Tech Investors: Pay Attention to Hype Cycles and Adoption Life Cycles

In this podcast, Motley Fool Live's This Week in Tech co-hosts, Tim Beyers and Tim White, discuss:

• How investors can think about adoption life cycles and tech investments.
• Where generative AI lands on the hype cycle.
• One key sign that a new product has "crossed the chasm" for widespread adoption.
• ChatGPT's "nice to haves."

To catch full episodes of all The Motley Fool's free podcasts, check out our podcast center. To get started investing, check out our quick-start guide to investing in stocks. A full transcript follows the video.

This video was recorded on June 10, 2023.

Tim White: If you think about the Apple I, maybe that's where the innovators were, people who were willing to take a risk on it at the time, an inflation-adjusted $5,000 or $6,000 on a computer that basically did nothing. You had to make it do anything you wanted to make it do. It really led to a lot of people at the time in that trough of disillusionment that happened with personal computers of like, no one will ever need a PC in their homes, right?

Tim Beyers: [laughs] Right.

Tim White: Big statements from big, fancy people.

Mary Long: I'm Mary Long, and that's Tim White, who co-hosts This Week in Tech on Motley Fool Live alongside Tim Beyers. Tim and Tim caught up on Motley Fool Money to discuss hype cycles, adoption curves, and why investors should pay attention to their key differences.

Tim Beyers: Let's talk about promising technologies. And you've seen a boatload of them -- so have I -- over the course of years. What do you think when we think about how long it takes genuinely like a really promising technology -- AI is in the news now, right? A genuinely promising technology, something like an artificial intelligence toolkit or artificial intelligence generally, generative AI like ChatGPT. How long do you really think that takes to become part of our daily lives? We're hearing about it constantly, but it's not really part of our daily lives yet.

Tim White: Again, we're talking about two different cycles. The hype cycle, which is people getting excited about things and the beginning of technology. Then that adoption cycle, where you have early innovators getting on board with things and then eventually adopting things in the general-public way at the end of that cycle.

All of the stuff we've seen from AI has been around for a long time. What made it cross the chasm to more people knowing about it and more people using it in their daily lives was making a chatbot free. And I think the free part is so important there, right?

Tim Beyers: Absolutely. We should talk about the difference between what we call the hype cycle or, more specifically, what Gartner -- which is a research firm which you may or may not have heard about, they're a public company, ticker symbol IT -- Gartner defines something called the hype cycle. Then there is the more commonly referred-to technology-adoption life cycle.

Each of them are bell curve. The hype cycle is really compressed bell curve, and the technology-adoption cycle is a bit more like a regular bell curve with a pretty big gap in it. That gap was defined in 1993. I hope I have this year right, Tim, but this is, if I have my history right, in 1993, Geoffrey Moore, who's a consultant and still to this day operates The Chasm Group. The consultant who wrote a book called Crossing the Chasm. And in Crossing the Chasm, Moore defined what a technology-adoption life cycle looks like. He said, there comes a point when those and all the enthusiasm, all this stuff in the hype cycle has to go across this big chasm where the people who are really excited convince the rest of us to say like, OK, this is real. We'll actually spend some money on this.

When we talk about these two different cycles, the differentiator, I think we both have identified, and we've talked about this so many times, is if you're talking about tech adoption, you're talking about solving what you and I have called a migraine-level problem. If you're talking about hype, what you're really talking about is some spending around excitement. Boy, this thing is neato, and I want to do some things with it.

This is why I'm going to come back to what you just said about free. Free is so important in the hype cycle part of the phase.

Tim White: If we think about hype is the idea that, wow, this technology could change everything. We'll never have to drive cars again. We'll never have to think for ourselves again. We'll never have to listen to the radio again because we can watch television. All these, like, it'll change everything. That's what the hype cycle is, and it ramps up to the peak of inflated expectations, as Gartner calls it, where everyone's like, this is going to do everything.

Then there's a crest at which suddenly, it doesn't deliver, and things start to fall apart, and actually turning that "this will change everything" just falls apart, and the technology is cool. But it doesn't solve anybody's particular migraine-level problem.

