Gartner Hype Cycle: The 5 Phases and How to Use It for Roadmap Timing

Gartner analyst Jackie Fenn first published the Hype Cycle in 1995 as a way to describe the gap she kept seeing between market enthusiasm for emerging technologies and their actual productive deployment. Thirty-one years and thousands of conference slides later, it is probably the most recognised and least used framework in technology strategy. Everyone can describe the five phases. Almost nobody uses it to make actual roadmap decisions — which is a waste, because the Hype Cycle is one of the best tools available for the single hardest timing question in product management:

When should we put an emerging technology on our roadmap — and when should we wait?

In the AI era, where new tools, models, and paradigms appear monthly, the timing question has never mattered more. This article treats the Hype Cycle as a practical roadmap and capital-allocation instrument, not as a descriptive summary of what’s happening in the market.

The Gartner Hype Cycle describes how emerging technologies progress through five phases: Innovation Trigger, Peak of Inflated Expectations, Trough of Disillusionment, Slope of Enlightenment, and Plateau of Productivity. First published by Gartner analyst Jackie Fenn in 1995, the cycle reflects how market expectations systematically overshoot before reality catches up, and is used to time adoption, investment, and roadmap decisions for emerging technologies.

Sign up (free) to tag emerging-tech bets by Hype Cycle phase

The Five Phases of the Gartner Hype Cycle

The Hype Cycle tracks a technology’s maturity through five distinct phases, plotted against expectations (y-axis) and time (x-axis):

1. Innovation Trigger — A breakthrough demonstration, academic paper, or prototype generates interest. No commercial products yet; usage is proof-of-concept.
2. Peak of Inflated Expectations — Early success stories combine with media attention to generate unrealistic expectations. Every vendor claims to have the new capability, often with the thinnest of implementations.
3. Trough of Disillusionment — Expectations collide with reality. Early deployments fail or underwhelm. Media narrative flips from breathless to sceptical. Vendors quietly pivot or go bust. Many technologies never exit the trough.
4. Slope of Enlightenment — Serious practitioners work out what the technology is actually good for. Second- and third-generation products emerge that solve the problems of the first wave. Best practices start to cohere.
5. Plateau of Productivity — The technology becomes mainstream, commoditised, and part of standard operating practice. Adoption accelerates; ROI becomes predictable.

Two things to notice about this curve:

• The Peak of Inflated Expectations is much higher than the Plateau of Productivity. This is deliberate — it reflects that expectations always overshoot before reality catches up. Amara’s Law captures the same dynamic: “we tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.” The Hype Cycle is Amara’s Law drawn as a curve.
• The time axis is elastic. Some technologies (e.g., cloud computing) take 10–15 years to reach the plateau. Others (e.g., blockchain beyond Bitcoin, VR/metaverse in 2024) are stuck in the trough indefinitely. 3D printing is the canonical cautionary tale — Peak in 2012, Trough through 2016–2020, now slowly emerging for specific industrial applications. AI-era technologies may compress this further still.

The Economist has noted (summarising Gartner’s own retrospectives) that only about 20% of emerging technologies traverse the full curve to the Plateau of Productivity — the majority stall in the Trough and never emerge, either because the technology never matures enough or because a substitute renders them obsolete. This is a brutal base rate that most “AI strategy” conversations ignore. If 4 in 5 hyped technologies don’t make it, your allocation discipline matters more than your enthusiasm.

How to Read the Hype Cycle for Your Roadmap

This is where most articles stop. They describe the phases and tell you the technology is on one or other of them. The useful question is: given what phase a technology is on, should we invest in it, and how?

The counter-intuitive conclusion: the Trough of Disillusionment is when you should be placing your dedicated-team bets, not the Peak of Inflated Expectations. Everyone is disillusioned; prices are cheap; second-wave vendors are quietly building the products that will plateau in 3–5 years. This is exactly the timing window that well-governed PE portfolio companies capitalise on and that less-disciplined companies miss.

