In 1995, Jackie Fenn — an analyst at Gartner — published a research note that introduced a visual framework for tracking how new technologies evolve in public perception. She called it the Hype Cycle. The shape was distinctive: a sharp rise, a dramatic peak, a steep crash, a slow recovery, and a final plateau. The curve did not describe the technology's actual capability. It described the gap between what people believed the technology could do and what it actually could do at any given moment. That gap — the delta between expectation and reality — follows the same pattern for virtually every transformative technology in modern history. And the pattern has not changed in thirty years.
Five phases. The Innovation Trigger is the moment a technology becomes publicly visible — a breakthrough paper, a product demo, a provocative headline. The technology is real but immature, and the earliest adopters begin experimenting. Then comes the Peak of Inflated Expectations, where media coverage, venture capital, and public enthusiasm push expectations far beyond what the technology can currently deliver. Everyone is talking. Few understand the constraints. The third phase is the Trough of Disillusionment — the crash that follows when reality fails to match the hype. Early products underperform. Startups fail. Media coverage turns negative. Investment dries up. The technology is now unfashionable. The fourth phase, the Slope of Enlightenment, is the quiet period where serious builders do serious work. Products improve. Use cases clarify. The gap between expectation and capability narrows. The final phase is the Plateau of Productivity — where the technology delivers genuine, sustainable value to a broad market. Not the utopian vision from the Peak. Not the despair from the Trough. Just real, compounding utility.
The pattern repeats with almost mechanical regularity. The internet followed it: peak hype in 1999, trough in 2001–2003, plateau by 2005–2010 as Google, Amazon, and Facebook built businesses on the infrastructure the bust had left behind. Blockchain followed it: peak in late 2017 as Bitcoin touched $20,000 and ICOs raised $5.6 billion, trough through 2018–2019 as 90% of tokens went to zero, with stablecoins and enterprise blockchain emerging on the Slope of Enlightenment by 2020. Virtual reality followed it: peak in 2016 when Facebook acquired Oculus and VR was declared "the next computing platform," trough in 2017–2019 as consumer headsets gathered dust, slope emerging in the early 2020s through enterprise training and surgical simulation. And AI — specifically generative AI — entered its Peak of Inflated Expectations in 2023 when ChatGPT reached 100 million users in two months and every company on earth declared itself "AI-first." Whether the Trough has arrived by the time you read this sentence depends on when you are reading it.
The framework's sharpest insight is not the shape of the curve. It is the investment implication. The Trough of Disillusionment is where the best opportunities emerge — precisely because most investors, most media, and most corporate buyers have moved on. Amazon was trading at $6 in 2001. Bitcoin was at $3,200 in December 2018. The technology had not changed. The expectations had collapsed. The founders who kept building through the Trough, and the investors who kept funding them, captured the value that the Peak's tourists had abandoned.
Section 2
How to See It
The Hype Cycle is operating whenever public sentiment about a technology diverges dramatically from its actual capability — in either direction. At the Peak, expectations exceed reality. In the Trough, reality exceeds expectations. Both gaps create mispricings that disciplined actors can exploit.
You're seeing the Hype Curve when the conversation about a technology shifts from "what can it do?" to "what can't it do?" — and then, after the crash, shifts back to "what can it actually do, specifically, for this use case?"
Investing
You're seeing the Hype Curve when a technology sector's funding collapses after a period of frenzied investment. Blockchain startup funding dropped 73% from 2018 to 2019. VR startup funding fell 60% between 2017 and 2019. The capital retreat is the Trough's financial signature — and the companies that survive it with product and revenue tend to generate the best venture returns of the cycle.
Startups
You're seeing the Hype Curve when a founder in a hot category finds it easy to raise money but hard to find paying customers. During AI's 2023 Peak, seed rounds for AI startups closed in days while enterprise procurement cycles remained six to twelve months. The gap between investor enthusiasm and customer adoption is the distance between the Peak and reality.
Corporate Strategy
You're seeing the Hype Curve when a Fortune 500 company launches a high-profile "innovation initiative" around a trending technology — blockchain lab, VR division, metaverse strategy — and quietly shuts it down eighteen months later. The launch happens at the Peak. The shutdown happens in the Trough. The genuinely useful applications emerge years later, usually without a press release.