That's where I think you really transition over to adoption, where there's a particular problem that some piece of technology solves, and that's when it starts to become more mainstream.

Tim Beyers: I'm going to guess here, but if you had to say where generative AI is in the hype cycle, is it at the peak of inflated expectations where now the hallucinations, so I go, wait a minute, maybe this thing doesn't give us exactly what we thought it was going to give us?

Tim White: Yeah. I think it's somewhere along the top there. I think it's funny that in 2021, Gartner listed chatbots and the trough of disillusionment [laughs] I think the worst case, where everyone's like, oh chatbots. They were going to save everything. They were going to let us fire all of our customer service agents. Now they're terrible, and no one wants to use them anymore. They were in the trough of disillusionment in 2021, and then they magically vanished from the hype cycle for AI in 2022.

Tim Beyers: Right.

Tim White: [inaudible].

Tim Beyers: It's a Jedi mind trick. This is not the technology you're looking for.

Tim White: Right. I think we're absolutely at the peak of inflated expectations around generative AI right now. But I think we're starting to slide down the backside and toward that trough of disillusionment, where people are wondering. This is cool, but there's so many little gotchas. Will we actually be able to make this a real part of our business?

Tim Beyers: For sure.

Let's talk a little bit about that switchover when the hype moves into the adoption and what Moore defined as the chasm here. We said that successful products always solve a migraine-level problem. There's always something. If we can look back through history, and when we talk about the chasm, the chasm is defined by the types of customers that are using a product.

On the left side, so you think of a bell curve on the left side. The real enthusiasts are the innovators and the early adopters. Then you jump the chasm to what's called the early majority, then the late majority, then the conservatives, then the skeptics. In other words, that group on the right side of the chasm has to have a reason, like a real business case, in order to spend money, Tim.

I'm wondering if we think about this, there are different types of technologies that we've seen cross the chasm. Yesterday, when we were prepping for this, we talked a little bit about home computers, which really were, I mean, I know we've been at this for 30 years. When we were kids, home computers were, I mean, boy, it was a privilege to have one. I think you said, I felt the same way. We got our Apple IIe from my uncle, who was a very early adopter of computing, and I mean, it was unusual in the early 1980s.

Tim White: Yeah. I think if you think about the Apple I, maybe that's where the innovators were, people who were willing to take a risk on it at the time, an inflation-adjusted $5,000 or $6,000 on a computer that basically did nothing. [laughs] You have to make it do anything you wanted to make it do. It really led to a lot of people at the time in that trough of disillusionment that happened with personal computers of like, no one will ever need a PC in their homes.

Tim Beyers: Right.

Tim White: That was a big statements from big, fancy people.

But the early adopters, people like your uncle, my father, who bought an Apple II plus, were like, we need to have our children have a chance to use this because these computers will be the future. That was a huge privilege to be able to have a computer like that in my home. Of course, I immediately grabbed onto it and then really never let go.

But those early adopters are what give companies enough money and enough feedback. This is the beautiful thing about a first-version product is you get feedback from your customers, and then you can make your product better and better. That's where the Apple IIe, like you talked about, suddenly hit the education market and really exploded and took off and really made Apple up until the Macintosh came out.

Tim Beyers: Apple, in some ways, found its way at least into the chasm and started bridging across through things that made that computer or the computers that Apple was making a lot more useful for solving a business problem. I'll use the example of one of the great early apps that made Apple's machines incredibly useful for the business community. I mean, I know you know this one, but there's a lot of people who've probably have never heard of VisiCalc.

Tim White: Yeah. Dan Bricklin created VisiCalc while he was watching a presentation at Harvard Business School. He was watching this presentation and realized that the financial model that was drawn on a blackboard is something that he could create on his computer and started working on it on the side. It really became the first spreadsheet as we know it. Of course, led to Lotus 123 and Excel and all the things that we use today.

But VisiCalc really gave people a true reason. Solved a migraine-level problem of people needing to keep track of budgets and other kinds of things that we now use spreadsheets for. People suddenly said, "I do need to have a computer because I can use VisiCalc."

