A request of our research analyst partners: Ditch the feature phone – smart phone segmentation!

In my March blog ("Getting Smarter About Feature Phones") I noted varying usage rates between smart phone and feature phone owners, with survey data suggesting that 10 – 20% of feature phone owners download applications versus 40 – 60% of smart phone owners. However, I also noted that this was misleading with certain feature phones, like INQ’s “Facebook phone,” achieving usage that compares admirably to leading smart phones.

At this point, the presence of a user visible HLOS, like Android or Symbian, serves as a proxy for the level of effort that OEMs and operators put into developing devices optimized for data use, with broad application ecosystems, ease of content download, and all-you-can-eat data plans. However, with the emergence of new platforms such as Samsung’s Bada and Qualcomm’s own Brew MP that may not conform to traditional definitions, it becomes much harder to define a “smartphone.” Moreover, as functionally-optimized devices come to market addressing specific user behaviors, like Microsoft’s Kin for social networking, it becomes challenging to fit them into broader categories.
Stock definitions of “feature” versus “smart” appear increasingly useless when it comes to building a strong understanding of the fundamentals of user behavior. And it is against this backdrop that I am calling our friends in the analyst community to get creative with the lens through which we discuss devices. Below I propose a few possible directions:

Behavioral

Is the device defined by social networking features or optimized for entertainment? Is it multi-purpose, or designed for electronic reading? Recognizing respectfully that this was the original intent of the “feature phone” designation it has become a catch-all for everything that isn’t voice-only.

Functional

It is clear from user behavior that certain feature combinations take usage to another level. Screen size, touch and a full web browser all appear incredibly important. Perhaps it is time to start classifying devices more holistically? For example high- and low-end web access devices, sensing devices and quick messaging devices (thanks, AT&T!).

Data Plan

Qualcomm’s own research shows that once so-called “feature phone” owners adopt data services, their behavioral profile is comparable to “smartphone” users, albeit less frequent. This suggests these devices face specific external constraints including the commercial terms under which they are offered. Further modeling of the device base addressed by different data plans would be a huge step forward.

In writing this, I recognize that an understanding of available device platforms is critical, not least to the developer community. We should absolutely be tracking and monitoring all of the operating systems in market today, including emerging alternatives. It’s also vitally important that we continue to understand other platforms. What about Java and Flash? What about browsers?

However, what I am calling for is a smarter and more effective segmentation of the mobile device market that speaks to what users really do with their devices. If our partners in the analyst community can address this, I’m willing to bet that it’s not just Qualcomm that will be willing to listen.