The most expensive mistake in product management isn't picking the wrong feature. It's running the right playbook at the wrong time, pouring marketing spend into a product the market hasn't validated yet, or defending a mature cash machine as if it still had a growth story to tell. The product lifecycle is the map that tells you which game you're playing, so you stop optimising for the last stage and start managing the one you're in.

The quick version

  • Four stages, four jobs. Launch is about proof, growth is about scale, maturity is about defending margin, exit is about disciplined withdrawal. The right move depends entirely on which one you're in.
  • The map is decades old and still useful. Theodore Levitt set out the stages in Harvard Business Review in 1965; Geoffrey Moore added the deadly gap between launch and growth in 1991.
  • Stage tells you where to put your money. The BCG growth-share matrix turns "what stage is this in?" into "fund it, milk it, or kill it?"
  • It's a lens, not a law. Products skip stages, restart, or die young. Use the curve to ask better questions, not to predict the future.

The idea in depth

The four-stage curve is one of the oldest ideas in management that's still worth knowing. In "Exploit the Product Life Cycle" (Harvard Business Review, November 1965), Theodore Levitt argued that most successful products pass through recognisable stages, introduction, growth, maturity and decline, and that awareness of which stage you're in should change decisions on pricing, identity, and how you sell and distribute. His core point wasn't that the curve is destiny. It was that smart firms anticipate the next stage instead of being surprised by it: plan the growth push before launch demand even arrives, and plan the maturity defence before growth flattens.

That reframes the planning conversation. Instead of treating your product as a fixed thing, you ask, every cycle, "which stage are we in, and what does this stage actually reward?" A product in launch rewards learning speed. A product in growth rewards capacity and distribution. A product in maturity rewards efficiency and retention. When teams keep running the launch playbook, endless experimentation, founder-led selling, "we'll figure out the model later", long after the market has spoken, they burn the runway that scaling would have paid back.

flowchart LR
  A(["Launch / Introduction
slow sales, prove demand"]) --> B(["Growth
sales accelerate, scale up"])
  B --> C(["Maturity
sales plateau, defend margin"])
  C --> D(["Decline / Exit
sales fall, withdraw"])
  C -. "reinvest / reposition" .-> B
The classic four-stage curve, with the reinvestment loop that can postpone decline. Leaders Loop

Launch is a gap, not a ramp

The biggest weakness in Levitt's tidy curve is that it draws growth as a smooth ramp out of launch. In reality there's a cliff in the middle. Geoffrey Moore's Crossing the Chasm (1991) maps a product's early life onto a technology adoption lifecycle, innovators, early adopters, early majority, late majority, laggards, and points out that there's a chasm between the early adopters and the early majority. The two groups buy for opposite reasons. Early adopters want to be first and will tolerate a rough product; the early majority is pragmatic and waits until the thing is proven, supported, and works with what they already own. The signals that tell you launch is "working", enthusiastic visionaries, glowing pilots, are exactly the signals that don't carry you across to a real market.

During launch, then, treat your first wins as evidence of a niche, not of escape velocity. Pick one beachhead segment, make the product genuinely complete for them (support, integrations, references, the boring parts), and resist the temptation to read early-adopter love as product-market fit. That discipline is the bridge between launch and growth; without it, the ramp Levitt drew never appears.

Maturity and exit: where the money is actually made, and lost

Growth gets the attention, but maturity is where most of a portfolio's cash gets generated and most of its capital gets wasted. The cleanest tool for this is the growth-share matrix, which Bruce Henderson and the Boston Consulting Group developed in the late 1960s and set out in BCG's "The Product Portfolio" around 1970. It plots products on two axes, market growth rate and relative market share, into four boxes: question marks (low share, fast market, your launches and bets), stars (high share, fast market, your growth winners), cash cows (high share, slow market, your mature earners), and dogs (low share, slow market, your exit candidates). The intended flow is simple: milk the cash cows to fund the question marks; back the winners until they become stars; let stars settle into cash cows as their market matures; and divest the dogs.

quadrantChart
  title Growth-share matrix
  x-axis "Low market share" --> "High market share"
  y-axis "Slow market growth" --> "Fast market growth"
  quadrant-1 "Stars (back the winners)"
  quadrant-2 "Question marks (your bets)"
  quadrant-3 "Dogs (exit)"
  quadrant-4 "Cash cows (milk to fund)"
The BCG growth-share matrix: fund question marks, back stars, milk cash cows, divest dogs. Leaders Loop

Here the discipline is to manage maturity and exit as deliberate cash decisions, not emotional ones. A cash cow doesn't need a growth roadmap; it needs efficiency, retention, and a light touch on cost. A dog doesn't need one more turnaround attempt; it needs an honest exit plan, sunset the line, harvest the customers, or sell the asset, so the cash and the team move to something with a future. The hardest part of the lifecycle isn't launching; it's letting go on time, because the product that pays the bills today is the one nobody wants to kill.