The Two Governance Failures Around the Hype Cycle

There are two predictable failure modes, and I see both constantly:

Failure 1: The Peak-of-Expectations Big Bet

The CEO reads a Forbes article. The board wants to “be seen to be doing AI.” Suddenly the roadmap has a major strategic initiative with significant capacity committed — often with the best engineers pulled off their current products. Six months later the initiative has produced demos but no revenue, the core product has slipped because of the capacity drain, and the team quietly reassigns the engineers back. The “big AI bet” is rebadged as “AI-enabled features” and disappears into the backlog.

The root cause is almost never that AI is overhyped (it isn’t). It’s that the company committed capacity without doing a proper business case with targets, and allocated to an initiative at the Peak of Inflated Expectations when the technology wasn’t mature enough to deliver on the ambition. The governance failure is: the decision was driven by urgency theatre, not by strategy.

Failure 2: The Trough-of-Disillusionment Refusal

The inverse problem. A technology had a hype cycle, didn’t deliver on peak expectations, and is now being quietly dismissed. “Crypto is dead.” “Voice is overhyped.” “AR isn’t ready.” Meanwhile, second-wave vendors are building the actually-useful versions and the companies that will plateau with them in 5 years are quietly placing dedicated-team bets.

The company that refused to bet at the trough will now have to buy the plateau-ready product from a vendor — expensively. The governance failure is: short-term post-hype cynicism overruled long-term strategic judgement.

The Portfolio View: Different Products, Different Phases

As with the product life cycle , the Hype Cycle lens works best at the portfolio level. Your company’s strategy should probably include emerging-tech bets at multiple phases of the cycle simultaneously:

• A few discovery pods evaluating Innovation Trigger technologies that might matter in 5+ years (low commitment, high optionality)
• Dedicated minimum-viable teams on Trough-of-Disillusionment technologies where you have a strategic thesis (contrarian bets)
• Scaled squads integrating Slope-of-Enlightenment technologies into production products (where most of the real capacity should go)
• Operational teams adopting Plateau technologies as commoditised components (table-stakes, low margin)

Almost no company manages this deliberately. Most have accidental exposure — some projects at the Peak (that everyone’s excited about), some at the Plateau (that everyone assumes are done), and nothing in the contrarian Trough window where the real returns are.

The 2026 Hype Cycle: Where AI Actually Sits

AI isn’t a single technology on a single Hype Cycle position. It’s a suite of capabilities each at different phases:

• Foundation LLMs — Slope of Enlightenment, approaching Plateau. Commoditising rapidly.
• AI agents / orchestration — Late Peak or early Trough. Enormous excitement, implementation problems becoming visible.
• AI-first SaaS — Peak of Inflated Expectations. Every vendor claims AI-native; most are AI-wrappers.
• AI-generated code — Slope of Enlightenment in developer productivity; Peak in the broader “no-code” adjacency.
• AI in regulated industries (healthcare, financial services) — still at Innovation Trigger for meaningful deployment.

This matters because “we need an AI strategy” is not a coherent statement. Which AI? Where on the cycle? What commitment? The board conversation should be specific, not thematic.

Try RoadmapOne to structure emerging-tech bets as dedicated teams — free sign-up

Using the Hype Cycle as a Roadmap Governance Tool

Three practical disciplines make the Hype Cycle operationally useful:

1. Tag emerging-tech objectives explicitly

On your roadmap, tag each emerging-tech bet with its estimated Hype Cycle phase. “AI agents for customer service” tagged “Peak” is a different decision from “AI agents for customer service” tagged “Trough”. Forcing the explicit tag focuses the board conversation.

2. Allocate capacity by phase, not by hype

Most emerging-tech roadmap capacity should go to Slope-of-Enlightenment technologies where the returns are starting to be real but the opportunity is still open. A small discovery allocation for Innovation Trigger; a dedicated-team allocation for contrarian Trough bets; almost nothing for Peak technologies where everyone else is already spending. The Three Horizons lens is a useful sibling to this — H3 maps roughly to Innovation Trigger / Trough bets; H2 maps to Slope of Enlightenment; H1 lives at the Plateau.

3. Revisit quarterly, not annually

Hype Cycle positions move fast in 2026 — sometimes a category moves from Peak to Trough in two quarters. Reviewing your emerging-tech tags once a year means the tags are stale and your decisions are based on stale judgement. Quarterly tagging and review is the right cadence.

