The Illusion of Taking Action: Everyone's Moving. Where the Hell Are We Heading?
It's been a good few weeks. Tickets closed, a feature shipped, a couple of experiments running. You're busy, the energy is up, and it all feels like progress.
Then somebody asks where all of this is going. It could be an investor, a teammate, or your own voice at 11pm. You open your mouth and realize you don't have a clear answer.
That moment is what this post is about. Moving is easy to do and easy to feel good about. Knowing where you're moving to takes a different kind of effort, and it's easy to skip without noticing. You'll see why, what it costs, and how to stop without stalling.
The short answer
Action bias is the habit of doing something because doing nothing feels worse. Each task you start quiets the open question of where this is heading, and a few weeks later the question feels too late to ask. The way out is small. Put a short, regular stop on your calendar and ask one thing: is this still worth doing?
Moving is easier than asking where you're going
Not knowing is uncomfortable. You feel it as a tight stomach, restless hands, the urge to check something. Starting a task ends that feeling in seconds. You open the laptop, write the ticket, ship something small. The relief is real, so you do it again tomorrow.
Nothing in that loop checks whether the work points anywhere. The question is still sitting there, and you've stopped feeling it.
Building also got cheap. With AI tools a prototype takes an afternoon, so the pause that cost and effort used to force on you is gone. You can start ten things before you've decided whether any of them matter.
Say a big client asks you for a customer dashboard. Nobody checks whether other customers want one. You start building, because building feels better than asking. Keep that dashboard in mind, because the next three feelings all show up in it.
Three things that keep you from stopping
1. Busy feels like progress
Closed tickets, shipped features, a full calendar. You can count all of that. Where you're heading you can't count, so the countable stuff quietly takes its place. Ask yourself what changed for a customer this month. If your list of what you did is long and your answer to that is short, you've been busy. The dashboard has tickets closing every day and still no sign that anyone wants it.
If the drive behind all this output is a need to prove your worth, that goes deeper than any roadmap. I wrote about it in chronic productivity.
2. You're in too deep to stop
Every action leaves something behind: a hire, a codebase, a promise to a customer. Three months into the dashboard there's a designer on it, forty tickets and a launch date. Stopping now means admitting the last three months were a mistake, and that stings more than three more months of the same. Psychologists call this escalation of commitment: people keep investing in a path while the evidence against it grows. So "do we still need this?" becomes the one question you stop asking.
3. You don't want to be the one who slows things down
When everyone around you prizes speed, "wait, why are we building this?" makes you the blocker. So you keep it to yourself and match the pace. Everyone else is doing the same. The doubts come out in DMs after the meeting and almost never in it.
What it costs you
You end up defending a direction you never chose. A quarter went into that dashboard, and nobody sat down and decided to spend it.
The opposite failure exists too. I wrote about teams that talk about change and never touch it in What If Nobody Wants to Move the Needle?. Both ends hurt, and this post is about the end where you never stop long enough to check the map.
How to stop without stalling
Four habits do most of the work, and none of them needs an offsite.
- Put the stop on the calendar. Thirty minutes every two weeks, one question: where is this heading, and is it still worth doing? Short planned stops beat the big panicked one six months in.
- Ask the start-over question. If you weren't already doing this, would you start it today? If the answer is no, ask what you're protecting by continuing.
- Sort by how hard it is to undo. Amazon's own leadership principle says many decisions are reversible and don't need extensive study. So if you can undo it in a week, move fast. If it will eat a quarter, a hire or your reputation, write one paragraph on where it leads before you start.
- Set the stop rule up front. "If we don't see X by this date, we stop." Decide it while you're calm, so stopping later is just following the plan.
The pause is the hard part
The first thing you'll feel when you stop is restlessness. That itch is the not-knowing again, the same one that sent you to the laptop in the first place. Stay with it for a minute or two. Breathe, and notice where it sits: your stomach, your neck, your throat. Let the question stay open before you reach for a task to make it go away. It's the same skill I work on in Emotional Wisdom, and it applies in a roadmap meeting too.
Before your next sprint, kickoff or big build, take ten minutes with no laptop and write one sentence: where is this heading? If you can't write it, that's your next piece of work.
Key takeaways
- Before a big build, write one sentence on where it leads. If you can't, that sentence is the work.
- Ask "would I start this today?" about anything you're already deep in.
- Move fast on what you can undo in a week. Slow down for what eats a quarter, a hire or your reputation.
- Book a 30-minute direction check every two weeks and keep it to one question.
- Set a stop rule before you start, while you're calm.
- When a hard question sends you to the laptop, wait two minutes first.
Definitions
Action bias: The pull to do something, even when pausing or doing nothing would give a better result.
Escalation of commitment: Continuing to invest in a chosen course of action despite growing evidence that it is the wrong one.
Sunk cost: Time, money or effort already spent that can't be recovered, and that tends to push people to keep going.
Stop rule: A condition agreed before a project starts that says when you will end it.
Frequently asked questions
What is action bias in simple terms?
Action bias is the habit of choosing to do something over waiting or thinking, because acting feels better. The cost shows up later, when weeks of activity turn out to point nowhere in particular. It is strongest when the situation is uncertain and other people are watching, which is exactly when a pause would help most.
Is a bias for action a bad thing?
No. It works well for decisions you can reverse cheaply, like a small test, a draft or a one-week experiment. It goes wrong when you apply the same speed to choices that are slow and costly to undo, such as a hire, a rebuild or a public promise. Match your speed to how hard the decision is to take back.
How do I know if I'm busy or actually making progress?
Finish one sentence: "This month, a customer can now do ___ that they couldn't before." If you struggle to fill the blank, you've been busy. Progress shows up as a change someone outside your team would notice, and a long list of closed tickets rarely passes that test on its own.
What is the difference between a useful pause and analysis paralysis?
A useful pause has a time limit, one question and a decision at the end. Analysis paralysis has none of the three, so it keeps gathering information and never lands. Thirty minutes that end with "we continue," "we change course" or "we stop" is a pause. Three weeks of extra research with no decision is stalling.
Does AI make action bias worse?
AI can make it worse. AI tools lower the cost of starting, which removes a natural checkpoint. When a prototype took a month, someone had to justify it first. When it takes an afternoon, nobody asks. It becomes easy to build a lot of things before anyone checks whether they were needed, so the direction question matters more, not less.
How do I bring up the direction question without sounding negative?
Frame it as protecting the effort already spent. Something like: "I want to make sure the next three months of work land. Can we take 20 minutes to say where this is heading?" Asking about the destination is easier for people to hear than questioning whether the current work is good.
