Gartner’s “Hype Cycle”: I don’t think so.

A lot of technologies emerge nowadays and it is often difficult for a company to predict or analyze whether a promising technology is a must-have for the company, worthy to be considered for future investments. It’s not a negligible aspect, since adoption might lead you to become a sustainable player in the field or will sink you to the bottom of the swamp. It is normal for companies to rely on risk assessments and consult sources of information that give an indication of the future directions to take. One of the often-cited models that thrive business models is the Gartner’s hype cycle. We will give a short description of the hype cycle followed by indicating the weaknesses (understatement) of the model. Fact is that everybody tries to predict the future. Don’t be fooled: if we would be able to predict in such a precise way the hypes as presented by Gartner, we’d be foolish to organize proof of concepts, perform market analysis and last but not least, try to think for ourselves about the future that is awaiting us.

Let us start with a short explanation. The “Hype Cycle” was introduced in 1995. It projects, at a certain time, the progression in maturity of an emerging technology with respect to the visibility it acquires within the market space. The different phases are depicted in Figure 1.

Figure 1: The Hype Cycle from Gartner

The first part of the curve is driven mainly by attention in the media. They speculate on the use and benefits of an emerging technology. This is followed by the performance gains and the respective adoption.  The technology trigger generates the interest but no usable product exists. At the rise, products will emerge. At the peak of inflated expectations, the number of offerings for suppliers will increase, but the product has still many child diseases. However, companies are staring to take a look at it to see if they have to adapt their strategies. Because the technology does not live up to the expectations, which contrasts to the high expectations expressed by the media, the technology gest discredited and we pass to the trough of disillusionment. As the product gets improved, the technology climbs into the slope of enlightenment and finally hits the plateau of productivity, where mainstream adoption is achieved.

The reason why predictive models are always very attractive for companies lies within the fact that they intend to predict future trends. Companies and management that has no clue whatsoever what is needed for the future of their companies will depend on those models for their decisions. Wrong! Predicting the outcome of future technologies is a hazardous business, or as Bernard de Jouvenel (1967) expressed that knowledge of the future is a contradiction, since the future does not exist. We do not deny that methodologies have been developed which consider various sources to guestimate the outcome of technologies, but at the end, there are so many variables in play, that predictions simply turn wrong. The model lacks theoretical foundation, methodological procedure or empirical validity. We are simply not very good at predicting the future. You may indicate that some of the technologies, mentioned of the hype-curve, have become leading technologies. But we often forget the percentage of cases and predictions when the model was wrong. You have to keep in mind the complete picture, not only the winners. We mention some of the comments and criticism:

