404 Charts

Web Browsing as an Engagement Metric for Mobile

Screen Shot 2014-02-05 at 8.41.25 PM

Platform and browser share of web browsing statistics get used quite a bit. Yet, I’m not entirely sure this is the best metric, let alone the only metric, that is useful to indicate what has happening on mobile devices. We hear this word engagement thrown around frequently. But what does that really mean in a world where experienced smartphone and tablet owners spread their attention across a multitude of screens and platforms.

During the desktop web era, engagement was easy to gauge. The desktop web was really the only web in town. Even today it is interesting to see the global statistics for time spent engaging on the web favoring the larger screen PC form factor.

Screen Shot 2014-02-06 at 11.36.49 AM

The install base of smartphones has either just passed or is about to pass the install base of PCs (desktops and notebooks) yet the above chart shows that web usage is still dominated by larger screen PC like devices.

In the chart at the top from NetMarketShare it includes both tablet and smartphone operating systems. At a WW level, for mobile devices, iOS easily dominates web usage. Yet, this statistic alone, “engagement” is not limited just to browsing the web. What can not be tracked so easily (publicly) is engagement of a device or platform as a whole. It is likely Apple or Google know something about how iOS and Android are used every day by their customers but that data will never see the light of day.

Consumers browse the web, play games, download apps, watch videos, shop, talk on the phone, text/use messaging apps, use certain apps more than others, etc. I call this share of compute time. What I am interested in is what the share of compute time of a mobile device or tablet actually is with a consumer. Does Android or iOS have the larger share of compute time as a mobile platform? That is the ultimate question. We can circle around this data with app store statistics, engagement data from certain apps like Facebook, WeChat, WhatsApp, and mobile shopping data, etc., but a firm data point will be harder to come by.

That being said, the metric of web browsing will be an interesting one to watch. On my podcast with Benedict Evans he has said multiple times that we don’t even know it will be like going to a website in five years time. How we engage and browse the Interwebs could be entirely different in 5 years. We simply don’t know.

What we do know is that people will have a small computer in their pocket. It will be with them at all times. They will want to discover, pay, learn, play, be assisted, communicate, and any number of things with this device. The platforms that help consumers get the most out of these products will be the ones that win.


Screen Shot 2014-02-06 at 6.27.48 AM

This above chart is a look back as well as a look ahead. While forecasting is an imperfect science, and updates to forecasts are required annually, there is more market intelligence behind these numbers than meets the eye. Yet as you look at the above chart many questions need to be asked. Here are a few:

  1. What is the breakdown of screen sizes as a part of the smartphone forecast?
  2. What are the ASP brands as a percentage of smartphone sales over the time period?
  3. What are the sales breakdowns of PCs from consumer versus commercial?
  4. What is the mix of PC forecasts for both desktops and notebooks over this period?
  5. What is the break down of tablet screen sizes as a percent of the tablet forecast?

Those are just a few to start. When looking at forecasts, I like to think of each category as an onion. Unless the onion is peeled back and each layer further examined then the data is somewhat limited in its usefulness.

Looking at the above forecasts do not highlight the underlying trends in the market place. This is why, as an example, I broke out 9″ tablets as its own forecast line. Of course this form factor is included in the overall tablet forecasts but looking at it broken out brings with it more important observations.

This form factor, for example, begins to impede on the area where the consumer PC was once the only player in town. People are not replacing their PCs with small tablets, although small tablets are certainly many people’s first bigger screen computing device in emerging markets. The 9″ and greater screen size tablet will impact the future of the traditional PC more than the smaller tablets will. This is why I broke it out separately in that chart. There is still more to the story but looking at tablet forecasts within their respective screen size volumes and price bands are all key indicators of what is happening in the market.

Peeling back the onion in my forecasts it is likely 9″ tablets will outsell the notebook(clamshell) form factor by end of 2017 but highly likely it will happen by 2018.

This peeling back of the onion can’t stop there. I still have work to do on the tablet onion, which is my current focus. What are the price bands of tablets as a percentage of current and future market sales? What percent will be iOS, Android, Windows, and AOSP (Android open source), or something else? What are the major segments of tablets? Will it splinter beyond tablet PCs, kids tablets, media tablets, etc? Should we count the tablets that mount on walls at retail, or are used as menus at restaurants, or as interactive displays in museums and other public spaces as internet of things devices or as tablets? These are all things that require further research. I find as I go deeper on each of these things new questions arise that require even more research.

I know folks in the media like to bash forecasts. As I stated, it is an imperfect science. Yet many companies in the industry need and want to hear some sound logic and market research to help them shape their strategy going forward and justify investments currently being made for future products in certain markets. Although it is imperfect it is worth it to try and sort out.