Haven’t seen you in a while

Companies survive through multiple S-curves — the technology cycles that first accelerate, then plateau, then give way to the next wave. Intel mastered this for decades, riding from microprocessors to personal computers to internet infrastructure. Then they missed mobile. That stumble revealed something essential: the companies that endure aren't just good at their current curve. They're good at recognizing when it's ending.

Most founders optimize for growth within their existing trajectory. Reasonable, until it isn't. The trap lies in mistaking operational excellence for strategic resilience. You can execute flawlessly on a declining curve and still find yourself irrelevant.

Intel's mobile blindness wasn't a failure of engineering. Their x86 architecture was objectively superior to ARM in processing power and compatibility. But mobile demanded a different optimization function — battery life over raw performance, cost efficiency over backward compatibility. Intel kept solving the wrong problem with increasing sophistication.

Steve Jobs understood this when he killed the iPod at its peak to build the iPhone. Cannibalization wasn't a risk to be managed; it was a capability to be weaponized. The companies that navigate S-curve transitions successfully don't just tolerate internal competition — they institutionalize it.

Netflix provides the clearest template. Reed Hastings didn't gradually evolve from DVDs to streaming. He built streaming as a parallel business while DVDs still generated 90% of revenue. Two different teams, different metrics, different success definitions. When the curves crossed, Netflix was positioned on both sides of the transition.

The early signals of S-curve exhaustion are almost always disguised as temporary headwinds. Customer acquisition costs that "seasonally" rise and never fully retreat. Feature improvements that generate less user engagement than they used to. Competitive advantages that require increasing investment to maintain the same market position.

These patterns compound slowly, then suddenly. By the time they're obvious in financial statements, the next curve is already established — likely by someone else.

Smart operators develop leading indicators specific to their industry's physics. In enterprise software, it might be the ratio of new features to customer requests — when you're building more features but solving fewer real problems, the curve is maturing. In consumer products, watch for the point where user experience improvements stop translating to usage increases.

The organizational challenge is structural. The team optimizing your current curve needs different incentives than the team building your next one. Current-curve teams should maximize efficiency and defend market position. Next-curve teams should maximize learning and optionality creation.

This requires parallel resource allocation that feels wasteful in the short term. You're funding two different futures, knowing one will eventually obsolete the other. The alternative — sequential development — creates timing vulnerabilities that competitors exploit.

Amazon's approach is instructive. They didn't pivot from e-commerce to cloud computing. They developed AWS while e-commerce was accelerating, treating it as infrastructure for their core business. When AWS revealed itself as a larger opportunity, they already had years of development and customer feedback cycles completed.

Start before you have to. Curve transitions take longer than anyone estimates. If you wait until your current trajectory shows clear deceleration, you're building from behind.

Fund optionality, not certainty. Next-curve investments should be portfolio bets, not single big bets. Multiple small experiments reveal which direction has real traction.

Measure different metrics. Current-curve success metrics will systematically undervalue next-curve progress. Growth rates, user engagement, and profitability all lag learning rates and technical capability development.

Protect the transition team. Early-stage next-curve work will consistently underperform current-curve work on traditional metrics. Insulating the new work from existing performance management systems isn't just politically necessary — it's strategically essential.

The companies that compound across decades don't just ride S-curves. They orchestrate them. They turn technological transition from an external force to be survived into an internal capability to be deployed. That's how you build institutional durability in an economy where the average S&P 500 tenure is fifteen years and declining.