Overconfidence, sometimes. But more often avoidance. Avoidance of slow, unglamorous preparatory work. Procrastination disguised as flexibility. Vagueness mistaken for openness. Work avoidance wearing the costume of confidence.

The mindset sounds harmless.

“I’ll fix it live.”

Sometimes that instinct saves you.

Until it doesn’t. No explosion. Just no second chance.

After more than fifteen years of running workshops, here’s what I’ve learned the hard way. Improvisation doesn’t fail because it’s bad. It fails because it’s asked to do work it was never meant to do.

High-stakes contexts  are unforgiving of late course correction.

Of visible recalibration.

Of uncertainty about what actually matters when misalignment was preventable.

This doesn’t make improvisation the enemy. Far from it. Improvisation is a powerful skill. I’ve improvised live narration of a chess match with Garry Kasparov. Twice. In front of thousands. No script. No safety net. It worked. It was exhilarating.

But it worked because alignment was already there. The context was clear. The audience expectation was clear. The role was clear.

Improvisation was an asset. Not a repair strategy.

Heroics feel good.

Heroics are unreliable.

This is why I eventually built a tool for myself. Not a script. Not a template. A diagnostic.

FastCat exists to force alignment before the stakes rise. A short self-coaching process that surfaces what actually matters before you walk into the room. The context you’re operating in. Who the audience really is. Where the real tension sits.

Its purpose isn’t to remove improvisation.

It’s to stop improvisation from carrying responsibility it cannot reliably hold.

Because when the moment is critical, “I’ll fix it live” isn’t confidence.

It’s a gamble.