Endeavour Partners - Blog

This morning at 4G World in Chicago, Bill Morrow, CEO of Clearwire, described the need for 4G networks to support bandwidth hungry applications on smartphones, and we can’t agree more.

He demonstrated the difference between an iPhone’s performance using AT&T’s 3G network and an iPhone using WiFi to connect to Clearwire’s 4G WiMax network.  The difference was impressive:  Clearwire’s network performed beautifully for both streaming video and Google Earth, while AT&T’s network was sluggish.

Unfortunately, it is exactly this need of 4G smartphones for data capacity that will likely spell the downfall of WiMax.  By our estimates, smartphones will be responsible for roughly 50% of mobile data traffic within the next five years.  But the iPhone is not available with WiMax – and neither are any other mainstream smartphones.

Yesterday Kris Rinne opened the 4G World conference by laying out AT&T’s plan to forego HSPA+ and move directly to LTE, with commercial rollouts beginning in 2011.  This is slightly behind Verizon’s aggressive LTE deployment across 20-30 markets planned for late 2010.  The iPhone, RIM’s BlackBerry, Windows Mobile devices, and a wide variety of Android devices will all be available to run on these LTE networks.

WiMAX is a great technology, but the window of opportunity is short.  Ultimately, the availability of leading smartphones, apps, and content will determine 4G winners and losers.

There’s a discussion on Mobile Innovation about ‘When will Femtocells go Mass Market?’

We think it’s now ‘if’, not when:

1. the market in the developed world is moving to smartphones
2. substantially all smartphones will have WiFi
3. WiFi is already pervasive in the enterprise
4. WiFi is widespread in the home

So, who needs the extra cost and expense of a femtocell, when the WiFi infrastructure is already there? Vodafone is asking £160 (= US$265, €183) for its femtocell, the Vodafone Access Gateway:

Vodafone Access Gateway

A Belkin N150 Enhanced Wireless Router for BT is just £70 ( = US$117, €80) – less than half as much:

Belkin N150 Enhanced Wireless Router

Part of the discussion was triggered by a new report from Juniper Research, which puts the numbers at just 15 million worldwide by 2012. That’s hardly mass market: it’s much less than 1% of subscribers in the developed world. It’s not enough to put any sort of dent in the smartphone surge shortfall.

So for femtocells, it’s becoming if, not when…

One of the big impacts of smartphones has been a resurgence in WiFi as noted by, amongst others GigaOM and ABI; we believe that pretty much every smartphone worthy of the name will incorporate WiFi, to enable a whole bunch of convergence use cases, such as connectivity within the home and office for convergence. This may well extend to public spaces as well; we think that there’s a compelling case to be made for ‘drive by downloads’ to unload the cellular networks laboring under the impact of the smartphone surge.

Our Cambridge (MA) office is at Cambridge Innovation Center, at One Broadway, which is a great space, and the importance of WiFi is driven home by a recent post on the blog:

people are beginning to rely much more on Wi-Fi than they had in the past, with some companies dropping wired connections entirely.  Some cell phones, such as the T-Mobile Blackberries our staff use, can now make and receive all their calls over Wi-Fi

There’s a great discussion of the challenges involved, and the outcome, for which are extremely grateful, of much improved reception and performance, thanks to Ruckus‘ BeamFlex (reviewed here)

After all the testing and analysis, one system stood head and shoulders above the others: the ZoneFlex system from Ruckus Wireless.  I remember in particular one of the graphs comparing the performance of the various systems under heavy load.  The other systems showed a jumble of jagged lines representing drop-outs, while Ruckus sailed smoothly through with none.

Another funny story from this effort was that after we were done, we had come to know some other support people very well, but didn’t have a read on the quality of Ruckus support.  Why?  Because we didn’t have to call them!

There’s a wider theme there that harks back to some of our early work with Virgin Mobile; the best customer service is the one you never need…

iSuppli has just caught up:

The days of supremacy for the Portable Navigation Device in the global navigation market are coming to an end…

“Previously, smart phones were not seen as a threat to the dominance of PNDs…” (our emphasis)

I beg to differ:

Even as PDA’s fade from the technology horizon, another currently hot “P” device — the PND, or “personal navigation device” — may soon follow, says an MIT Sloan School of Management expert on business strategy, technology, and innovation.

PND products, such as Tom Tom and Garmin, may remain popular for a few years, “but life is getting tough very quickly” for PND makers, according to MIT Sloan Senior Lecturer Michael A. M. Davies. PND’s face a growing challenge not only from cell phones, which are increasingly equipped with high-end navigation programs, but from a less obvious competitor: automobile makers.

Davies wonders whether PND makers “will be able to spot a market turning point before it arrives. Right now, things look great for them, but that trend will reverse dramatically,” he explains. “The three-way battle between them, cell phone makers and car vendors is already underway…

One of our recent projects involved taking a very long term look at demand for wireless infrastructure in the United States – going out twenty years. A couple of really interesting things came out of it:

• just how long it takes for technologies to move from the lab to having a significant market impact – something I’ll cover in another post
• the balance between demand and supply for network capacity over the next few years

On the one hand, there’s a surge in smartphones, and smartphones that are powerful and easy to use drive significantly more data traffic. On the other hand, wireless network operators will soon be able to make use of the new spectrum that they have acquired in one form or another, and they will soon be able to start the transition to Long Term Evolution (LTE).

How do these play out?

• how much increased demand does a smartphone put on a cellular network?
• how much does new spectrum help?
• what is the payoff from deploying LTE?
• what does the relative timing look like?

For the first question, we are indebted to one of our specialist partners, who we worked with to cut their data so that we could figure out the differential load of diverse devices.

For the second and third questions, spectrum and LTE, we built a simple model that estimated the overall impact, taking into account the different circumstances of each of the major US network operators.

Although building out the network infrastructure is both challenging and capital intensive, it’s not enough, however. Getting the payoff also requires getting devices into the hands of customers; this is what constrains the benefits from more spectrum and new technology.

Smartphone surge and shortfall

This picture shows in schematic form the smartphone surge, and the resulting shortfall in network capacity. As smartphones become widespread, they’re going to drive a nearly tenfold increase in demand. The payoff from new spectrum and LTE will be significant, but not enough to absorb all of this demand; over the next several years growth in demand outstrips growth in supply, and they don’t come back into balance until sometime around 2015 or beyond.

Now this analysis focused on smartphones; it doesn’t take into account a couple of other things:

• laptops with data cards – not as numerous but can drive several times as much demand – or devices like Verizon’s incredibly cool MiFi which provide the same capability (and which I’m using to post this from here)
• other mobile devices or appliances, such as Amazon’s Kindle™, where we may see a lot of innovation through programs such as Verizon’s

These may also have a significant impact, depending on just how many people adopt them, and how their usage patterns evolve. But that’s a topic for another day…