Inversion

What it means

Inversion is the practice of approaching a problem backwards. Instead of asking how to achieve something, you ask how to guarantee its opposite - and then work to avoid those things. It’s a deceptively simple technique that consistently produces insights the forward-looking approach misses.

The idea is often attributed to the mathematician Carl Jacobi, whose advice was “invert, always invert” (man muss immer umkehren). Charlie Munger, Warren Buffett’s business partner, made it a cornerstone of his thinking: “All I want to know is where I’m going to die, so I’ll never go there.”

The reason inversion works is that humans are generally better at identifying what’s wrong than what’s right. We can spot problems more easily than we can design solutions. If you ask a team “how do we build a great product?”, you’ll get vague, aspirational answers. If you ask “how do we guarantee this product is terrible?”, you’ll get a vivid, specific, and immediately actionable list - poor communication, ignoring user feedback, no testing, unclear ownership. Avoid those things and you’ve already eliminated the most common causes of failure.

In the real world

In investing, Munger doesn’t start by asking “what’s the best investment?” He starts by asking “what would definitely lose money?” and avoids those things. This negative screen eliminates the worst options quickly, and what remains is a much better set to choose from. It’s not glamorous, but it works - avoiding stupidity is more reliable than chasing brilliance.

In healthcare, the question “how do we keep patients healthy?” is enormous and diffuse. Inverted - “what definitely makes patients sick?” - it becomes actionable: smoking, inactivity, poor diet, lack of sleep, chronic stress, social isolation. Public health campaigns that focus on eliminating known harms often achieve more than programmes chasing optimal wellness.

In daily life, inversion is the friend who asks “what’s the worst that could happen?” before you make a decision. Want a good relationship? Instead of listing ideal partner qualities, list the behaviours that would definitely destroy a relationship - dishonesty, contempt, avoidance, keeping score. Avoid those and you’ve cleared the ground for something good to grow.