How Apple turned a hardware plateau into a $100bn services business

Key takeaways

• Apple's Services segment did $26.3 billion in a single quarter in Q1 FY25, on gross margins around 75%. Hardware sits closer to 37%.

• The pivot wasn't a marketing repositioning. It was a decision to monetise the installed base — over 2.35 billion active devices — as a recurring-revenue asset.

• Apple didn't diversify away from the iPhone. It made the iPhone worth more per unit over its lifetime by attaching software margin to it.

• The brand didn't move an inch. The commercial architecture underneath it moved completely. That's the discipline most brand-led businesses get backwards.

• The transferable lesson: your next unit of growth usually isn't a new customer. It's a second commercial relationship with the customer you already have.

Around 2016, something inconvenient happened to Apple. iPhone unit sales stopped going up.

Not collapsing — plateauing. Which, for a company whose entire equity story had been "sell more iPhones next year than you did this year", is a specific kind of problem. The hardware S-curve had done what S-curves do. Replacement cycles lengthened. Emerging markets weren't going to save the average selling price. And every analyst covering the stock had a slide deck that ended with a question mark over the words "peak iPhone".

The interesting bit is what Apple did next. Because the obvious moves — cut price to chase volume, launch a genuinely cheaper phone, buy a media company, pivot to "we're a lifestyle brand now" — were all available. Apple did none of them.

It made a much more boring decision, and a much more commercially serious one. It decided to stop treating the iPhone as the product and start treating it as the distribution channel.

The brand story hasn't changed since 1997. That's the point.

When Steve Jobs came back to a near-bankrupt Apple in 1997, the first thing he did wasn't creative. It was structural. He killed most of the product line, cut the SKU count from something absurd to something like four, and drew a two-by-two grid on a whiteboard: consumer, pro; desktop, portable. One product per box.

That decision — specificity as a capital-allocation argument, not a design preference — is the thing that has held constant for nearly three decades. Every Apple product since has been an argument for concentrating resource on a small number of things done to a standard nobody else is willing to pay for.

Which is why the brand story around the Services pivot barely needed to move. Apple didn't have to convince anyone it now cared about music, or television, or payments, or fitness. It had to convince them that of course the company that made the device would also make the software layer that ran on it. The narrative was pre-loaded. The brand had done that work in 2001 with iTunes and in 2008 with the App Store.

What changed wasn't the story. What changed was the P&L behind it.

The commercial architecture: turning an installed base into an annuity

Here's the shape of the argument, stripped of the keynote gloss.

Hardware is a lumpy, capital-intensive, low-teens-to-high-thirties gross margin business. You sell a phone once, then wait 3-4 years for the customer to buy the next one. Every unit needs to be manufactured, shipped, warehoused, and returned. The whole thing is exposed to component prices, currency, and one very concentrated assembly partner.

Software delivered to a device the customer already owns is none of those things. It has near-zero marginal cost of delivery. It doesn't require a factory. It doesn't ship. And on Apple's disclosed numbers, it runs at gross margins somewhere around 74-75%, roughly double the hardware line.

The strategic question wasn't "how do we sell more iPhones?" It was "how much recurring, high-margin revenue can we attach to each iPhone already in a customer's pocket?"

Apple then built the answer as a portfolio, not a product:

• The App Store — a 15-30% take rate on a marketplace it doesn't have to stock.

• Apple Music, TV+, Arcade, News+, Fitness+ — subscription products with the entire customer acquisition cost already paid for by the hardware sale.

• iCloud storage — the most elegant of the lot, because it monetises the customer's own photos, which get bigger every year.

• AppleCare and payments — attach-rate margin on transactions the customer was going to do anyway.

Each individual line is unremarkable. Netflix has a subscription business. Spotify has one. Amazon has a marketplace. Apple's edge isn't that any single Service is best-in-class. It's that the acquisition cost for all of them is effectively zero, because the customer bought the acquisition channel and put it in their own pocket.

Why this is a brand argument, not a diversification argument

The lazy read of this is "Apple diversified beyond hardware". That's wrong, and the wrongness matters.

Apple didn't move away from the iPhone. It made the iPhone worth more. The lifetime value of a single iPhone customer, on Apple's own economics, is now materially higher than the sticker price of the device — because you're not just buying a phone, you're being enrolled into a set of recurring relationships that the phone unlocks.

The brand was the enabler of that. If the trust wasn't there — if customers didn't already accept that Apple would handle their photos, their payments, their kids' games — none of the attach economics would work. Every Service is a small permission the customer grants, and every permission is underwritten by a brand that has spent two decades not doing anything embarrassing with their data.

This is the bit most brand-led businesses miss when they read the Apple story. They think the brand is the moat. It isn't. The brand is the permission. The moat is what you build with the permission — and what Apple built was a recurring revenue system attached to an installed base of 2.35 billion active devices.

What you can steal from it

You are almost certainly not Apple, and you almost certainly don't have a two-billion-device installed base. That's fine. The structural lesson doesn't require scale — it requires sequencing.

Three things worth taking:

1. Your next unit of growth is usually behind you, not in front of you. Most brand-led businesses spend disproportionately on acquiring the next customer, when the more valuable move is opening a second commercial relationship with the customer they already acquired. Ask what the second product looks like before you spend on the second customer.

2. Brand is the permission slip. Architecture is the annuity. A strong brand that isn't attached to a repeatable commercial mechanism is a nice piece of equity waiting to be under-monetised. Apple didn't get more Apple. It got more attach.

3. Don't confuse a narrative change with a P&L change. Apple's brand story barely moved through the Services pivot, which is why it worked. If your reposition needs a new tagline before it needs a new revenue line, you're doing marketing. If it needs a new revenue line and the tagline barely has to move, you're doing strategy.

The iPhone stopped growing units almost a decade ago. Apple's revenue has kept going up anyway. That's not a marketing story. It's a commercial architecture story that happened to be wearing a brand.

Strong brands. Strong businesses. Strong people. That's the only architecture that holds.