(9/3/2011: Publishing this 4 years after I wrote it)
Man, people sure are dumb.
Back in January, a market research company called iSupply published an estimate of the bill of materials (BOM) cost of the iPhone, based on the publicly available specifications. Their analysis suggested that Apple was selling the iPhone for 50% more than the cost of the parts in side. After the iPhone shipped, iSupply followed up on their analysis by doing a teardown of an actual iPhone and creating an estimate based on the actual components they found inside. Their new estimate was quite close to their original estimate. At about the same time, Business week published a story based on a very similar estimate from another firm.
These numbers have all sorts of people bent out of shape. There are those who resent the fact that Apple seems to be making a fat lot of cash on the product they are lusting for. Then there are those who seem to feel a need to apologize for just how much cash apple seems to be making by explaining that those estimates are first, only estimates, and two, don’t take into account important costs, like engineering, marketing.
John Gruber bitches that iSupply’s estimate doesn’t take into account things like “Build quality, packaging, shipping,” and also that iSupply includes an estimate of the cost of the operating system @ $7 a unit.
Charles Arthur of the Guardian also has a few beefs with the iPhone teardown analysis, complaining that it leaves out both the hard work and some intangible magic that went into creating the iPhone in the first place.
John & Arthur, here is a clue. iSuppli knows more about the manufacturing costs of cell phones and other electronic goods than you and I do. They have a lot of data at their command for estimating the costs of hardware, software, manufacturing, packaging etc, all of which are part of their report. They sweat the fine print too — the original published estimate includes a $0.40 line item. I would guess their estimate of the OS cost is based on the typical license cost for a “smartphone.”
Are these estimates perfect, HELL NO, which is why they are called estimates. John and Arthur are right to make the point that too many people are accepting them as fact, but they are the best chance anyone outside of Apple & its manufacturing partner have at coming up with a basis for comparison with other similar products (Yes yes, the iPhone is in a class by itself, but face it, people are comparing it to the Treo or the Q).
Does apple sell the iPhone version of OSX, again, hell no, but the makers of other phones are paying license fees for the OS they are using.
Is the iPhone experience a step above that of smartphones? Again, hell yes, but that isn’t really the point of a BOM exercise. The R&D to make the iPhone as cool as it is is what is called a fixed cost. The R&D cost of the iPhone doesn’t change based on volume. It’s money Apple had to spend, whether they end up selling 10 iphones, or 10 million.
The BOM costs are a major part of the variable costs of the iPhone. Each iPhone Apple sells increments the variable cost to the company. The BOM isn’t the only variable cost for the iPhone, there are support costs, etc. These costs can be sub