Bulletin 8 February 2019. On the Internet of Things, and when is a market not a market?

Don’t question the number

One thing I’ve noticed in this topsy-turvy world of estimates and opinions is just how diverse they can be. A cursory glance at Internet of Things market size predictions from 2015, 16, 17 and 18 (hat-tip to Louis Columbus for rounding everything up so neatly every year) shows two themes. First, how it is difficult to identify a baseline (are we talking consumer IoT or Industrial IoT, is the latter the same or different to IoT in manufacturing/utilities, etc etc). 

But second, how wildly different the figures actually are. “Gartner predicts 6.4B connected things will be in use worldwide in 2016, up 30% from 2015, and will reach 20.8 billion by 2020,” said one 2016 article. But wait, a year earlier, we heard that “The number of connected devices is projected to grow from 22.9B in 2016 to 50.1B by 2020.” 

Now, this is not the moment to pitch Gartner against the World Economic Forum: no doubt both would say, “read the small print” in which case we could go into some kind of “my definition is better than your definition” posturing. I’m not sure why such comparisons are rarely, if ever made; perhaps we fear some kind of response, or (worse) perhaps nobody really cares anyway. 

What? Am I really saying that such huge efforts of market sizing fall upon deaf ears? Not precisely; equally I know just how many times I’ve sat through a presentation with some huge figure attached to a market area, to think to myself, “I guess that’s quite big, then.” Sizing a market defines a market; once the exercise has taken place (and the market is defined), the size becomes less relevant. 

Which is why I don’t think anyone ever really questions the figures. Once we have a market, or an industry, or a solution space or what ever you want to call it, once it has a life of its own, in some ways it defines itself as a self-fulfilling prophecy, or simply as something nobody ever got sacked for talking about. In this case, probably the most important consequence is for chip and networking manufacturers, as they justify their strategies to investors and indeed, analysts — which makes for a strange feedback loop. 

Which is interesting, because it may not actually exist in this case. Eh? Aren’t I shooting myself in the foot, given that I’ve written reports on the topic? Let’s think about this. First, computers have two choices, either to sit outside of the rest of the world, or to be in some way embedded: that isn’t new. Second, sensor-based monitoring and remote control have existed since the invention of the thermostat, or RADAR, or (insert your preference here): not new again. 

So can we really talk about a ‘market’ based on the continuation of either of these themes? Back in the day, when I was spending more time on this stuff, I argued for a closed-loop model, i.e. if you can’t sense, analyse, decide and influence a connected device, you probably haven’t got IoT. The rest is just more sensors, which might be worth measuring in itself but it possibly wouldn’t command the kinds of figures ($520B, according to Bain last year) to make for a ‘market’. 

Like it or not, the industry needs ‘markets’. We may not agree with them, in which case we can write things that disagree (and in a strange way, endorse the conversation). They may outlive their usefulness, if they ever really existed at all (cf current rhetoric around cloud/hybrid/multi-cloud; I’m also reminded of predictions around ‘smart phones’). At least we can all take solace in the fact that the market for markets is growing at a CAGR of 30% Y/Y, and looks like it will continue at least unto 2030. Whatever that means. 


More Smart Shift: The information fire hydrant

The next big section in Smart Shift covers how software builds upon the massively powerful computer architectures we see today. First off (following a digression into the golden age of steam, with a mention of my old colleague and ferry master Rob Bamforth), I look at the vast volumes of information we have been generating. 


That’s all for this week, thanks for reading!

Cheers, Jon

Note. This week’s photo comes from a Cisco IoT event in Barcelona, which saw a bunch of seasoned experts staring into rubbish bins. 

Also published on Medium.

Bulletin 8 February 2019. On the Internet of Things, and when is a market not a market?

Leave a Reply

Your email address will not be published. Required fields are marked *