Wi-Fi: Past Performance is No Indication of Future Results

Predicting the future is never easy, but relying purely on the past as an indicator of what might happen next is hardly a good idea – especially with respect to rapidly-evolving technologies like Wi-Fi.

You might think, since we're more than 20 years into the wireless-LAN revolution, that things would be pretty settled and stable in the Wi-Fi world at this point. I'm often reminded that practical radio itself is well over 100 years old (and the theoretical underpinnings of radio much, much older than that!), but the rate of innovation, and across the board, from theory to basic technologies, standards, chipsets and other components, OEM/ODM subassemblies, products, systems software, pricing, distribution, marketing programs, applications, and, OK, maybe not industrial design, has never been greater. This, of course, implies that opportunities for players, also across the board, in the WLAN space have also never been greater, but such is, granted, debatable - and at the core of two conversations I recently had.

Case in point #1 - I got a call from a reporter who wanted to talk about Wi-Fi chipset vendor Quantenna. He was looking for a comment from an analyst regarding the chances the firm might have in the market, citing statistics he got from another analyst that the Wi-Fi chipset leaders appeared to be gaining market share, implying that a smaller player like Quantenna is simply wasting their time. And, indeed, perhaps we should be expecting a shakeout in Wi-Fi chips, as these items become more complex. But as you might surmise from my comment above regarding the rate of innovation today, nothing could be further from the truth. Innovation at any level in the food chain can open doors to new opportunities. The "me, too" phenomenon, typical of commoditized markets and driven purely by manufacturing economies of scale, is not at work here today. Even with 802.11ac poised to become "it" for a good number of years, there's no evidence of slowing innovation in Wi-Fi, and significant evidence of enhanced and growing market opportunities. I'll shortly review a couple of APs from a new player in this space - yes, a new entrant into W-Fi. And things are hardly boring there - it's still all about diversity, in technologies, chips, systems, markets, strategies, and applications. Close the patent office? I don't think so. Quantenna has a great shot in applying their technologies and strategies as the transition to 802.11ac accelerates - remember, they've already announced a Wave 2 product. Success, then, is far from assured, but the level of competition present in Wi-Fi is a solid indicator of, again, opportunity, with victory going to those who are best at playing the game, not simply those who are the biggest.

Case in point #2 - I spoke with a senior executive of a company that's offering customer-facing wireless services to storefront retailers, but this company eschews Wi-Fi and instead keeps customer handsets on cellular. The argument here centered on whether Wi-Fi would appear in enough handsets anytime soon, so as to obviate the need for femtocells (or similar approaches to provisioning wide-area wireless) in this setting. Beyond the obvious disadvantage that cellular has with indoor location and tracking, I simply had to disagree - it's clear to me that Wi-Fi will indeed appear in virtually every handset over time, and, especially with incentives, everyone is going to be upgrading at some point, and probably sooner rather than later. Wi-Fi-equipped smartphones are available for free (with a contract, of course) from many carriers right now.

My counterparty here noted that he had been waiting 20 years for that kind of market penetration, but my comeback was that it's been only relatively recently that the standards have settled enough to permit the broad (and economical) integration of Wi-Fi into the VLSI in handsets. Given a consolidation in handset platforms (which I believe is essential for reasons of cost; differentiation will be largely in software, not hardware), broad deployment of Wi-Fi in residential, commercial (even if only for use by staff and local applications), and public-access settings, and the need for carriers to provide a suitable offload technology in high-demand, high-population-density settings (like, oddly enough, indoors), Wi-Fi will in fact be in essentially every handset whether the consumer knows it or not. A femtocell is basically a Wi-Fi AP using a different radio technology and different spectrum. Femtocells might be present, but my argument is that Wi-Fi will and indeed must be present. And, of course, there's the issue of a given femtocell supporting  every local carrier, which isn't going to happen anythime soon, if ever. And, of course, Wi-Fi provisions vastly more capacity than cellular ever will. Case closed, at least from my perspective.

I don't mean to sound critical here; argument is essential in reaching at least conclusions, if not the truth. But in both of these cases, my conversation partners were using past performance to try to predict the future, and regardless made good cases for their beliefs. But, as anyone who's an investor in the financial markets knows, past performance is not necessarily (read: never) an indication of future results. Analysts like myself are in the business of predicting the future. Since no one (using currently-available forecasting technologies, anyway!) can do that with anything even approaching always-better-than-even-money accuracy, it's important to illustrate the reasoning behind any forecast, said reasoning ultimately being of far greater long-term value than any specific forecast by itself. I hope this posting has made at least a small contribution towards that end.

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.
Must read: 10 new UI features coming to Windows 10