One of my favorite problem-solving questions is: “What’s wrong with this idea?” It drives the conversation away from why the idea might work, and toward why it might fail. That perspective is invaluable when building.

Question: “What’s Wrong With This Idea?”

I learned this idea from Paul Arden:

“If, instead of seeking approval, you ask, ‘What’s wrong with it? How can I make it better?’, you are more likely to get a truthful, critical answer."― Paul Arden, It’s Not How Good You Are, It’s How Good You Want To Be

People don’t like to criticize. They’d rather agree. Add in power dynamics such as boss-subordinate or junior-senior. Engineering leaders have both working against them! On top of that, criticizing an idea requires emotional safety, which is sometimes in short supply.

Less-experienced folks often won’t even think to challenge the idea. They might have good reasons to doubt based on their life experience or reasoning skills. But the thought to challenge it won’t enter their heads unless invited.

Asking this question invites that challenge. It’s signaling: “I care more about getting the right answer than being right.”

That challenge is incredibly valuable. We want to hear it now from our team. Not from a customer or competitor when the product is failing in production.

In case it isn’t clear, ask this about your ideas, not others. You’re modeling vulnerability.

Here are three mental models that have helped me think about this question.

Mental Model: Crossing a One-Way Street

They say that a programmer is a person who looks both ways when crossing a one-way street. Asking question is a way of “looking both ways.”

People get hit by cars going the wrong way down a one-way street. Do this job long enough, and you’ll see it happen. We want to try and see as far ahead as we can, now.

Mental Model: “How am I going to lose this case?”

I watched a legal procedural where the lead prosecutor asked, “How am I going to lose this case?” This is another version of this question.

Her team of lawyers listed a bunch of problems: the defendant was too likable, the victim had a checkered past, and there were no witnesses. And then, the team started solving these problems.

Contrast that with a prosecutor who says: “Open-and-shut case.” Which team do you want to be on?

Mental Model: Inverted Thinking

Another way to think about this is through the lens of inverted thinking. With inverted thinking, we can think through the full implementation of our idea and ask, “We built this and it failed. What happened?”

Some possible answers:

  • We missed a use case
  • We didn’t know what the customer wanted (no acceptance criteria, or bad acceptance criteria)
  • We didn’t deliver what the customer wanted (failed to meet the acceptance criteria)
  • It took too long to deliver (the competition beat us, or we ran out of funding)
  • The stakeholder who cared about it went away (customer churn)

Asking “What’s wrong with this idea?” can surface these hazards early on.

Using the Question With AI

I’ve found this question mitigates one of the main flaws of current AI: a desire to please, and an assumption that I’m smart.

As AI researchers have written, one of the core instructions to an LLM is, roughly, “If the human thinks you’ve answered the question, that’s good.” And also, “Assume the human knows more than you.” This introduces an obvious problem: sometimes my ideas are bad, and AI is trained to not tell me that.

Instead, ask AI “What’s wrong with this implementation?” You might get a very valuable answer.

Call to Action

The next time your team is considering an idea, break through the positivity with the question: “What’s wrong with this idea?”