Personal Decisions
You're seeing the Hype Curve when career decisions cluster around trending technologies. In 2017, every product manager wanted to work on blockchain. In 2023, every engineer wanted to work on AI. The career FOMO at the Peak produces an oversupply of talent chasing a technology whose practical applications have not yet matured — and an undersupply of talent in the Trough, precisely when the most important building work begins.
Section 3
How to Use It
The framework's primary application is timing — understanding where a technology sits on the curve to calibrate investment decisions, hiring plans, product strategy, and expectations.
Decision filter
"Before making a strategic commitment to an emerging technology, ask: where on the Hype Cycle is this technology right now? Am I reacting to genuine capability or to the delta between expectations and reality? The Peak rewards attention. The Trough rewards commitment. I want to learn at the Peak and invest at the Trough."
As a founder
The Hype Cycle creates a tactical playbook for timing. At the Peak, raise money — capital is cheap and abundant for anything in the trending category. Use the Peak to build, not to scale. Your product is immature and the market does not yet understand what it needs. In the Trough, double down — your competitors are shutting down, talent is available at reasonable cost, and the customers who remain are the ones with real problems the technology can solve. The Trough is where product/market fit emerges because the tourists have left and only the genuine use cases survive.
The founders who built the defining internet companies — Amazon, Google, PayPal — did their most important work during the 2001–2003 Trough. Jeff Bezos used the dot-com crash to negotiate better supplier terms, reduce operational costs, and build the infrastructure that became AWS. The Trough did not weaken Amazon. It removed the noise that had been obscuring the signal.
As an investor
The Hype Cycle is a pricing framework. At the Peak, valuations reflect expectations, not capability. The median AI startup in 2023 raised at valuations that implied penetration rates no enterprise software category had ever achieved. At the Trough, valuations reflect disillusionment, not potential. Blockchain companies in 2019 raised at valuations that implied the technology would never find real-world applications — which has since proven false in stablecoins, supply chain verification, and digital identity.
The disciplined investor studies every technology at the Peak, builds relationships with founders during the Trough, and writes cheques when the Slope of Enlightenment begins — when the use cases have clarified, the unit economics are provable, and the valuation has not yet re-inflated. Sequoia's investments in Stripe (payments infrastructure after the fintech hype) and a16z's early bets on Coinbase (crypto infrastructure during the 2018 trough) both followed this pattern.
As a decision-maker
Corporate technology adoption should be explicitly mapped to the Hype Cycle. At the Peak, experiment with small budgets and limited scope — run pilots, build prototypes, develop internal expertise, but do not commit enterprise budgets or restructure operations around immature technology. In the Trough, assess which applications survived and which disappeared. The applications that persist through the Trough are the ones with genuine product-market fit. At the Slope of Enlightenment, commit — the technology is mature enough for production deployment, the vendor landscape has consolidated, and the implementation risks are well understood.
The companies that adopted cloud computing in 2008–2012 — after the initial hype but before full mainstream adoption — gained structural advantages that late adopters in 2016–2018 could not replicate. Early adoption during the Slope, not the Peak, creates lasting competitive advantage.
Common misapplication: Treating the Hype Cycle as a precise timing tool. The curve describes a qualitative pattern, not a quantitative prediction. Gartner does not tell you when the Trough will end or how long the Slope will take. The internet's Trough lasted roughly three years (2001–2003). Blockchain's has lasted longer. AI's duration is unknown. The framework tells you what phase you are likely in. It does not tell you when the next phase begins.
Second misapplication: Assuming every technology reaches the Plateau. Some technologies stall permanently in the Trough — 3D television, Google Glass, Second Life — because the underlying capability never develops enough to match even modest expectations. The Hype Cycle is not a guarantee of eventual productivity. It is a pattern that applies when the technology is fundamentally sound but expectations have outrun implementation.
Section 4
The Mechanism
Section 5
Founders & Leaders in Action
The leaders below demonstrate two approaches to the Hype Cycle: one who deliberately built through the Trough and emerged stronger, and one who timed an entire company's strategic pivot to a technology ascending the Slope of Enlightenment.
