3 ms·
Android Economics
- justincormack 15y agoGoogle have an incentive to downplay revenues for this purpose, so I would imagine that everything possible is removed.
- VengefulCynic 15y agoWhile they certainly have incentive to downplay revenue for the purposes of paying Oracle, let's not forget that all of the contributing revenue factors would be legally discoverable. I think that we can assume that the numbers contributed in court reflect the revenue fairly accurately - Fortune 500 companies tend to have lawyers whose degrees don't come in specially-marked boxes of cereal. Also, Google has a lot of very good incentives to state their revenue exactly as is legally relevant (albeit, not a penny more, in this case).
- huggyface 15y agoI don't believe they need to downplay at all -- in any rational analysis there is no reason they would include, for instance, someone clicking on an ad on a website. It is not reasonable to call that "Android revenue", just as it isn't "iOS revenue" if someone clicked on an ad on The Verge in Safari.
- johncoogan 15y agoThis is a great analysis. I would be very interested to see the standard deviation of Android revenues/device/yr compared with Facebook ARPU. Obviously this is laughably unobtainable, but it's fun to speculate about who has more upside compared to the mean. I wonder what the 95th percentile generates? On a side note, has anyone thought about what Mail.ru might do before and after the Facebook IPO? Sorry if that's off topic.
- Drbble 15y ago> I just don’t know how to adjust this value so by leaving it at the average, I can avoid introducing more assumptions. This is a fallacy of assuming the midpoint equals the centroid. The correct way to handle absence of key information is to widen the error interval, not to assign zero weight to unknown factors. Either those factors have some partially known weight limit and you can estimate a range for the average, or those factors have completely unknown weight and you cannot make any rational estimate of the average.
- siglesias 15y agoIn business estimates, this is how it's done. You leave the reader to construct the error margins by clearly delineating the assumptions. The magnitude of the error bars would be left to his judgment anyway.
- huggyface 15y agoNo one has provided any evidence at all that the $544 million figure (itself essentially invented) is in any way a subset of the $2.5 billion "mobile run rate". From a purely rational perspective, there is no reasonable reason why Google needs to include in-game advertisements, or web ad revenue, in Android revenue. Does Microsoft include Google search revenue in Windows revenue? Does a purchase of a hat in Team Fortress count as Windows revenue? I suspect -- with no more evidence than anyone else -- that the $544 million is but the sum of partner payments for the Google applications or accelerated access, added with Google's cut of the Nexus line revenues. Advertising, though....Android itself has zero advertisements. I don't see ads in Google Maps, gmail, or in any system service or function.
- siglesias 15y agoI think what's salient is where that number is relative to Apple and Google's revenues per iOS device, which also don't include in-game advertisements or ad revenue. Edit: Miscomprehended the derivation, see reply below
- huggyface 15y agoThose derived values are based upon the likely erroneous assumption that I just detailed. How does Google make any money on iOS devices if you excluded advertisements? Quite the opposite -- they pay Apple to be the default search engine.
- siglesias 15y agoI see the point you are trying to make now, but I'd like to offer two points about why this revenue derived from the lawsuit might include ad revenue estimates: 1) I would imagine that regardless of whether it is in Google's interest to count ad revenues in their public reports of profitability, it is most certainly their internal metric for profitability. They simply must have some metric for deciding if the Android investment is effort is worth pursuing, or if certain Android strategies can be considered successful. Would Rubin's declaration that Android is profitable be true if he were only counting early access fees, and Nexsus sales, etc? I don't have the numbers, but I would doubt it. 2) The 3 to 1 profitability of iOS against Android seems to make some sense if you consider that iOS, as of October, commanded over 3x the mobile traffic share of Android. (see http://bit.ly/tHlYe6 http://bit.ly/tHlYe6 )