Hacker News Reader

Comment by lukevp | Hacker News Reader

From years of observing first-hand how hard it is to develop software, I’m shocked that Apple continually makes products that fit their aesthetic (eg. The crown and headband evoking the Apple Watch) and have coherent and mostly-baked software on launch like this. Is there a book about how they’re able to operate at scale and deliver such consistent things? Do they just have an insanely good product team? How do they document requirements? It’s fascinating to me, because I think it’s their real competitive advantage - they get to the finish line with product that would make sense to a single, invested, super smart person, instead of a pile of inconsistencies and incoherencies that contradict each other as you move through the application. The only other product I feel is delivered at Apple’s level is CloudFlare, and I think that’s why they’ve grown so fast.

View on HN

I know it's a joke, but I think it's still a good illustration: https://i.imgur.com/XLuaF0h.png

From everything I've heard, Apple's org structure is a lot more focused and UX-driven from the top down than basically any other large tech company. That means you get a lot less compromising and in-fighting.

The other side is how Apple makes money compared to other companies. Meta makes money mostly by selling you (the user) to advertisers (the real customers). Same with Google (70%ish ad revenue). Microsoft is a hot mess of competing B2B fiefdoms (the vast majority of their revenue comes from businesses, not consumers). Amazon only makes a few products directly for consumers, and mostly makes money from being a marketplace + AWS. That leaves Apple fairly uniquely positioned as a premium consumer hardware company that also sells consumer software services to enhance the experience and... drive more hardware sales.

If you just draw the org structure for Apple differently in the link, it's same as Amazon..

Look again. Amazon is a tree, Apple is hub and spoke.

I don't think so.

Apple builds an insanely strong core from which any and all other teams can (maybe must) build on top of and use to power their team's platform.

You have a core messages team for example - off of which Mac, iPhone, and Apple Watch teams integrate. Do that for each core service or product (photos, safari, notes, music, etc).

The Amazon structure is such that everything flows from the person above you, work is not shared between teams.

One is compositional (Apple). The other is inheritance (Amazon).

If you look at it closely, any first level node with a child node at Apple can be bypassed by the central boss. The boss is connected to every other node.

You could say the same about Dyson or Oxo or Bodum. Having a single esthetic and coherent product line is a choice.

On the face of it, it sounds like a company would want coherency, but in practice it means leaving a lot of money on the table for every single of your products. Any consistent design decisions means anyone who's not pleased by it has no option in your lineup to go to. That's of course a gamble Apple is willing to take.

For instance, Apple consistently uses glass and metal in their product design, the only exception I can think of being airpods and their mouse. I personally prefered the plastic for weight and comfort, so that set me out of the market for their headphones and I went to Sony for instance.

Same for laptops with touchscreens, if you care any bit about them Apple is a no go to you. Same for so many things.

PS: on inconsitencies and incoherencies, Windows is the king of the kitchen sink approach, and it's consistently stayed ahead of Apple in numbers. Another instance on how being opiniated isn't only a positive aspect from a business perspective.

porsche also.

Your point is valid, but that ... diversity? ... comes at a _very_ steep price. Not only for the manufacturer, also for the consumer! With Apple I _know_ a couple of things about any future purchase, without ever having used, held or felt the particular product. Namely: 1) Build quality will be significantly higher than competitors, 2) It will hurt my feelings to look at the bill after checkout, 3) It will work as advertised (or slightly better).

Yes there also will be quirks I'll be unhappy with, which I _could_ (given enough time and experimentations) find a better solution for. But fundamentally I'll be getting a solid product, that works well enough. It is _that_ promise, which fundamentally allows Apple to do an online store and do a mostly directly-to-customer retail operation.

If you increase diversity, you make this "buy any product and you'll be fine" thing so much more complicated, loosing more sales than the niches even contain. I'd guess almost all people just want it to work :TM:, at least that's me, and that's me deciding the phone of my mum

I totally see this point, in particular as an opportunistic Apple customer myself. Apple doesn't need to do marketing regarding some aspects of their products (e.g. build quality) and can focus their messaging on more specific parts.

On the other hand, that doesn't seem to be what most consumers long for. Apple had two runaway successes with the iPod and then the iPhone, but to note, the iPhone is not the best selling phone brand nor dominates the international market. Same way Apple is not the #1 computer maker.

So yes, focus and consistency brings tremendous value, but there are many other ways to get that level of value, and many customers will have different priorities.

(I also don't think any brands need to be #1, being profitable should be enough, but I am more and more interested by how Windows has stayed at the top for so many years. In particular Apple progressively feels closer to Sony when it was at its peak: superbe products, strong ecosystem binging, coupled with also a heavy dose of strategy tax on how to benefit from what the other companies are doing)

Hypothesis: Apple is a hardware company that also does software, and that ethos allows them to ship more complex products over much longer time scales than their competitors.

I agree. Sometimes it feels Apple ships software too slowly. For example we have to wait for iOS 17 to be released to get a Whisper level dictation on the iPhone while Whisper has been out for almost a year. But maybe shipping at the pace of hardware is a good thing. Maybe we shouldn't deploy our sites on each commit

If I were to speculate I'd say it's because they don't ship things until they are finished. I'm sure a lot of the polished launches we have seen have been late by 6, 12 or 24 months. If the market isn't ready, the product doesn't ship. If the technology isn't ready (say, a high enough resolution VR display) to create a wow product then the product doesn't ship. I think it takes a great organization to not "accidentally" ship bad things. You need to prevent bonuses, egos, sunk costs, and self-deception from shipping something bad prematurely.

Creative Selection by Ken Kocienda is the closest, modern look inside Apple's product development that I've found. Focus is on the development of the software for the original iPhone and is a great book.

Yes! this book is a first-person account of product development at Apple, showing how both the design and the features of a system tool keep progressing under the selective pressure of iterative review sessions, with creative and specific challenges provided by managers at multiple levels, all the way to the top (Steve Jobs at the time of the book's tales). It's a great book, very well written, the only one I've read that provides insight into (some of) Apple's creative process. When they say that Steve's DNA has profoundly shaped the company, I imagine this is one aspect of what they mean.

It's Steve Jobs' legacy and a function of the authoritarian, highly controlled process he created. A democratic self-directed environment like Google or design-by-committee like Microsoft could never produce such a product line. Even an Intel with billions more in funding cannot recreate TSMC factories.

You might enjoy Ken Kocienda's "Creative Selection", which gives his first-person account of Apple's creative process.

They're absolutely awful at online / web-based services though. Embarrassingly bad at that kind of software.

I wonder how much can be attributed to Radar, their homegrown bug tracking / project management service.

Fascinating... Where can I read more about this Radar tool?

not only that but how they manage to keep it all relatively hush hush until the big reveal.

Reply on news.ycombinator.com