Tim Beyers: This is probably, I would say, the very beginning of some businesses deciding as things were crossing the chasm here. I can use a computer to manage my business. I actually don't need to use a paper ledger anymore. I can automate some of this. We've never gone back from that. You end up with these little use cases that end up being worth spending quite a lot of money on.

Let's talk about the through line here, because there's enthusiasm and then there's practical desire to spend. You just pointed this out, that you need the enthusiasm, you need the cheerleading to get people thinking about the practical. But when do you think that flips?

I'm going to bring up another one that we talked about yesterday. There are moments where a technology has all sorts of promise, and you do have a lot of cheerleaders, and it ends up going all wrong. I think you know where I'm going with this one because it's on our list. There's the CASE tools, which I know we've talked about before.

Tim White: Yeah. In the '90s, there was this huge rush toward computer-aided software engineering.

Tim Beyers: Yes.

Tim White: Something about CAD. You may have heard CAD as computer-aided design. CASE was computer-aided software engineering. It was this idea that you can take a piece of software and make a drawing, like a diagram of what you want your software to look like. Press a button, and it will generate all the code for you to do that.

Of course, that never really turned out to be true. In the same way that the current generation of generative AI can't really write all of your code for you. It certainly can help, just like the CASE tools could help. But in the end, I think a lot of people realized that the CASE tools were really just adding time and not actually eliminating work.

Tim Beyers: I'm going to come back to the free tools in a minute here, because some of the economics of what's changed is making the technology-adoption life cycle arguably a little more compressed. But at that time, the cheerleaders were so vocal about this that there was a lot of investment in things like Unified Modeling and tools like Rational Rose. We think, this is going to change everything. We're going to have businesspeople, marketers, and salespeople are going to be able to define what business process they need. They're going to learn Unified Modeling Language, and they're going to draw the workflow that they need. Then the code is just going to magically pop out. And it just became an exercise in disappointment here.

Coming back to free, which is where we are now. A lot of tools, due to a whole confluence of things, your open-source movement and so forth, we can try a lot of things for free right now. Generative AI, ChatGPT, we're trying AI for free, and we're just getting enthralled with it. Do you think because of this prevalence of free tiers, that what used to cost us something, like it cost you something to be a cheerleader in the 1970s, 1980s, and now it doesn't cost you anything anymore. Does that dramatically alter the economics of the technology-adoption life cycle?

Tim White: I think it does because either expectations can be very low. If you're spending $3,000 on something unless you're a super early adopter. I'm looking at Apple Vision Pro. [laughs] Unless you're a super-early adopter, spending that kind of money, you have very high expectations that this is going to be a product that's going to solve some problems for you. Whether that problem be boredom, [laughs] entertainment, whatever.

But if you get it for free, your expectations are at literally the bottom. It really helps to get innovators in the door. If they can get people to use things for free, give them feedback, get increasingly better and better products out the door to the point where eventually, they can charge for things because they actually have a product that does meet expectations.

I think Linux is another great example of a tool that was initially free, and it was a very limited operating system when it first came out. But because of a lot of work that happened in the '70s and '80s, creating free software for Unix operating systems, it immediately had a bunch of tools that solved people's problems. And it was the peak of the time when Linux came out, when companies like Sony and HP were charging really large amounts of money for licenses for Unix.

Tim Beyers: You don't have to be that kind. You could say obscene.

Tim White: [laughs]

Tim Beyers: Right.

Tim White: I just remember like AT&T Unix was costing upwards of $1,000 per machine that you installed it on at the time. It was just crazy, because people would have racks and racks of these machines that they would all have to have licenses for. Linux absolutely changed the game by saying no, you can literally spin up a computer with an operating system on it, put it on the Internet for no money.

Tim Beyers: But what's interesting about this, Tim, is if the cost is eliminated up front, the worry I have is that you'll see more bad products because there really is no gating factor. Cost is not a gating factor anymore. You just release it out into the wild, and it can be a terrible product.

