Apple Failure Modes
Apple has avoided the types of failures that have beset so many tech giants. From the HP I dearly loved and the IBM we once feared, to Palm, Nokia, Blackberry, and many more… Will Apple eventually follow a similar trajectory and either disappear or recede into the shadows? Or can Tim Cook continue to keep the Steve Jobs Apple 2.0 miracle alive almost a decade after the magician’s passing?
The Monday Note has been on an irregular hiatus as I labor on a book chronicling my picaresque half century in the tech world. While I only spent ten of those years inside Apple, gravity exerts its pull and the book sometimes feels centered on the company that allowed me to fulfill two dreams: Coming to the US and leading a product engineering organization.
Writing about the early days at Apple led me to contemplate how the ambitious but struggling company became today’s $2T enterprise, how it avoided the “failure formulas” we’ve seen in so many grandees of the industry.
Nokia, Palm, and Blackberry followed a relatively simple failure recipe. When the first generation iPhone was announced, they dismissed the threat, impugning Apple’s ability to play in their arena. Then Android devices arrived, and the giants refused to back down: ’We know what we’re doing, just look at our numbers!’.
My good old HP is a much more complicated story. On the technical side, it allowed its superb desktop computing business to be disrupted by “cheap” 8-bit processors, but the real problems were cultural and political: A revolving door in the CEO suite, a Board of Directors that spied on each other, no coherent corporate strategy leading to catastrophic acquisitions followed by spinoffs…
No company has been as powerful and then fallen as far as IBM. Once known as The Company, its mainframe products and services dominated business computing, its management methods were exemplary. (In the mid-seventies I was given a copy of the all-encompassing Manager’s Guide and was in awe with the depth and scope of the work.) Then, the PC happened, a product category IBM initially seized, only to lose it by letting clones powered by Microsoft software flood the market and kill its margins.
A decade later when the Internet and networked servers changed the game, IBM wasn’t ready and almost went bust, only to be saved by Lou Gerstner…at least for a while. Unfortunately, Gerstner’s successors were unable to harness the relentless growth of Cloud Computing, and now the company has fractured. The current CEO, Arvind Krishna, recently decided to split IBM into “Two Market-Leading Companies with Focused Strategies”. The larger entity keeps the IBM name, the smaller as yet unnamed company rids IBM of a low-margin, low hope, ferociously competitive IT infrastructure business.
Microsoft offers an interesting counterexample of success after it made an historic, expensive miss. Late to the smartphone game, the company gave Nokia special licensing terms for its Windows Phone OS, only to see the partnership flounder. Despairing, Microsoft bought Nokia for $7.2B in 2013 and took a $7.6B writeoff two years later, followed by another $900M the following year. The clean-up job was left to Satya Nadella who took the reins from Steve Ballmer in 2014. Since then, Microsoft has prospered as the company has focused on software and Cloud services for organizations. As a part of that refocus the Microsoft stores, modeled after the Apple Store, have been shuttered.
While these failure stories hold some lessons for Apple, some of them are actually reassuring.
For example, it takes more than one substantial mistake for a large company to begin its decline. The Apple Maps debut and “Antennagate”, as examples, were embarrassing but didn’t do any lasting harm. To be sure, two mediocre iPhone vintages in succession would have a deleterious effect on image and finances, but even that could be survived, especially in today’s quasi-saturated market. And as the Microsoft example shows us, seriously missing an industry wave (smartphones) can be overcome by jumping on a new one (the Cloud aided by the Windows/Office flywheel). This may shed light on Apple’s efforts to give more momentum to the Services business, a flywheel in its own right.
Apple’s iCloud is a different story. True, “cloud” is a very broad term and many of the company’s cloud services are so taken-for-granted as to be almost invisible. For example, iPhone photos live in the petabytes or exabytes of cloud storage that propagates nicely to users’ devices. The same is true for Music and more.
While iCloud as a product has come a long way since the 2008 MobileMe, the Exchange For The Rest Of Us that embarrassed Steve Jobs, it’s often sluggish and buggy (even now as I attempt to use Pages “as we speak”). It lacks the power and polish that Google and Dropbox have to offer. That said, one shouldn’t expect Apple to offer iCloud services in the way that Amazon Web Services, Google Cloud, and Microsoft Azure do. In fact, Apple in part depends on AWS and others for its own infrastructure — a contentious internal topic.
Apple’s record with Artificial Intelligence (another broad domain) is surely a sore point in the Board Room. Although the company was “there” first with Siri, the company watched as Google and Amazon surpassed them to become the leaders in Intelligent Assistant applications. In everyday life, one can see modest progress in Siri’s usefulness and pervasiveness, and we can hope Senior VP of Machine Learning and AI Strategy John Giannandrea, a Google alumnus with a distinguished résumé who joined Apple in 2018, will set things right.
Apple’s strengths are not to be discounted when considering failure modes. Its hardware, software, and supply chain management is unrivaled. But let’s focus on a less lauded advantage, the power of its organizational structure.
To simplify, there are no divisions at Apple, no iPhone, Mac, or AirPod “subcompany”. Instead, there are functions as sketched by the Apple Leadership chart (helpful job details are accessed when clicking on the names)
When Apple develops a new product — I’ll avoid titillating possibilities — work is organized around projects. A project group is formed by drawing on functions such as Software Engineering, Operations, Hardware Technologies, and so on. Some team members, for activities such as Product Design or Operations, may work on more than one project. The group exists as long as the project exists and is disbanded if the product is canceled or put on the shelf.
One of the things that beset HP was its divisional structure with the unavoidable rivalries, territorial disputes, and fights over resources. Customers, of course, don’t care about divisons, they care about products. Apple’s robust, flexible, functional organization helps everyone focus on products and customers.
It’s an extremely valuable Steve Jobs legacy.
Does this mean Apple is immune to large scale failure, that it won’t someday take the path HP or IBM did?
In a quest for the next engine of growth, Apple could take big risks such as trying to enter the auto industry, either in a frontal assault against Tesla, Toyota, and “Deutsche AG” (German car makers), or in more original forms of individual mobility. Or it could be tempted by the humongous amounts of money spent on healthcare.
And no matter how powerful its organizational structure is, Apple, like every company, is susceptible to personal mediocrity: Insecure B-grade managers hire C-grade players who won’t challenge their authority or their “expertise”, and products suffer as a result. We know the old organization joke: When upper layer people look down, they see brains; when brains in the lower layers look up, they see #$$holes. For an organization, the beginning of the end comes when the brains realize the upper layers are colonized by incompetents and get into Why Bother Mode. I don’t know enough about the company’s hiring and firing practices but, in my nervous mind, this is the biggest risk to Apple. From a distance, it’s impossible to know how hard Apple works to avoid a form of degenerative failure.
- The whole point of certification is that it independently and impartially verifies that you are complying to a standard.You can find a problem together with the