Ever since Steve Jobs unveiled the iPad 2, people have busy trying to compare his statements to the truth, especially this one: “Many have said (iPad) is the most successful consumer product ever launched. Over 90% market share and our competitors were flummoxed.”
Where was Steve getting this 90% market share data point? Strategy Analytics showed iPad’s market share at 75% in 4Q10 and falling fast.
The thing about Steve Jobs is that he rarely outright lies, instead opting to look at data in a way that he thinks makes most sense and which contains some shred of validity. I don’t think Steve was far off from the truth saying iPad had 90% market share. Using conservative figures and assumptions, I calculate iPad’s tablet market share at 90% in 4Q10, and nearly 95% for 2010.
However, in coming quarters, iPad’s market share will fall, but not for the reasons you might think.
The main problem with market share data is that there is no easy way of measuring how many tablets are purchased by consumers. Instead, market research firms rely on company figures (i.e. Apple earnings reports, Samsung press releases) and other estimates to reach unit numbers that are better described as “shipped” rather than “sold”. There is nothing wrong with this procedure as long as it is clearly labelled and, more importantly, the accompanying attention-grabbing headlines indicate this terminology. Instead, bloggers and reporters jump to conclusions that are often misguided and misleading.
So why is it OK that companies report units shipped as units sold? It all comes down to accounting.
Companies need to determine inventory and cost of good sold figures in order to calculate earnings. Sounds simple enough. Diving deeper into purchasing contracts would show the more intricate interactions between a buyer and seller. Without jumping into the accounting bunny hole, let’s look at Apple’s most recent 10-K:
“(Apple) recognises revenue when persuasive evidence of an arrangement exists, delivery has occurred, the sales price is fixed or determinable, and collection is probable. Product is considered delivered to the customer once it has been shipped and title and risk of loss have been transferred. For most of (Apple)’s product sales, these criteria are met at the time the product is shipped. For online sales to individuals, for some sales to education customers in the U.S., and for certain other sales, (Apple) defers revenue until the customer receives the product because (Apple) legally retains a portion of the risk of loss on these sales during transit.”
An iPad on a freight plane headed to a Walmart warehouse is no longer counted as an iPad in Apple’s inventory, instead it is counted as an iPad in Walmart’s inventory. Apple is able to recognise that iPad as sold and recognise the accompanying revenue (and profit).
So how should one account for market share data with this shipped/sold methodology in mind? I support the idea that proper market share data should make an attempt of calculating how many units have been sold to consumers. There is no exact science to this, and to a certain degree, even Apple and Samsung may not know day to day how much of its product is actually sold (although this process is becoming more efficient thanks to advancements such as Walmart’s lean inventory practices which notifies Apple soon after you buy iPad). Given the tablet market’s young age and small sales figures, we can reach market share according to units sold to consumers with a large degree of confidence.
Let’s start with the most recent market share data by Strategy Analytics published on January 31, 2011:
Global Tablet 4Q10 OS Shipments:
iPad: 7.3 million
Android: 2.1 million
Others: 0.3 million
According to these figures, iPad’s market share fell to 75% in 4Q10. Let us dive deeper into these numbers.
1) Since Strategy Analytics is using shipped units, I want to turn these numbers into units sold. To do this, I need to estimate the number of tablets that are currently in a company’s distribution channel, (the location between Apple’s factories and your living room). Apple’s iPad distribution channel is easy to calculate, since Apple’s CFO Peter Oppenheimer gave it to us on Apple’s most recent earnings call. iPad channel inventory was up 525,000 iPads during the quarter with a total of 4-6 weeks of sales in inventory (roughly 2.5-3 million iPads). Apple had already filled its distribution channel two quarters ago and the 525,000 additional iPads were put in the channel to fill additional points of sale. iPad was available in 46 countries with an additional 15 countries selling iPad in the beginning of 2011. To recap, Apple shipped 7.3 million iPads in the quarter, which includes stuffing 525,000 iPads into the already filled channel, leaving 6.7 million iPads actually sold to consumers (iPads shipped – iPads put in channel inventory = iPads sold). To be ultra conservative, I will assume they had to fill their channel more than they said, leaving approximately 6.4 million iPads bought by consumers during the most recent quarter.
2) Let us do the same calculation for Android sales: 2.1 million android tablets were shipped during 4Q10. How big is that distribution channel? Taking a look around the marketplace, there really wasn’t much in the way of Android tablets besides Samsung’s Galaxy Tab and we know from reports that the Galaxy Tab was available in 94 countries and 200 different wireless carriers. We also know that the Galaxy Tab was introduced close to the end of 3Q10, and market share data shows only 0.1 million android tablets were shipped during 3Q10. Ladies and gentlemen, this is called a channel stuff. Android’s 2.1 million shipped figure for 4Q10 was primarily Samsung filling its extensive channel (which I estimate to be at least 1.5 million Galaxy Tabs). Available in twice as many countries as Apple, I assume that the channel is nearly 50% the size of iPad’s channel (for the sake of being conservative). Samsung’s Galaxy Tab channel has been relatively small in the U.S., but in Asia and Europe, the Galaxy Tab’s retail reach has been much more extensive. Backing out the channel inventory, leaves me with around 0.6 million Android tablets sold during 4Q10.
3) Tablets marked as “Other” I will largely ignore since that is a non-factor considering a typical freight plane will hold a few thousand tablets at a time. I consider the 0.3 million tablets as largely channel fill and random promotions, so I will include 0.1 million “other” tablets sold during 4Q10.
Revised market share for tablets that were actually sold during 4Q10:
iPad: 6.4 million
Android: 0.6 million
Others: 0.1 million
Revised iPad market share in 4Q10: 90%. For 2010, using the same procedure, I calculate that Apple sold approximately 12 million iPads, with Android selling 0.7 million tablets and 0.1 million “other” tablets sold to consumers, giving iPad approximately 95% market share for 2010. Steve Jobs wasn’t far off with his 90% iPad market share statement.
With all of this shipped vs. sold terminology in mind, it is important to think about how this will impact iPad’s market share going forward. With tablets from RIMM and HP shipping sometime in the near future, and additional Android tablets like the Motorola Xoom, it is obvious that there will be more tablets shipped in 2011 than in 2010. Running back of the envelope numbers: if Apple can ship 20 million iPads (conservative) in 2011, RIMM ships 3 million, HP ships 2 million, and Motorola ships 2 million, iPad’s market share is now down to 74% and falling fast.
However, if we take a look at units sold, I am confident that iPad will be doing much better.
It will be interesting to keep an eye on how actual sell-through turns out for non-iPads. The tricky part in this whole market share discussion is what happens to tablets that don’t sell? Will price reductions beef up sales? What if non-iPads are literally given away? Does every non-iPad that gets manufactured eventually finds its way into the hands of a consumer? Why does all of this matter? One word. Developers.
A tablet stuck in a distribution channel or sitting on the shelve at Walmart will not lead to many sales of a developer’s application. A stack of tablets given out for free because no one wanted to pay for it doesn’t exactly sound like prime ground to stake app development dollars, and we all know that software (and the accompanying third-party app ecosystem) will ultimately decide long-term tablet market share.
Use caution when reading the dozens of upcoming tablet market share reports in 2011 or you will misunderstand what is actually happening in the marketplace.