Growth Can Hide Fragility

Growth is the most convincing costume fragility ever wears. Revenue rising, team expanding, pipeline full — every one of those signals is real, and none of them tells you whether the structure underneath can absorb a change in conditions.

I know this the expensive way. For roughly nine years I built and ran a renewable-energy business. [VERIFIED FACT REQUIRED — timeline and scale details pending verification.] In its best-looking years, the business was also accumulating the exact exposures that would later break it: dependence on policy conditions I didn’t control, financing whose assumptions were never re-tested, and a growth rate that consumed the margin for error faster than it created new margin.

Momentum answers the wrong question

The question growth answers is: is the current strategy working under current conditions?

The question that decides survival is different: what happens to this structure when the conditions change?

Almost every review process I ran — and most that I now see inside other businesses — measured the first question with precision and left the second to instinct. Not because owners are careless, but because the first question produces numbers on a schedule and the second one produces an argument nobody wants to have during a good quarter.

Why this matters more with AI in the room

AI adoption is a conditions change. It moves cost structures, shifts what customers tolerate, and revalues skills inside the team — all beneath a revenue line that can keep looking normal for a surprisingly long time.

A business that has never practiced asking the second question will experience AI the way my business experienced its own conditions change: as a surprise that was, in retrospect, fully documented.

The failure wasn’t hidden. It was filed under success.

What I would examine now

If I could re-run my own case, I would look at three structural readings, none of which appear on a standard dashboard: how concentrated the sources of survival are, how many assumptions the financing needs to stay true simultaneously, and how long the business could operate while being wrong about its main bet. I am building those readings into the frameworks I test here.