• Kernohan (2104) stated that it presents the graph as an external, “natural”, process - separate from human intervention. The cycle (and accompanying guidance) is sold as an investment aid. You “understand” the hype cycle, you don’t “use” it. It’s a map of the future. A future that cannot be changed. It reinforces our greatest secular myth - that “it will all turn out right in the end”. Technologies, no matter their idiocy, will eventually sail up onto the plateau of productivity. The difficulties - why, that’s just the trough of disillusionment. Soon the world will somehow see the light (without intervention, mind you) and the slope of enlightenment will be scaled. Our own experience tells us this is not true, but so desperate are we for it to be true we believe it anyway. Most technology is awful, it doesn’t work and it causes us endless pain trying to make it work. People will get remunerative careers in helping us to get within touching distance of the initial promise. Eventually they will write books and articles, run conferences and workshops, and the problem will be filed as completed.
• They have tied to stitch in some scientific models, I guess to increase the credibility. However, when we take a look at the S-performance curve, easily to be constructed by accumulating the adoption categories over time, we notice that there is an intended relationship towards the early adopters. But when early adopters, also known as opinions leaders, are confronted with solutions that do not bring any relative advantage with respect to the current situation, it is hardly to believe that they will positively influence the majorities, as indicated in the model. They try to link the chasm (Geoffrey Moore) to the trough of disillusionment.  
• They are, of course, not blind to their own mistakes. Therefor, they try also to classify the failed predictions into categories, which are denominated as special hype cycle circumstances. We mention phoenix technologies (e.g. artificial intelligence), which continuously cycle through enthusiasm and disillusionment. Another category is ghost technologies (e.g. interactive TV and Video On Demand), which are put on hold because they have not delivered their promises. And last but not least, it is mentioned that as part of the normal evolution of technology, the target audience may change to what was originally intended. The problem lies within the model of prediction itself, not. Kahneman (2011, Thinking Fast and Slow) mentioned that the odds of good investments in the stock market over a longer period of time are not higher when made by a darts-throwing monkey than by your fund managers. However, the illusion continues. There are so many variables that can influence the outcome of a technology that makes it very hard to model, hence the numerous explanations for failures. 
• Khan (2012) mentioned a practical example regarding big data: “Sorry, Gartner, with all due respect, you're dead wrong. Actually, the expectations from big data are underinflated as far as business, science, government, and education are concerned. That's because the benefits these market segments get from big data are not theoretical, they're real.” (referring to the hype cycle of 2012). Big data is already well along on the so-called Plateau of Productivity as its countless success stories already prove. Skeptics who doubt this are like people who once derided the PC as not being a serious computer. Today, it is those big data skeptics that we should not take too seriously.
• Fisher (2014) indicated that the Hype Cycle is most useful when it’s viewed as descriptive, not prescriptive. The fact that a particular technology is at the peak of the Hype Cycle, or the bottom of the trough, doesn’t mean that there’s anything wrong with the technology. Good technologies, products and vendors all go through this cycle. Just because some poor technology happens to be in the Trough of Disillusionment doesn’t mean it should be dropped or dismissed.
• Mullany (2016) also indicated that the median technology does not obey the hype cycle. We only think it does because when we recollect how technologies emerge, we're subject to cognitive biases that distort our recollection of the past. He mentioned that (1) we are very bad in predicting the future, (2) an alarming number of technology trends appear, (3) about 20% of all technologies just dies, (4) the insight might be right but the proper implementation lacks, (5) recycling from recurring technologies is observed, (6) many major technologies flew under the radar of the hype cycle (x86 virtualization, noSQL, open source) and (7) lots of other technologies make tremendous progress when no one is looking at it. 
• Some examples where the predictions went wrong. Personal digital assistant phones (2002) were out ruled, also better known as smartphones. Tablets sales (driven mainly by iPAD) went straight trough the roof, no hype cycle to detect at all.

For the Gartner’s hype cycle, from the more than 200 unique technologies that have ever appeared on a Gartner Hype Cycle for Emerging Technology, just a handful of technologies have been identified early and travelled somewhat predictably through a hype cycle from start to finish. Now, despite the fact that we have criticized the hype cycle doesn’t mean you have to throw it away, but at least we have a pretty good idea now of its severe limitations and shortcomings. The questions to be answered within our own company are rather related to how we must deal with predictions of technology in general. That is; how dependent have we become on that kind of information to determine our future (investments). We must not waste energy in being negative about the difficulty to predict emerging technologies, but be excited about the continuous progress being made exists to drive our own future. Are we armed enough to define and drive our own future? This does not diminish the fact the Gartner enumerates emerging technologies. They are simply wrong in the fact to try to squeeze them into a predictive model.

Conclusion

Companies are eager to know whether to invest in an emerging technology or not. The outcome is not negligible: it may lead you to become a sustainable player in the field or will sink to the bottom of the swamp. That’s why a lot of companies use the hype cycle of Gartner. However, as we have shown within the article, we are simply not very good at predicting the future. From the more than 200 technologies appeared on the hype cycle, just a handful has gone through the predicted hype cycle. We would like to believe it is much more, but we're subject to cognitive biases that distort our recollection of the past. Therefore, it is a must to handle information regarding the future with great care and even suspicion. We must not waste energy in being negative about the difficulty to predict emerging technologies, but be excited about the continuous progress being made exists to drive our own future. 

References

De Jouvenel, Bertrand. (1967). The art of conjecture. New York: Basic Books.

Kernohan, D. (2014). A New Order:A collection of short articles and miscellanea from The Followers of The Apocalypse. Seattle: Amazon Media.

Kahneman, D. (2011). Thinking, fast and slow. New York: Farrar, Straus and Giroux.

Mullany, M. (2016, December). 8 Lessons from 20 Years of Hype Cycles. Retrieved from https://www.linkedin.com/pulse/8-lessons-from-20-years-hype-cycles-michael-mullany/

Fisher, S. (2014, August). Why the Gartner Hype Cycle is Like Your Last Girlfriend. Retrieved from https://www.laserfiche.com/simplicity/why-gartner-hype-cycle-your-last-girlfriend/