Tim White: Sure. I do think that there's some fear of flooding the market, and I think we're certainly seeing that with AI tools now. Just like a couple of years ago, we saw that with cryptocurrencies, like there's a different currency every week, and that's because the cost of entering the market was zero, right? It cost you nothing to make a new one, and so everyone made one.

I think that's still going to be true, but the good news is, as we've often discussed, user experience trumps everything. If you've got a really easy-to-use tool that's very simple and very reliable, that will win over a tool that is otherwise similarly priced, e.g., free. I think you end up competing a lot on user experience.

Tim Beyers: Let's talk about when do we know, like as investors... A lot of bad products can come to market quickly because the cost to introduce products now has gone way, way down. Free is the new model here. How do we know when a product or a company has found its way across the chasm?

There's a couple of indicators I think we can talk about here. I'll kick it to you first and tee you up with this one.

I think when you have seen, either in a vertical industry or a set of customers, something you can define, you could point and say, those people have made it very clear that they need this product. In the case of like the original Mac, the desktop publishing as a practice and the graphic design community said, "You can have this computer if you take it from my cold, dead hands."

Tim White: I think what you just said is the classic business version of crossing the chasm, which is as soon as your salespeople start telling the IT department to shove it when the IT department says no, you can't have that, [laughs] that's when you've crossed the chasm.

And a great example, of course, is when the iPhone came out. Suddenly, every sales exec had one of those. Everybody had to have one, and they really wanted to use them for everything, for mail and for all this stuff. And of course, the IT department freaked out and said they're not secure, you can't use that, you can have it. And of course, President Obama notoriously wouldn't give up his BlackBerry.

Those are the things that you know you've crossed the chasm, when people are demanding that they use them in their business environment, even if there's strong resistance.

Tim Beyers: Yeah, so there is, the loyalty indicates to you that look, this solves my problem. What we said before, this is a migraine-level problem for me, and there's absolutely no way you're taking this away from me. Some sign that a group has said, absolutely, there's no way you're taking this away from me. That's the evidence of a migraine-level problem being solved.

When we think about this, I'll take an example of a company that I think has crossed the chasm. Not recently -- it's been a while -- but I do think there's ample evidence to say, just using a software product, I think MongoDB crossed the chasm a really long time ago. Because there is a number of instances where it's so easy to develop a piece of software and attach that database to it that developers are never letting that go.

Tim White: Yeah, I think that's true. And of course, any of the cloud hosting companies are in that place where people want to host on [Alphabet's] Google Cloud, want to host on [Microsoft's] Azure, want to host on [Amazon's] AWS, and there's a lot of people just assuming that that's going to happen now.

I think when you assume that's going to happen, that's a big difference from like I interviewed the CTO of a company called TEU years ago, and he said, "When I first said we're not having any servers around, we're doing everything on AWS, everyone thought I was crazy and now they're like, wow. Yeah, that's how you do things now."

Tim Beyers: Yes. When you reach that "of course that's what you're doing" moment, I think that's evidence that you have crossed the chasm.

As investors, if we're wrapping this up a little bit here, I think what we've said as we were talking about this, it's probably better as a public-market investor, to be on the other side of the chasm. In fact, I would say it is universally better to be on the other side of the chasm. The left side of the chasm where you're still working with cheerleaders, that's a good place for venture capitalists.

Tim White: I mean, the true huge money gets made from there, but the true huge money also gets lost from there. They're making a lot of very expensive bets on that before-adoption side of the chasm, and most of them don't pan out.

Tim Beyers: There's a common term that gets bandied about a lot, particularly among executives who specialize in things like product-led growth. Venture capitalists use this too. They call it a tipping point, and the phrasing you'll hear sometimes is called product-market fit. Product-market fit. And product-market fit means just really dumbing it down.

Tim, this is the way I think about it. It's we've got a product and we found a migraine-level problem, and those two have met, and now there's an explosion of demand.

Tim White: We actually have a product that people want to buy eventually, and all we need to do is figure out how to get more people to buy it, not to get anyone to buy it. [laughs]

Tim Beyers: We need to be able to satisfy demand at scale, get those people satisfied, grow the pool of those people, and then get other people around them talking about it.