Bezos is the most instructive Hype Cycle operator in modern business history. Amazon went public in 1997 at $18 per share, rode the dot-com Peak to $106 by late 1999, and cratered to $6 in the Trough of late 2001. The internet had not stopped working. E-commerce had not stopped growing. What collapsed was the expectation that every online retailer would become a billion-dollar business overnight. Bezos responded to the Trough by cutting costs, renegotiating supplier contracts, and investing in infrastructure that competitors — many of whom were shutting down — could not match. Amazon Web Services, conceived during the Trough years, launched in 2006 and became the most profitable division in the company. Bezos treated the Trough not as a crisis but as a competitive clearing event. The tourists left. The builders remained. By the time e-commerce reached its Plateau of Productivity, Amazon had structural advantages that no later entrant could replicate.
Huang positioned NVIDIA for the AI Hype Cycle a decade before it arrived. The 2012 breakthrough — when a GPU-trained neural network won the ImageNet competition by a massive margin — was AI's Innovation Trigger. Huang bet the company on it, pivoting NVIDIA's strategy from gaming GPUs to AI training infrastructure. Through AI's quiet Slope of Enlightenment (2014–2022), NVIDIA invested billions in CUDA, data centre chips, and developer ecosystem while the broader market treated AI as a niche academic pursuit. When generative AI hit the Peak of Inflated Expectations in 2023, NVIDIA was the only company with the hardware, software stack, and production capacity to serve the explosion in demand. The stock rose from $140 to over $500 in a single year. Huang's edge was not timing the Peak — it was building through the Slope so that when the Peak arrived, NVIDIA was the infrastructure layer the entire cycle depended on.
Section 6
Visual Explanation
The curve's distinctive shape — sharp ascent, crash, gradual recovery — captures the core mechanism: expectations overshoot reality at the Peak, undershoot it in the Trough, and converge at the Plateau. The callout box on the right makes the investment thesis explicit: the best opportunities emerge not at the exciting Peak but in the unfashionable Trough, when expectations have collapsed below the technology's actual improving capability. The five phase labels along the bottom serve as a diagnostic checklist — a technology can be mapped to its current phase by evaluating the gap between public sentiment and actual maturity.
Section 7
Connected Models
The Hype Cycle draws on models of technology adoption, cognitive bias, and strategic timing. The six connections below map the intellectual ecosystem: which models describe the same pattern from a different angle, which models explain the psychology that generates the curve's shape, and which models describe the strategic implications of timing decisions relative to the cycle.
Reinforces
Amara's Law
Roy Amara's observation — "We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run" — is the Hype Cycle expressed as a single sentence. The Peak of Inflated Expectations is short-run overestimation. The Plateau of Productivity is long-run underestimation corrected. Amara's Law explains why the curve has the shape it does: the human tendency to misjudge technology's timeline produces predictable overshooting and undershooting.
Reinforces
Technology Adoption Lifecycle
Geoffrey Moore's adoption lifecycle — innovators, early adopters, early majority, late majority, laggards — maps onto the Hype Cycle's phases. Innovators and early adopters drive the Peak. The "chasm" between early adopters and the early majority corresponds to the Trough. The early majority's adoption drives the Slope of Enlightenment. The two frameworks describe the same phenomenon from different angles: the Hype Cycle tracks expectations, the adoption lifecycle tracks actual users.
Leads-to
Paradigm Shift
Technologies that survive the Hype Cycle and reach the Plateau of Productivity often constitute paradigm shifts — fundamental changes in how an industry operates. The internet was a paradigm shift. Mobile computing was a paradigm shift. AI may be one. The Hype Cycle describes the turbulent path a paradigm shift takes from initial recognition to widespread productive adoption. Not every technology that enters the cycle produces a paradigm shift — but every paradigm shift passes through the cycle.
Section 8
One Key Quote
"We always overestimate the change that will occur in the next two years and underestimate the change that will occur in the next ten."
