I'm writing a CSS book

JavaScript archives Compatibility Thidrekssaga QuirksBlog Donations CSS book Politics Mobile About COH

contact

Apple’s cycles

Over the past week or so I found I have seriously misjudged the iPhone 4S announcement, and I’d like to set the record straight. After reading a few interesting pieces I realised I’d misunderstood Apple’s patterns and cycles, and I saw how brilliant the 3GS move is.

I decided to write this little summary to firmly plant Apple’s pattern of cycles in my mind. No original thoughts here — I just collate ideas from other sources.

The design cycle

John Gruber:

The gist I get [...] is that a new form factor was never in the cards for this year’s iPhone. It may or may not have ideally launched a few months sooner, but the plan was always for an iPhone 4 successor that looked like the 4 but had improved internal components. [...]

Apple isn’t going to make a new form factor just for the sake of newness itself — they make changes only if the changes make things decidedly better.

But why did Apple feel a better design wasn’t necessary this year? I’m sure they’ve meanwhile improved on the 4’s design.

Horace Dediu explains. Apple doesn’t need an iPhone 5 right now because its current replacement target market are iPhone 3G(S) users who’re ready for a new model. To them, the 4(S) is still new.

iPhone 4 users, for whom the lack of a new design could conceivably be a problem, aren’t ready to upgrade yet because they’re still on a contract. As to non-iPhone users, they don’t care whether a specific design is new or not, since it’s their first anyway.

So Apple did not create a new design because the market didn’t ask for it.

Besides, in his Critical Path #10 show (this entire series is required listening) and a blog post Horace unveils a very compelling theory that explains even more of Apple’s rationale:

1. Apple has a massive amount of hardware on its balance sheet, so huge that it can’t be explained merely by Apple Store inventories and iCloud data centres. Horace proposes it’s machinery.
2. He argues that Apple itself owns the means of iPhone production. The machines used for creating iPhones in China are Apple’s, while the Chinese companies are mere cheap labour suppliers.
3. This means that Apple must have something close to a five-year plan for iPhone production: the new machines have to be ready to run once the new iPhone is unveiled, while the next generation of machines is already being designed.
4. Apple upgrades its machinery once every two years. That’s why the iPhone 3G and 3GS, as well as the iPhone 4 and 4S, have the same design. It’s cheaper that way because the machines and the logistics for creating the casing and touchscreen don’t need to be changed.

This gives Apple another reason not to release a new design this year.

The design cycle is in fact two-yearly: 2008 and 2010 saw a new design; 2009 and 2011 didn’t. Why two years? Because that’s how long the average contract between consumer and operator runs.

Well duh! It’s so simple once it’s spelled out to you.

The sales cycle

In Q2 20.3 million iPhones were sold; in Q3 only 17.1 million. Thus sales shrank in Q3, while they grew in Q2. In previous years this pattern was reversed.

This led to quite a bit of confusion. “Gone are the years when iPhone sales tapered off in the quarter preceding a new iPhone,” John Gruber predicted, but those years weren’t gone after all. Horace Dediu got his forecasts for both Q2 and Q3 wrong because he thought his model was faulty and later found out it wasn’t.

Part of the Apple cycle is lower sales in the quarter leading up to the new model because consumers know an announcement is imminent. Since the announcement is usually in June the sales dip usually affects Q2 figures.

This year people figured out there would be no June announcement but a September or October one, so the sales dip was postponed to Q3.

That’s all. The fundamental pattern is unchanged, though the timing shifted due to the production delay. I feel modest pride in having forecasted this right back in August.

Q4 will once more show monster growth figures that will bring Apple back to about 20% of the smartphone market. Not more than that, though. This year Apple still lacks a cheap phone.

The price cycle

I expected an iPhone Nano, and I was very disappointed when it wasn’t announced. Tomi Ahonen in particular has long called for a cheap iPhone, and I agreed with him that this would be the way Apple would conquer the low-end market.

Well, I was wrong. We got a $375 iPhone 3GS instead. That’s too expensive for a true low-end phone.

Still Apple made a brilliant move here. It just took me a while to get the point. Sorry for being so slow.

The iPhone Nano concept is modelled on how other phone vendors think. Release a new model with some but not many new features, and give it a low price, so that it’s clearly positioned as a modern low-end device. Add a splash, a bang, and a marketing campaign, and your phone is ready to roll.

Apple has a tendency to disrupt such processes. Witness the iPhone 3GS.

I still think the current 3GS price is too high for the low-end market, especially in countries where operator subsidies are illegal. I still think Apple’s market share will not grow considerably beyond 20% in the next year.

But that’s not the point.

The point is that the 3GS will be produced forever, but that each year its price will drop a little. The same goes for the newer models.

Apple wants a low-end phone, but it also wants to deliver top quality, which requires a full design process. The 3GS has already been through that process, and production costs get lower each year. Machinery has been depreciated, component costs will continue to fall.

At the iPhone 5 announcement the iPhone 3GS’s price will drop again. And when the iPhone 5S is announced in 2013 it’ll drop again. Once it reaches $25 it’ll be within reach of even the poorest of the poor. 2018? 2019?

People of the world, buy a 3GS once the price becomes bearable to you. And rest assured there’ll be an upgrade every two years at the same price point.

Bloody brilliant. And so elegant.

The broken cycle

John Gruber:

Apple is a company of patterns and cycles. These product cycles keep the machine functioning at a steady pace. They broke one pattern with the iPhone 4S: all previous iPhones were released in June.

