Danny Sullivan argues that the conclusions about Android revenues based on the Oracle v. Google court document are flawed. That’s certainly possible, but it doesn’t sound as flawed as Sullivan seems to believe.
Google says the following in its proposal:
Specifically, in the event of a finding of patent infringement of the ’104 patent, Google is willing to stipulate that un-adjusted damages for the ’104 patent through 2011 are $2.72 million, and in the event of a finding of patent infringement of the ’520 patent, that un-adjusted damages for the ’520 patent through 2011 are $0.08 million.
They then go on to say:
Similarly, in the event of a finding of patent infringement of the ’104 patent, Google is willing to stipulate that future damages for the ’104 patent should be 0.5% of Android revenues through the expiration of that patent on December 22, 2012, and in the event of a finding of patent infringement of the ’520 patent, that future damages for the ’520 patent should be 0.015% of Android revenues through the expiration of that patent on April 7, 2018.
Just looking at the ‘104 patent, if Google says the damages (if they’re found guilty of patent infringement, which they haven’t been yet) from 2008 (when Android launched) to 2011 are worth $2.72 million and then says going forward, the damages should be 0.5% of Android revenues, that implies that $2.72 million is equal to 0.5% of Android revenues from 2008 to 2011. From that point, it’s simple math.
That 2.8 million figure isn’t said to be based off of Android revenues. My understanding is that those figures are based off what a court-appointed expert believes the patents are worth.
But the keyword in my mind from the document is “Similarly”. It’s true that Google doesn’t say the $2.72 million number is related to revenues (though it’s not entirely clear how else you would quantify the “damages”). But they then dive right into explaining that they should pay 0.5% of Android revenues for further damages and lead with “similarly”.
Maybe it’s a poor choice of wording, or maybe it was a inadvertent tip of their hand. Google doesn’t like to break out Android revenue numbers specifically (hence, this whole debate) and it’s usually the case that companies don’t do that because they’re not very impressive (at least compared to rivals).
Further, why would Google come up with a future rate based on revenues and base the past rate off of something completely unrelated and as such, seemingly arbitrary?
In saying that the $550 may be “hugely inaccurate”, Sullivan notes:
Maybe a key source is advertising revenue. But is that advertising revenue based on what appears within apps? Advertising revenue on what appears if people do a search on Google with an Android phone and click on an ad? What about if they search on Android with some other search engine and select an ad? Is that revenue that’s counted toward Android?
That’s silly. Of course the majority of the money Google is making from Android comes from advertising revenue. They don’t charge to license out the software. And paid app sales (for which they get a 30% cut) are widely documented to be ho-hum. Even in Apple’s more robust App Store paid app ecosystem, the 30% cut doesn’t equate to all that much money.
From a revenue perspective, Google is an advertising company. That’s true on desktop and that’s true on mobile.
Further, Sullivan’s thought about people using other search engines on their Android devices is just strange. It’s a fringe scenario at best. And still, Google would know their own numbers. Why/how would they be tracking rivals’ click-through numbers? Rev share that Yahoo and/or Microsoft are paying them? Again, that’s fringe at best.
Sullivan’s entire argument doesn’t really back up his title: “No, Google Doesn’t Make Four Times More Off The iPhone Vs. Android”. He admits a few times that no one actually knows outside of Google. So maybe Google does make four times the money off of the iPhone versus Android. Maybe it’s five times.
Google clearly doesn’t like this story line. Too bad. Shut everyone up by releasing the actual numbers or don’t and deal with it.
In simply reading the court document, Google is saying that they should pay 0.5% of Android revenues to Oracle if they lose the suit. And that the number they should pay for previous years is $2.72 million. Why would they think they should pay less in the past than in the future (as Sullivan is implying)? Why wouldn’t these numbers be directly related? And if they’re not, why call the two numbers “similar”? I’ll go with common sense on this one.