How the Hype Cycle Relates to Other Lifecycle Models

Don’t confuse the Hype Cycle with the product life cycle or with Rogers’ Diffusion of Innovations . They are complementary:

• The Hype Cycle plots market expectations for an emerging technology over time. It tells you about the sentiment and maturity of a category.
• The Product Life Cycle plots sales of a specific product over time. It tells you where your commercial product sits on its own revenue journey.
• Diffusion of Innovations plots adoption of an innovation through a population. It tells you which customer segment you are selling to at any moment.

A single AI product can simultaneously be at the Peak of Inflated Expectations (Hype Cycle), in the Introduction stage (Product Life Cycle), and sold to early adopters (Diffusion). All three lenses are useful; each answers a different question.

The PE / Board Diagnostic for Hype Cycle Exposure

When I look at a portfolio company’s roadmap from a governance position, I ask:

1. What emerging-tech bets are on our roadmap? Named explicitly, not just “innovation.”
2. What phase of the Hype Cycle is each of them on? If nobody on the leadership team can answer, the bet isn’t being governed.
3. Are we over-concentrated at the Peak? Big capacity commitments to Peak technologies is a yellow flag — is this strategic conviction or FOMO?
4. Are we under-exposed at the Trough? A portfolio with no contrarian bets is a portfolio that will miss the next wave of plateau-era winners.
5. Are we doing the grown-up conversation about the technologies we’ve chosen NOT to bet on? Silence on a whole category means nobody has actively decided; it’s drift rather than strategy.

How RoadmapOne Helps

RoadmapOne lets you tag objectives by Hype Cycle phase alongside Run/Grow/Transform and Three Horizons. The analytics surface — in one view — how much capacity is committed to each phase of each emerging technology. Most boards find this clarifying because the conversation shifts from “do we have an AI strategy?” to “do we have the right capacity allocation across the Hype Cycle?” The second question is one you can actually answer.

Frequently Asked Questions

What are the five stages of the Gartner Hype Cycle?

Innovation Trigger, Peak of Inflated Expectations, Trough of Disillusionment, Slope of Enlightenment, and Plateau of Productivity. Each phase represents a different combination of market expectations and actual technological maturity.

How long does each phase of the Hype Cycle last?

It varies enormously by technology. Some categories move through all five phases in 3–5 years; others spend a decade in the Trough. In the AI era, compression is likely — some AI sub-categories may complete the cycle in 24–36 months.

What is the Trough of Disillusionment?

The phase after the Peak of Inflated Expectations, where expectations fall below the eventual steady state because early deployments have underwhelmed. Counter-intuitively, this is often the best time to make dedicated-team bets on technologies you believe will reach the Plateau — capital is cheap, competition has thinned, and second-wave vendors are quietly building the winning products.

Should I invest in a technology at the Peak of Inflated Expectations?

Generally no. Peak-of-Expectations investments are overpriced, over-hyped, and often based on immature capabilities. The exception is if the technology is genuinely core to your strategic thesis and you have the discipline to survive an 18-month period when expectations collide with reality. Most companies don’t.

Does every technology go through the Hype Cycle?

Not all. Some technologies never reach the Innovation Trigger in any meaningful way — they remain academic curiosities. Others ascend the Peak and then never reach the Plateau — they stall permanently in the Trough. Gartner estimates only about 20% of hyped technologies reach the Plateau of Productivity. The pattern is a useful default, not a deterministic path.

How is the Hype Cycle different from the product life cycle?

The Hype Cycle tracks market expectations for a technology category over time. The product life cycle tracks sales of a specific product over time. A product built on a Hype Cycle technology can be in the Introduction stage of its own life cycle while its underlying technology is on the Plateau.

Sign up (free) to tag and govern your emerging-tech bets

Conclusion

The Hype Cycle is not a predictive tool — nobody can say with certainty where a technology will land. It is a governance tool. It forces the leadership team to be explicit about which emerging technologies they are betting on, what phase those technologies are on, and what scale of commitment matches the phase.

The best portfolio companies I work with use it quarterly as a simple tagging discipline on their roadmap. The worst use it only as a PowerPoint artefact at the annual offsite. The difference between the two — in long-term capital efficiency — is enormous.