But why was the pattern broken?

One could argue that the break was deliberate because Apple actively wanted to move its announcement from June to October. I’ve never heard that theory defended anywhere, though.

The delay took everybody by surprise back in May, and Apple’s communication in the first three to four months of this year sure pointed to the expected June announcement.

The other theory is that Apple hit some internal snag and postponed the 4S launch in order to fix it. Market-wise this wouldn’t give them any trouble: fans would wait, sales dips would be postponed.

But what kind of snag did Apple hit?

Tomi Ahonen offers an intriguing theory (under Why Delay). It hinges on the rumour that the next iPhone was going to be SIM-card-less, allowing the user to roam from network to network.

Now Apple is certainly working on a SIM-less phone: it’s even patented one. There was a rumour back in August that the next iPhone would have no SIM card.

The rumour could hold some truth. Cut the middle man, talk straight to the consumer, and give him the unlocked, universal phone he wants. From Apple’s and the consumer’s perspective this makes sense.

But operators hate SIM-less phones with a passionate hatred. The SIM card is their most important tie with consumers. Without the SIM card and a direct connection to consumers the operators would become dumb data pipes — exactly the fate they’re struggling so incompetently to avoid.

So, Tomi says, the operators told Apple to add a SIM card slot.

And Apple had to comply. It couldn’t afford to ignore the operators.

Apple cannot sell twenty million iPhones per quarter without operator subsidies and operator shops. €650 is too much for the average consumer who’s used to “cheap” phones. Besides, there aren’t enough Apple Stores.

How are iPhones sold? A consumer finds his contract is expiring, and he wants a new phone anyway. He goes to his operator’s store and sees a shiny iPhone. He notes the “price” is an affordable €150 or so — with a contract. He selects it. The store clerk is fine with that: his employer makes far more than the price of the iPhone plus mobile network traffic costs off the contract. He himself gets a slight commission. He happily helps the consumer with the paperwork.

Apple needs the operators’ subsidies and store networks. The operators refused to subsidise and sell a SIM-less phone, so Apple was forced to backtrack and add a SIM card slot, which caused the delay.

That’s Tomi Ahonen’s theory, and he’s not without experience and contacts in the operator world.

The internals cycle

How is the 4S different from the 4? Dean Bubley explains:

What Apple has actually been doing is working on a new hardware platform which will probably endure for several generations of its devices. That has likely taken a *huge* amount of work: chipset and hardware level integration is massively complex and needs lots of fine-tuning. It's quite possible to have to start again several times if the outcomes aren't perfect - something that Apple has the unique luxury of doing as it's not really that pressured cashflow-wise to get something out, even if it's compromised.

Dean thinks Apple is moving to Qualcomm chips throughout, and that the iPhone 4S is the first model in the new line. Besides, it had to support both GSM and CDMA. (That was part of the reason to move to Qualcomm, in fact.)

That’s quite a bit of work, and it could of course be that the reason for the delay is found here.

John Gruber adds:

[Apple added] a two-year cycle that starts with a new form factor (3G/4) followed a year later by a new phone with the same form factor but significantly improved internals (3GS/4S). If next year’s phone is named “iPhone 5” then I’ll expect a lookalike iPhone 5S in 2013.

The 5 will have a different design but largely the same hardware configuration. The 5S may bring another internals leap. And so on.

Missing feature?

Still, the iPhone 4S has somewhat fewer features than people predicted. Mostly that’s because the predictions were wrong, but Tomi Ahonen adds an an interesting note.

If, as he says, Apple had to change the 4S design in order to accomodate a micro-SIM card slot, something else had to be removed.

We don’t know what that something else is. We don’t even know if Tomi is right. Still, I’ll keep an extra eye open for any features the 4S might turn out to need. Maybe we’ll find a clue later on. Or not.

Apple’s cycles

Apple is on two separate, staggered two-year cycles: the design cycle and the internals cycle. They follow the standard two-year contracts of consumers with operators.

Each design will continue to exist indefinitely and will become cheaper and cheaper in the process, allowing anyone to start buying iPhones eventually.

Possibly something went wrong in the execution of this year’s cycle, causing a production delay.

The quarter just before the new iPhone announcement always shows less sales than the previous one because consumers know a new model is coming up and postpone their buy. If the announcement is delayed, the sales dip is also delayed.

There are a few open questions:

• Will the announcement date return to June, or will it stay in October? If it goes back to June, we’ll know something was wrong this year. If it stays in October nothing is proven either way.
• What will happen to the iPhone 4 at the next announcement? Will it continue to exist, or will Apple retain only the 4S, as it retained only the 3GS?
• Is it only the outer casing design that requires expensive changes in machinery, or do internal changes require one, too? I wouldn’t know where to even begin to answer this question.
• Will the iPhone 5S, due to be launched in 2013, have its internals revised as heavily as the 4S? Or will Apple be able to use this year’s improvements longer? Dean Bubley seems to think the latter.
• What’s the minimal price for a 3GS once machinery is fully depreciated and component costs hit rock bottom? The price will mostly be made up of labour costs and Apple’s margin.
• iOS 5 runs on the 3GS. We were explicitly told so, and now we also understand why. But will the 3GS run iOS 6? And iOS 7? One day the 3GS will not be able to run the latest iOS version any more. How will Apple avoid having to maintain several versions of the OS in ten years or so?

I don’t have strong opinions about any of these questions. I’ll just use the answers to refine my knowledge of Apple’s cycles.