— [Bill Gates](/people/bill-gates), The Road Ahead (1995)
Gates published this the same year Fenn published the Hype Cycle, and the two ideas are mirrors. The two-year overestimate is the Peak of Inflated Expectations — the compressed timeline in which investors, executives, and the media expect a new technology to transform everything immediately. The ten-year underestimate is the Trough-to-Plateau trajectory — the long, quiet period during which the technology actually matures, finds real use cases, and transforms industries in ways that the Peak's breathless predictions could not have anticipated.
The framing is useful because it gives the overestimate and underestimate specific time horizons. Two years is roughly the duration from Innovation Trigger to Trough for most technologies — the period during which expectations inflate and collapse. Ten years is roughly the duration from Trigger to Plateau — the full cycle from initial recognition to sustained productive adoption. Gates was not being poetic. He was describing a structural pattern in technology adoption that Gartner would spend the next three decades documenting in granular detail.
The implication for operators: calibrate commitments to the ten-year timeline, not the two-year one. The company that builds AI infrastructure expecting a ten-year deployment horizon will make fundamentally different architectural decisions than the company that expects "transformation within 24 months." The first survives the Trough. The second writes off the investment and blames the technology.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
The Hype Cycle's greatest value is not prediction. It is pattern recognition. The framework does not tell you when the Trough will end or when the Plateau will arrive. What it does — and what no other widely-used technology framework does as cleanly — is remind you that the pattern exists at all. Every generation of technologists believes that the current revolution is different, that this time the hype is justified, that the crash will not come. The Hype Cycle is a thirty-year empirical record demonstrating that the crash always comes. The technology that survives it is often genuinely transformative. The expectations that preceded it never are.
The most actionable insight is the asymmetry of attention. At the Peak, everyone is paying attention and valuations are high. In the Trough, no one is paying attention and valuations are low. The companies, technologies, and talent that are available in the Trough are available precisely because the crowd has moved on. Amazon hired aggressively during the dot-com Trough — engineers who had been laid off from Pets.com and Webvan joined a company that was building the infrastructure for the next two decades of e-commerce. NVIDIA acquired key AI talent during the years when AI research was considered a backwater. The Trough is a clearance sale on assets whose long-term value the market is temporarily unable to see.
The framework's honest limitation is survivor bias. We remember the technologies that made it through the Trough to the Plateau — the internet, mobile, cloud computing — and we extract lessons from their survival. We forget the technologies that entered the Trough and never emerged: Google Glass, 3D printing for consumers, the metaverse (as of this writing), smart home hubs, autonomous vehicles at scale. The Hype Cycle describes a pattern, not a guarantee. The Trough is a filter, and most technologies do not pass through it. The discipline is not to invest in every technology in the Trough. It is to invest in the technologies whose fundamental capability is improving even as expectations decline — where the gap between dismissed expectations and improving reality is widest. That gap is where the returns live.
The practical test I apply: is the technology's core performance metric improving on a predictable curve, independent of public sentiment? During the 2018 crypto Trough, Ethereum transaction throughput and developer tooling continued improving. During the VR Trough, headset resolution and latency continued improving. During the AI Trough of 2017–2019 (before the generative AI peak), model accuracy on standard benchmarks improved every quarter. When the performance curve keeps climbing while the hype curve is falling, the two lines will eventually cross — and that crossing is the Slope of Enlightenment. When the performance curve also stalls, the technology is not in a Trough. It is at a dead end. That distinction is the difference between a buying opportunity and a value trap.
Section 10
Test Yourself
The scenarios below test whether you can identify which phase of the Hype Cycle a technology occupies and whether the decisions being made are calibrated to the phase or distorted by it. The key diagnostic: are expectations running ahead of capability (Peak), behind capability (Trough), or converging with it (Slope/Plateau)?
The distinction matters because the right action at the wrong phase produces the wrong outcome. Aggressive investment at the Peak overpays. Aggressive investment at the Trough captures value. The framework rewards patience and punishes enthusiasm uncalibrated to timing.
Is this mental model at work here?
Scenario 1
In early 2023, a Fortune 500 company announces a $500 million 'AI Transformation Initiative,' hires a Chief AI Officer, and issues a press release declaring that AI will 'fundamentally reshape every aspect of our business within 24 months.' The company has no existing AI infrastructure, no training data strategy, and no deployed AI features. Twelve months later, the initiative has produced two internal chatbot prototypes and a revised data governance policy.
