I get to sit in on a lot of analyst meetings and the increasing theme of many of them is that the market is moving back towards vertical integration for PCs and smartphones where the entire solution is controlled by one vendor.
In fact, today we were just talking about the fact that the cell phone market may eventually consist of Apple, Google, Microsoft, and RIM phones and that it won’t matter who builds the hardware.
What prompted the discussion about AMD was a storytalking about how Apple was once again looking at the company to supply alternative solutions to the ones they are currently buying from Intel. Based on what I’m hearing this might actually happen this time. And it could then lead to a future decision where Apple bought the company if Apple’s experience was a good one.
Let’s talk about the move to vertical integration in the PC and device space and what it means to you.
Vertical Integration: The Apple Impact
If you look at Microsoft, Google, and Apple closely you’ll see that they all are slowly vertically integrating. Apple is now designing their own processor for the majority of their consumer products and tightly manages and owns their platforms, Google just announced their own chip group and has been building their own data center hardware for some time. Microsoft has their own chip group as well, owns products top to bottom like the Xbox and Kin phone, and very tightly specifies the Windows 7 Phones and all of the PC products sold in Microsoft stores.
I think this is largely the result of Apple’s huge success in creating massive margins, impressive loyalty, and dominating lucrative markets by controlling the customer experience. People have voted with their dollars that when it comes to a choice between freedom/low price and a good customer experience, a significant number will pick the latter and the other vendors are learning this lesson.
This suggests that increasingly the major players (or whoever has the most power) will control the product and own it, but likely not build it.
Power will shift to the vendors who own the solutions along with the related margins if the Apple model holds.
However, this is in some ways a return to the past where interoperability and migration between vendors becomes an increasing problem, but where solutions that are contained within vendors become easier to use and maintain.
Given the increased focus on putting information in the Cloud and the tendency for the browser to be the great equalizer, I doubt things will get as bad as they were in the mainframe era. But we can already see that moving between Apple, Google, and Microsoft smartphone products isn’t as easy as moving between Dell, HP, and Lenovo PCs.
This will likely create some of the same bidding problems that often occur with other closed platforms: migration costs will often offset even reasonably large bidding differences between vendors. The other result is a consolidation of technology companies under the umbrella firms that are putting the solution together.
Apple Buying AMD
One of the biggest potentials would be Apple buying AMD. This clearly wouldn’t have been possible had AMD retained their FABs because the cost and risk to Apple would have been too high.
However, now that Global Foundry and other Foundries are in place to handle this FABless approach to building SOCs, AMD is both relatively inexpensive and reflects a better risk/reward balance.
The reason Apple might want AMD is for Fusion, which will be the first blended high performance GPU/CPU part. Steve Jobs already values the graphics side of his platform more than the number crunching side. This would give him exclusive access to the only blended part in the market and one that particularly favors Apple’s graphics intensive platforms.
While Apple has never been a fan of AMD, they actually liked and used ATI a great deal and the idea of getting out from under Intel is likely appealing to them as well. Apple likes its suppliers to jump to Apple’s commands and Intel is simply too big and powerful to do that to Apple’s satisfaction.
One of AMD’s sustaining advantages over Intel is their willingness to bend to OEM’s needs and this should, assuming Apple starts using their solutions, endear AMD to Apple. But Apple also likes unique advantages and will likely, assuming the initial engagement is successful, drive them to buy AMD to deny this technology to others. It would also allow them to insure their higher priced offerings aren’t easily compared to lower priced offerings from Dell and HP using similar technology.
Some already brand Apple’s solutions as inferiorbecause they can be directly compared today to lower priced but higher performing products from Apple’s competitors. This would help eliminate that risk.
Wrapping Up: Owning You
In the end this is all a fight for who owns you, the customer, and it is nice being chased by powerful companies. However, we’ll all have to be a little more careful as to which products and platforms we pick in the future because of this change.
This is because you don’t want to be owned by a vendor you don’t like working with, given how increasingly difficult it will be to easily change vendors in a vertically integrated world.
I expect this vertical integration to take a number of years and be nearly complete by the end of this decade. But the choices you make today may have a lot to do with the vendor you are tied at the hip to tomorrow. And because it will be very difficult to untie that link, you need to start thinking of these choices both strategically and in the context of where you want to be in the future. Apparently we find ourselves in increasingly exciting times.