Find what keeps making smoke
Planted 02026-05-31
Stop fixing symptoms and find the conflict underneath them.
Imagine your life, your team, or your company is a complicated machine.
Smoke is coming out of six different places.
One person says, “Patch that pipe.”
Another says, “Tighten that bolt.”
Another says, “Buy a new machine.”
Stop. Shut up. Be quiet.
Before we run around, let’s ask a simpler question:
What is actually going on inside the machine?
Because the smoke is not the problem. The smoke is how the problem shows itself.
The real question is: what is happening underneath that keeps making smoke?
That is the whole point of this process. It is a disciplined way to stop fixing symptoms and find the conflict that keeps producing them.
1. What should change?
Don’t start with:
“What should I do?”
That question is too early.
Start with:
“What keeps happening that I don’t like?”
Not theories. Not labels. Not diagnoses.
Just observable facts.
“Communication is bad” is not good enough yet. That is a foggy statement. You have to bring it down to earth.
What do you actually see?
“The team finishes the work, but stakeholders don’t buy in.” “I start projects with enthusiasm, then abandon them.” “Sales conversations make me uncomfortable, so I avoid them.”
Now collect several of these.
Then ask:
Are these separate problems, or are they all shadows cast by the same object?
That is the interesting part.
You look at one undesirable effect and ask:
If this is true, does it make that other thing more likely?
If yes, why?
And if the “why” is vague, keep going.
You are not trying to make a fancy diagram. You are trying to understand the machine.
2. Where is the hidden conflict?
Most persistent problems are not caused by stupidity.
They persist because two reasonable needs are pulling against each other.
Think of two people pulling on the same rope.
One side says:
“I must do X.”
The other side says:
“No, I must do the opposite of X.”
At first, it looks like one side must be wrong.
But usually both sides are protecting something real.
For example:
“I need to write from my expertise, otherwise I feel like a fraud.” “I need to write about broader topics, otherwise people won’t care.”
So the apparent conflict is:
“Stay narrow and credible” versus “Go broad and relevant.”
Now here is the trick.
You usually do not solve this by choosing one side.
You solve it by finding the assumption that made the two sides look incompatible in the first place.
Maybe the hidden assumption is:
“My expertise only counts if I stay inside my exact technical domain.”
Well, is that true?
Maybe not.
Maybe expertise can be used as a lens.
Maybe a data expert does not have to write only about data. Maybe they can use data thinking to explain business, decisions, incentives, organizations, careers, or human behavior.
Then the conflict starts to disappear.
Not because you compromised.
Because you found the weak beam holding up the whole false structure.
3. What should it change into?
Once you find the bad assumption, you need a new move.
Not a slogan.
Not a wish.
A move that changes the rules of the problem.
For example:
“Use data expertise as a lens for explaining broader human and business problems.”
Now you are not choosing between technical credibility and broad appeal.
You are using technical credibility to create broad appeal.
But you still have to check the idea.
Ask:
If we did this, what would improve first? What would that make possible? What else would change? Could this create a new problem?
This is important.
A beautiful idea is not enough. You have to see whether it actually produces the future you want.
You are testing the idea against reality.
4. How do we make the change happen?
Fine.
Now you have a beautiful theory.
But how do you actually move the wrench?
Ask:
What stops this future from existing already?
Each obstacle tells you something that must become true first.
Then get concrete.
What is true right now? Why does that require action? What exact action should happen next? What result should that action produce?
This is where many good insights die.
People understand the problem. They even see the better direction. Then nothing changes, because no one translates the insight into the next physical action.
So make it physical.
Not “be more consistent.”
Instead:
“Every Friday, choose one broad business problem and explain it through the lens of data.”
Not “get stakeholder buy-in.”
Instead:
“Before presenting the analysis, interview three stakeholders and ask what decision they are actually trying to make.”
The test is simple:
Could a camera see you doing it?
If not, it is not an action yet.
The whole thing in plain English
The process is this:
- List the bad things that keep happening.
- Make each one observable.
- Connect them by cause and effect.
- Find the deeper conflict between two legitimate needs.
- Expose the assumption that makes the conflict seem unavoidable.
- Break that assumption with a better move.
- Check whether the move really creates the future you want.
- Work backward into concrete actions.
Or, even simpler:
Don’t name the problem. Watch what it does. Don’t fix the smoke. Find out what keeps making smoke. Don’t compromise between two bad choices. Find the assumption that made you think those were the only choices. Don’t stop at insight. Turn the insight into the next thing someone can actually do.