Kindle Sales Theory is Flawed
Monday, August 10th, 2009
In February, a Kindle sales theory was proposed by Citi Investment Research (PDF) using Sprint activation numbers.
Thanks to CIR Telco Analyst Mike Rollins, we have uncovered a key disclosure in Sprint’s September Quarter 10Q filing. P. 42 in the Wholesale, Affiliate and Other Revenue sector. Here’s the text:
“Certain wholesale devices are activated on the network by our wholesale partners prior to selling the device to the end customer, which resulted in approximately 210,000 such additions being activated on our network during the third quarter 2008.”
Additional sleuthing on Mike Rollins’ part suggests that there could have been 100,000 wholesale device activations in each Q1 and Q2 of ’08, and our combined view is that these wholesale device activations refer specifically to the Kindle. Tie these points together with the knowledge that Amazon fully sold out its Kindle supply by mid-November, and it’s hard to escape the conclusion that Amazon sold approximately 500,000 Kindles in 2008.
I just couldn’t understand how one would think that all 210K were specifically Kindle related. I wrote to Sprint Investor Relations asking a very simple question.
Can you confirm that the wholesale device activations referenced are solely Kindle?
The other day, after a bit of persistence, I was able to get the following response from Sprint Investor Relations.
We cannot offer much additional commentary on this topic. I can tell you that there are other data centric devices included in our reported wholesale and affiliate sub count than the Amazon Kindle. Neither Amazon nor Sprint provides details on the number of kindle devices. The comment below about the 210k devices at the end of 3Q08 references those that were activated but not yet in the hands of an end user.
This clearly states that the activations do not specifically relate to Kindle. However, it doesn’t provide any insight as to what percentage may be Kindle related.
The later comment regarding the device activations is odd. My question did not address this in any way, shape or form. Instead, this is a … spontaneous additional comment.
Perhaps it is a simple clarification but in some ways it feels like activations are divorced from the sales cycle. That the activations could be a supply number (for all devices) instead of being mapped to sales. Please share your thoughts here since I find the subject murky.
However, in light of this data, I believe the 500,000 figure is generous. I don’t blame Citi for trying to construct a sales volume theory. It is Amazon and Jeff Bezos who are to blame. Speaking to shareholders at the Seattle Art Museum, Bezos had this to say about Kindle sales.
I’m not sure we will ever reveal all the numbers. Our point of view is that there is a competitive advantage to keeping the numbers close. You may have to remain curious on that point and I certainly understand the curiosity, especially since I look at the numbers so avidly every morning.
I don’t find this a compelling argument and can only surmise that sales are weak. Otherwise, the impact Kindle sales would have on the business would be large enough to meet the materiality definition, and thereby be necessary to any shareholder report.
I understand that Amazon is in a fight with Google for the future of digital books and they want to play their hand like they’re holding pocket aces. But the player who consistently over bets is usually bluffing.








Eric Engleman at TechFlash recently sat down with Ian Freed, VP for Kindle. The 
Oprah endorses Kindle. OMG! Stop the presses. OMG! This is huge. OMG!
