tecosystems

Free Hardware and The Rise of Android

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

Yesterday on Twitter, I asked whether anyone could identify a particular point on a basic time series chart plotting of Google’s publicly disclosed Android daily activation figures. No one provided the answer I had in mind, though Corey Gilmore came close. In the field’s defense, they had little chance of success initially because the first chart provided applied the datapoint to the wrong point on the timeline, meaning that they were scrutinizing the wrong month for clues (mea culpa, guys). Here’s the chart, which plots the number of daily Android activations by time:

Guesses ranged from Android 1.5 to the Palm Pre release to the availability of the first non-G1 device, all of which were close. Corey’s answer was Google I/O 2009, which was in fact the right occasion, but not the precise event I was looking for. That was this:

At the conclusion of their first major conference, I/O, in 2009, Google announced that each attendee would receive a free HTC manufactured G2 Android phone. What I became curious about when examining the data last night was what the practical impact of this was on Android usage and penetration. While there’s been much debate and hand wringing over the long term implications vis a vis expectations, the important question to me is: was this worth it?

The chart above can’t definitively answer that, though the elapsed time between the G2 issue and activation growth is plausible given the realistic latency between device acquisition and actual application generation. Google, fortunately, has provided us with harder evidence. A year later at I/O, they issued high end HTC Evo handsets. Last year? Samsung 10″ Android tablets, a Chromebook and a Verizon LTE mifi. The unsustainable nature of this trajectory aside, Google clearly believes in device seeding as a strategy. Besides I/O, they’ve handed out Nexus One’s at everything from Linux Foundation events to OSCON. Clearly they perceive some benefit to the practice.

Obviously it helps with ticket sales. Likely as a byproduct of the free hardware, I/O sold out in 59 minutes last year. But that’s not likely to be the justification; Google could reasonably expect to sell out the conference giving away nothing more than the traditional, useless conference backpack. And while the marketing spin to giving G2’s away was as a “thank you” to the attending developers, there are cheaper ways to convey that.

The primary motivation instead is likely developer recruitment. In the wake of its dramatic ascent it’s easy to forget, but at the 2009 I/O show Android was an also ran, an afterthought next to the dominant iPhone. Product only gets you so far, as Google understood then, Microsoft presumably understands today and HP understood too late. Success in the mobile world is increasingly a function of the ecosystems you can sustain around your platform, which means that developer recruitment is what will determine success from failure. And as developers generally advantage platforms they use, what better way to make sure they’re using Android than to hand it to them, for free?

The available data above doesn’t prove that Google’s “Oprah Moment” was the catalyst that sent Android to the moon. Even if that chart went hockey stick the month after I/O, it’d be necessary to point out that correlation does not prove causation. But a year later Google was lighting up a hundred thousand devices a day, and based on the current trajectory they’ll eclipse a million well before I/O this year.

While we can’t quantify the impact that hardware giveaways have had in that growth without more data, then, the evidence suggests that the role was substantial.

One comment

  1. […] Free Hardware and The Rise of Android Yesterday on Twitter, I asked whether anyone could identify a particular point on a basic time series chart plotting of Google’s publicly disclosed Android daily activation figures. No one provided the answer I had in mind, though Corey Gilmore came close. In the field’s defense, they had little chance of success initially because the first chart provided applied the datapoint to the wrong point on the timeline, meaning that they were scrutinizing the wrong month for clues (mea culpa, guys). Here’s the chart, which plots the number of daily Android activations by time: […]

Leave a Reply

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