An honest limitation. The curve has real critics, and they've been loud since the start. In "Forget the Product Life Cycle Concept!" (Harvard Business Review, January–February 1976), Nariman Dhalla and Sonia Yuspeh argued the model's empirical basis is shaky and that, used carelessly, it becomes a self-fulfilling prophecy: a manager reads "maturity," cuts investment and marketing, and causes the decline the curve "predicted." Products also skip stages, plateau for decades (think Coca-Cola), or get a second life through repositioning. Treat the lifecycle as a diagnostic lens that prompts good questions, not as a clock you're obliged to obey. The stage is a hypothesis about your market, and like any hypothesis it can be wrong.

A worked example

Consider "Rota," an illustrative shift-scheduling app for hospitality venues. (The numbers below are made up to show the mechanics, not real figures.)

Launch. Rota lands with a handful of trendy café owners, classic early adopters, who love tinkering with a new tool and forgive the rough edges. Sign-ups look healthy. The trap would be to read this as growth and hire a sales team. Instead the team treats it as a beachhead: they pick "independent cafés with 5–20 staff," and spend a quarter making the product genuinely complete for that one segment, payroll export, a phone app the floor staff will actually open, and three reference customers. Unglamorous, and it's what crosses the chasm.

Growth. Now the pragmatic early majority starts buying, because the product is proven and supported. Sales roughly triple year on year. The job changes overnight: from learning to scaling. Rota invests in onboarding, server capacity, and a repeatable sales motion. In BCG terms it's a star, winning, but cash-hungry, and the right call is to keep feeding it rather than milk it early.

Maturity. Three years on, most of Rota's addressable cafés are signed. Growth flattens to single digits; the work is now retention and margin. Rota becomes a cash cow: it funds the company's next bets rather than demanding endless feature spend. The mistake here would be a panicky "growth at all costs" reinvestment that destroys the margin the business actually runs on.

Exit. One of Rota's side modules, a standalone tip-pooling tool, never found a market and sits as a dog: low share, no growth, a steady drain on engineering. The disciplined move is to sunset it, migrate the few users into the core product, and redeploy the engineers to a new question mark. Killing it isn't failure; it's the matrix working as intended.

Frequently asked questions

How do I know which stage my product is actually in?

Look at the shape of your growth, not its level. Accelerating sales and an expanding total market is growth; a flat or replacement-only market with sales plateauing is maturity; a sustained downward drift is decline. The trickiest call is launch-vs-growth, where Moore's test helps: are you still selling to visionaries who love being first, or to pragmatists who bought because it's proven? Crossing that line is the real start of growth.

Can a product go backwards or restart the cycle?

Yes, that's the part the smooth curve hides. Repositioning, a new segment, a major redesign, or a price-model change can lift a mature product back into growth (Levitt called these "market stretching" moves in his 1965 piece). The lifecycle isn't a one-way escalator; the reinvestment loop is a legitimate strategy, not cheating.

Isn't the model too old to be useful?

The four stages are old; the underlying logic isn't dated. What's changed is speed, software lifecycles can compress launch-to-maturity into a couple of years instead of decades, and the fact that products are now updated continuously rather than replaced. The stages still tell you which job to optimise for; they just turn over faster.

What's the most common mistake leaders make with it?

Two, and they're mirror images. One is reading early-adopter enthusiasm as product-market fit and scaling into the chasm. The other is refusing to accept maturity or decline, defending a cash cow with growth spend, or pouring money into a dog out of attachment. Both come from running the wrong stage's playbook.

How is this different from product strategy?

Strategy is the choice of where to play and how to win; the lifecycle is the timing layer underneath it. The same strategic bet needs different execution at launch versus maturity. See Product strategy & vision for the choice itself, and use the lifecycle to decide how to run that choice over time.

Related in the Toolkit

Where to go next