Tim White: I think in terms of investing, one of the main takeaways that I am always suggesting to people is to look for things that are right after that adoption has really hit and companies that are really ready to really hit really hard on some big question.

One that hit that way for me personally in my investment career was HubSpot. That was a product that has a lot of different features now, but at the time, it was mostly a CRM, so customer relationship management, and email marketing platform.

At the time, Salesforce was dominating that industry, and no one thought that another product could really crack into the market. But HubSpot found the product-market fit of small businesses, very small businesses, solopreneurs, designers, folks like that who just need something more than a spreadsheet and a simple WordPress website, but they're not wanting to pay the premium to use Salesforce. And they have utterly dominated that market in the last few years.

Tim Beyers: Yeah, and they show no signs of slowing down, and they've been able to... Where these companies get really interesting and ones you want to hold for a long time, and HubSpot is this company. Once you solve one migraine-level problem for a particular type of audience, that same audience gives you permission to solve another migraine-level problem for them.

And HubSpot, boy, did they lean into that like very few other companies I've ever seen. It was inbound marketing. And then you had those customers saying, hey, could maybe you help with my sales pipeline? Yes, we can. They built a hub around that. And then they built a hub around web design, and they built a hub around support.

So customers, where this ends up again drawing the through line between hype and when you actually get adoption when you're at the hype stage and you've found cheerleaders, they are very excited about what your potential is. When you're on the other side of the chasm, you are now pain relief for a well-defined customer base, and that customer will come back to you and say, what else can you do?

Tim White: As long as you can continue to deliver on that -- which not every company can, including Salesforce; they have done it for a long time, and now maybe they're perhaps struggling a bit. If you can continue to deliver on that, then you can continue to increase your revenue per customer and solve customer pain, and they will stick with you for a long time.

Tim Beyers: Let's end this by making -- because we always like to make reckless predictions here. This one's a bonus. We didn't talk about this up front here.

I think we both agree that generative AI is still stuck in the hype cycle. It's still on the hype side of the equation that we haven't yet seen a general product-market fit for generative AI yet. Tim, if I had to give you a time frame, how long do you think it takes generative AI to get genuine product-market fit where it is solving a migraine-level problem?

Tim White: I think it's already solving some problems for some people today. There are people who get benefit from using ChatGPT as is right now. But for free, right?

Tim Beyers: Yes.

Tim White: That's the thing. Where I think we really want to think about is when is the product worth enough money to someone that if it went away, the David Gardner snap test, if this goes away, will I be like, well, it went away for free, [but] I'm willing to pony up $5 a month, $20 a month, $50 a month to keep using it. That's where I think the real heart of your question lies.

I think that could happen in less than a year if the people who make these tools continue to push. And there is definitely an arms race between tools now, which definitely leads to accelerated tech excitement.

Let me try this again. There's a lot of people competing on this right now, which definitely leads to tech going very fast in terms of how well it gets better, but will that be really useful to a lot of people in their daily life soon? I don't know. Apple was very careful not to say "AI" in their big announcements, and I think that's telling that they don't think it's there yet.

Tim Beyers: I'm going to take that side of the prediction equation here and play, as I sometimes do, play get-off-my-lawn guy here for a second and say, I think it's at least three years. And the reason I say that is because I think you need to identify what kind of data and what kind of data problems are so specific and so hairy that they need AI to solve them. I don't think we've defined that yet.

To your point, I think ChatGPT has found a whole bunch of nice-to-haves, and that's interesting, and that's where the cheerleaders live. But I think the need-to-have, must-pay-for, don't-do-this-and-we-feel-severe-pain. Those data problems, I don't think they've been well defined yet, Tim, and so I'm giving it three years. But then again, I get curmudgeonly at this stuff.

Mary Long: As always, people on the program may have interest in the stocks they talk about, and the Motley Fool may have formal recommendations for or against, so don't buy stocks based solely on what you hear. I'm Mary Long. Thanks for listening. We'll see you tomorrow.