Scenario 2
In 2019, a venture capital firm closes a $200M fund focused exclusively on blockchain infrastructure. The firm's thesis: the 2018 crypto crash destroyed speculative tokens but left the underlying technology improving. The fund invests in stablecoin infrastructure, enterprise blockchain for supply chains, and developer tooling. By 2022, three of the fund's fourteen portfolio companies have reached $100M+ valuations.
Scenario 3
A startup founder pitches investors on a virtual reality social network in 2016. The pitch deck projects 50 million monthly active users by 2019, citing Facebook's Oculus acquisition and the 'inevitable shift to spatial computing.' By 2019, the total installed base of VR headsets worldwide is approximately 5 million units. The startup has 12,000 active users and has shut down.
Section 11
Top Resources
The Hype Cycle literature spans technology strategy, innovation economics, and behavioural finance. Start with Fenn's original framework and Gartner's annual reports, layer in Perez for the deeper economic theory, and use Amara and Gates for the pithy distillations that make the pattern memorable.
The definitive treatment by the framework's creator. Fenn and Raskino expand the original research note into a full strategic playbook: how to identify which phase a technology occupies, how to calibrate investment and adoption timing, and how to avoid the predictable errors that each phase induces. Essential for anyone using the Hype Cycle as a decision-making tool rather than a descriptive metaphor.
Perez documents the recurring pattern across five major technological revolutions: canals, railways, steel, automobiles, and information technology. Each followed the same arc — an installation period characterised by financial speculation and bubbles (the Peak), a crash, and a deployment period characterised by productive adoption (the Plateau). The Hype Cycle is a micro version of the macro pattern Perez describes across two centuries of industrial history.
Moore's framework for navigating the gap between early adopters and mainstream adoption maps directly to the Trough of Disillusionment. The "chasm" is the Trough experienced at the company level — the period when early adopter enthusiasm fades and mainstream customers have not yet arrived. Moore's prescriptions for crossing the chasm — targeting a beachhead segment, building the whole product — are the tactical complement to the Hype Cycle's strategic overview.
Gartner publishes annual Hype Cycle reports across dozens of technology categories. Each report plots specific technologies on the curve with estimated years to the Plateau of Productivity. The reports provide the most granular, current mapping of where individual technologies sit on the cycle — useful as a calibration tool for investment and adoption timing decisions.
Gates's observation about two-year overestimation and ten-year underestimation provides the most concise articulation of the Hype Cycle's core mechanism. Written in 1995 — the same year Fenn published her framework — the book documents Gates's own experience navigating the gap between expectations and reality during the internet's early adoption phase. The prescience of his framing is itself evidence that the pattern is structural, not incidental.
The Hype Curve — Five phases map the gap between expectations and reality for emerging technologies. The Trough, not the Peak, is where the best opportunities emerge.
Tension
FOMO Components
FOMO is the psychological engine that inflates the Peak. Social proof (everyone is investing in AI), scarcity (limited allocation in the hot round), urgency (the window is closing), and loss aversion (you'll regret missing this) combine to drive expectations beyond any rational assessment of the technology's maturity. The Hype Cycle describes the macro pattern. FOMO Components describe the micro mechanism inside each participant's decision-making that generates the pattern.
Reinforces
Second-Order Thinking
First-order thinking at the Peak: "This technology is amazing, I should invest now." Second-order thinking: "This technology is amazing but immature, valuations reflect hype rather than capability, and the best entry point will come after expectations collapse." The Hype Cycle rewards second-order thinkers who can resist the Peak's emotional pull and wait for the Trough's better pricing. Second-order thinking is the cognitive discipline that converts the Hype Cycle from a descriptive framework into an actionable strategy.
Leads-to
Inflection Point
The transition from the Trough to the Slope of Enlightenment is a strategic inflection point — the moment when the technology's improving capability finally crosses the threshold of practical utility for a broad market. Identifying this inflection point is the highest-leverage timing decision the Hype Cycle framework enables. Companies that commit at the inflection — not too early (the Trough), not too late (the Plateau) — capture disproportionate returns.
