S-Curve Lifecycle Analysis

The Investor Who Arrives on the Plateau

Space-sector capital has a timing problem, and it rarely looks like the problem it is. A fund commits to a launch technology at the moment every industry deck still labels it “emerging,” writes the check on the expectation that cost-per-kilogram will keep falling at the rate of the last five years, and then watches the improvement curve flatten just as the rest of the portfolio is being built around the extrapolated price point. The fund was not wrong about the technology. It was late to a phase transition that had already quietly happened.

The mirror image is equally common. A national agency waits too long to commit to a successor architecture because the incumbent’s marginal gains still look impressive on year-to-year slides, and by the time the displacement becomes visible in a program review, the industrial base on the other side of the curve is already consolidating without them. Neither error is a mistake of ambition. Both are failures of temporal positioning — of reading where a technology actually sits in its own maturation, rather than where the last budget cycle said it was.

S-curve lifecycle analysis exists for exactly this calibration problem. It gives the practitioner a structured way to ask not “is this technology mature?” but “where on its maturation trajectory is it, and what does that position tell me about the next strategic move?”

Rogers, Foster, and the Shape That Would Not Go Away

The S-curve entered strategy from two adjacent traditions. Everett Rogers, in Diffusion of Innovations (1962), made the sigmoid shape canonical for the spread of new ideas and products through social systems: slow adoption by a small pioneer group, acceleration as early and late majorities follow, and a plateau as the remaining population of laggards proves stubborn. Rogers’ curve was descriptive, built out of agricultural extension studies, medical prescribing patterns, and the spread of household technologies, but its reach was broader than its data: the shape recurred so consistently that strategists began treating it as a structural regularity rather than a stylized fact.

Richard Foster, writing from within McKinsey’s innovation practice, redirected the same shape toward corporate technology strategy in Innovation: The Attacker’s Advantage (1986). Foster’s innovation was not the curve itself but the pairing: two S-curves, overlapping in time, with an incumbent technology approaching its ceiling just as a successor begins its climb. The crossover between them — the moment at which the successor’s worse-but-improving performance passes the incumbent’s better-but-stalled performance — became Foster’s central diagnostic tool. His argument was blunt: incumbents do not fail because they manage their current curve badly. They fail because they manage it so well that they keep investing in it past the point at which the successor curve has already established the new frontier.

The space sector inherited the framing more or less intact. Chemical-propulsion roadmaps, broadband-satellite throughput charts, solar-cell efficiency tables — almost every technology narrative in the sector is implicitly an S-curve story, whether or not the practitioners drawing it have read Foster.

What the Curve Sees That a Maturity Rating Does Not

The analytical move at the heart of S-curve analysis is the substitution of a trajectory for a snapshot. A technology readiness level is a verdict at a moment — TRL 7, TRL 9, flight-proven. It tells the reader where the technology is. It does not tell the reader where it is going, how fast, or how much room remains before it stops.

S-curve analysis answers those second-order questions by insisting on three disciplines.

What the three disciplines produce together is something a static maturity score cannot: an estimate of how much room the technology has left before the next phase transition, and therefore a timing signal for strategic commitment. It is the difference between knowing a technology is mature and knowing that one is about to stop improving.

Two Launch Curves in the Same Field of View

Consider the analytical move applied to a generic comparison between two launch architectures operating in the same market window. Technology A is an expendable heavy-lift system with a long flight history. Its cost-per-kilogram-to-LEO trajectory is instructive: a long, slow decline through the 1980s and 1990s, a modest step improvement during the commercial launch boom, and then a clear plateau. For several consecutive years, marginal improvement has been incremental. The ceiling is structural: expendable staging imposes a mass-fraction penalty that no amount of vehicle-level optimization can remove. Technology A is late in its maturity phase.

Technology B is a fully reusable medium-lift system. Its cost-per-kilogram curve looked unimpressive for the first decade of its development — it sat in emergence, with demonstration flights that were expensive per mission and cheap only in ambition. Then reuse rates began climbing. As individual boosters accumulated flights without refurbishment beyond expected envelopes, amortization of hardware costs dropped per-launch pricing faster than most incumbents forecasted. The slope is steep. The ceiling has not been approached. Technology B is in early rapid growth.

Now overlay the two curves on the same metric axis. The crossover on cost per kilogram has already passed: B is below A, and the gap is widening. But extend the analysis to a second metric — payload capacity for ultra-heavy dedicated missions — and the picture inverts. A retains a niche that B, at its current reuse-optimized configuration, cannot yet serve. The lifecycle position of each technology depends on which performance metric defines the curve.

The signposts for the next phase transition fall out of the analysis. For Technology A, watch for customer migration on volume missions (already happening) and, eventually, for the decision by a historical operator to retire the architecture rather than refurbish it. For Technology B, watch the reuse-rate asymptote: the moment per-vehicle flight counts stop climbing linearly, cost-per-kilogram improvement will taper and a new curve — perhaps a larger, more aggressive reuse architecture, perhaps something structurally different — will begin its own emergence.

The non-obvious insight, the one a static maturity rating could not produce, is that the lifecycle position of each technology depends on the metric chosen, and that a portfolio strategy built on a single metric will misread the market. An operator who optimized only for cost-per-kilogram would exit Technology A prematurely and miss its surviving ultra-heavy niche; an operator who optimized only for payload would miss the volume migration entirely.

Where It Earns Its Keep and Where It Misleads

The method’s strength is timing. No other tool in the strategic library gives the practitioner as clean a signal on when a technology is about to stop improving, or on when a successor is about to cross over. For capital-allocation decisions with long commitment horizons — fleet purchases, launch-pad investment, component-supply contracts — lifecycle positioning is often worth more than absolute performance assessment.

It is also the natural complement to disruption theory. Disruption tells the strategist what kind of successor to expect and through which market segment it will enter. S-curve analysis tells the strategist when to expect it, by reading the incumbent’s slope against the successor’s. The two methods in combination are stronger than either alone.

The weaknesses are real. S-curves are recognized retrospectively far more easily than prospectively, and calling the current phase in real time is genuinely hard — plateau and pause can look identical for several years. Metric choice smuggles judgments; the same technology can appear in different phases depending on what is measured. The method assumes a single dominant trajectory and handles poorly technologies whose performance dimensions are evolving at different rates. It oversimplifies when multiple partial substitutes are emerging rather than one successor. And it can encourage deterministic thinking — the curve suggests inevitability, when in fact external shocks, policy reversals, and adjacent technological convergences can reshape the trajectory.

The complement for that last weakness is scenario planning: pairing the curve with a handful of alternative futures forces the practitioner to identify what would redirect the trajectory, rather than assuming the trajectory is fate. For policy and governance topics, where performance is not quantifiable in the same way, the S-curve is largely inapplicable; comparative policy analysis and institutional design analysis carry the load there.

For the Practitioner