Developed by Jackie Fenn at Gartner (When to Leap on the Hype Cycle, 1995) and expanded with Mark Raskino (Mastering the Hype Cycle, 2008).
Places a technology on a curve from trigger through inflated expectations, disillusionment, and gradual enlightenment to a productive plateau, to time investment and expectations.
One technology moving along the hype curve, and what each phase means for buying
The line is the market's expectations over a technology's life: a trigger, a peak, a trough, a slope back up, and a plateau. The square is AI-drafted loan memos; move it along by the months since the first convincing demos, and the band says what that phase means for a bank deciding whether to buy. Illustrative.
Expectations in the market
What it means for buyingTrigger: Early, and few outside the labs have heard of it. Watch it, and fund a small experiment only if it touches the core business.Peak: Everyone is talking about it, prices and promises are at their highest, and references are thin. Wait, and watch who goes live.Trough: The first projects disappointed and attention left. Evaluate now: vendors are hungry and the real uses are getting clear.Slope: Peers report results on specific uses and expectations rise again for sound reasons. Invest, starting with the proven uses.Plateau: Mainstream and dull. Buy it as a standard product, on price and support.
An example
The technology committee of a regional bank had budget for two new bets and six technologies on its list, each with a champion. Every vendor pitch said now was the moment.
The chief information officer asked for one page per technology: when the first demos appeared, what the trade press was saying, and how many banks of their size had it in production. Then the committee placed each on the hype curve. AI-drafted loan memos were at the peak: constant pitches, no references. Blockchain trade finance was in the trough, its industry pilots closed and its vendors discounting. Back-office automation bots were on the slope, with two peer banks reporting steady savings on loan boarding. Mobile check deposit sat on the plateau, and a voice assistant peers had launched years before had never recovered.
The committee funded the bots for the two proven uses, bought the next check deposit release on price, put loan memos on a watch list for a year, and stopped the voice assistant's budget line.
Why it works
Expectations for a new technology move faster than the technology does. The first demos promise too much, attention piles on, early buyers pay top prices for immature products, and when the first projects disappoint, attention leaves just as the real uses are being worked out. Buying on the noise means buying at the top and quitting at the bottom.
Placing a technology on the curve separates the noise from the evidence. The useful questions are concrete: how long it has existed, what the press says, and who has it working at scale. Each position then implies a stance, from wait to evaluate to invest. Allowing that some technologies never climb out of the trough keeps the curve from becoming a promise that patience always pays.
The thinking behind it
Jackie Fenn Jackie Fenn, an analyst at the research firm Gartner, introduced the hype cycle in a 1995 Gartner report, When to Leap on the Hype Cycle. A new technology rises to a peak of inflated expectations, falls into a trough of disillusionment, and only then climbs to steady productive use. It is a judgment chart, not a measured law: analyses of Gartner's own charts have found that few technologies pass through the whole cycle.
Mark Raskino Mark Raskino, also a Gartner analyst, wrote the book-length account with Fenn, Mastering the Hype Cycle: How to Choose the Right Innovation at the Right Time (2008). It sets out the pattern in which expectations rise to a peak, fall into a trough of disillusionment, and recover to a plateau of productivity, and how an organization can time its adoption. The model offers no data-based test of where a technology stands.
How to run it
- Name the technology precisely. Define it as a use, such as AI-drafted loan memos, since a broad category sits at several points at once.
- Collect the evidence of position. Note when the first demos appeared, the tone of the press, and how many peers have it in production.
- Place it on the curve. Put it at the trigger, the peak, the trough, the slope, or the plateau, and write the evidence beside it.
- Take the stance its place implies. Watch at the trigger, wait at the peak, evaluate in the trough, invest on the slope, and buy as a standard on the plateau.
- Ask whether it will ever leave the trough. Look for uses that are improving, and stop funding a technology that shows none.
- Revisit every year. Place each technology again with fresh evidence, and record which ones moved.
Where it goes wrong
The curve is read as a law. Few technologies follow the whole path, and some peak and vanish. Treat each placement as a judgment backed by evidence, and keep the never-recovered corner in view.
The placement follows the champion. A sponsor who wants the budget will see the slope where others see the peak. Place technologies against the same three facts each time: age, press tone, and production references.
Waiting becomes the policy. A committee that waits at every peak and every trough never adopts anything until competitors have. When the evidence says slope, fund it, starting with the uses peers have already proved.
When to use something else
- Crossing the Chasm. When the question is selling a new technology, and which group of buyers to win next.
- Technology S-Curve. When the question is the technology's real performance over time, more than the market's expectations of it.
- Real Options Analysis. When the choice is how much to spend now for the right to invest more later, with numbers on each path.
Where it sits in SOLVE eX
Go to the source
- When to Leap on the Hype Cycle, Gartner Group report, 1995
- Mastering the Hype Cycle: How to Choose the Right Innovation at the Right Time, Harvard Business Press, 2008