My 2025 Apple Report Card

Wednesday, 25 February 2026

This week Jason Snell published his annual Six Colors Apple Report Card for 2025. As I’ve done in the past — for the report-card years 2024, 2023, 2022, 2021, 2020, 2019, 2018 — I’m publishing my full remarks and grades here. On Snell’s report card, voters give per-category scores ranging from 5 to 1; I’ve translated these to letter grades, A to F, which is how I consider them. (See footnote 1 from last year’s report if you’re curious why it’s not A to E.)

As I noted last year, “Siri/Apple Intelligence” is not a standalone category on the report card. I know Snell is very much trying to keep the number of different categories from inflating, but AI has been the biggest thing in tech for several years running. If it were a standalone category, last year I said I’d have given Apple a D for 2024. This year, I’d have given them an F — an utter, very public failure. (Their AI efforts in 2025 did end on a mildly optimistic note — they cleaned house.)

Mac: C (last year: A)

If there were separate categories for Mac hardware and MacOS, I’d give the hardware an A and MacOS 26 Tahoe a D. The hardware continues to be great — fast, solid, reliable — and Apple Silicon continues to improve year-over-year with such predictability that Apple is making something very difficult look like it must be easy.

Tahoe, though, is the worst regression in the entire history of MacOS. There are many reasons to prefer MacOS to any of its competition — Windows or Linux — but the one that has been the most consistent since System 1 in 1984 is the superiority of its user interface. There is nothing about Tahoe’s new UI — the Mac’s implementation of the Liquid Glass concept Apple has applied across all its OSes — that is better than its predecessor, MacOS 15 Sequoia. Nothing. And there is much that is worse. Some of it much worse. Fundamental principles of human-computer interaction — principles that Apple itself forged over decades — have been completely ignored. And a lot of it just looks sloppy and amateur. Simple things like resizing windows, and having application icons that look like they were designed by talented artists.

iPhone: A (last year: A)

iPhone 17 Pro and Pro Max are, technically, the best iPhones Apple has ever made. They’re very well designed too. The change to make the camera plateau span the entire width of the phone is a good one. It looks better, allows a naked iPhone 17 Pro to sit more steadily on a flat surface, and lets one in a case sit on a surface without any wobble at all. Apple even finally added a really fun, bold color — “cosmic orange” — that, surprising no one, seems to be incredibly popular with customers.

The iPhone Air is, from a design perspective, the best-designed iPhone Apple has ever made. It’s a marvel to hold and carry. One rear-facing camera lens is limiting, but it’s an excellent camera. Not 17 Pro-quality, no, but excellent quality, yes. Battery life is amazing given the physical constraints of the iPhone Air’s thin and lightweight design. The main two dings against the iPhone Air are that (a) Apple didn’t offer it in a fun bold color like the 17 Pro’s orange, and (b) Apple, bafflingly, hasn’t advertised the Air. I’ve seen so little promotion of the Air that I’d wager most iPhone users in the market for a new phone don’t even know it exists until they walk into a store and see it there.

The no-adjective iPhone 17 is the best iPhone for most people, which is exactly what the no-adjective iPhone ought to be. Back in March, the iPhone 16e introduced both a terrific lowest-price new iPhone — including the then-current-generation A18 chip — and a significant shift in strategy from the SE models of yore. The SE iPhone models were only updated sporadically, going 2–4 years between revisions. The “16” in “16e” is a pretty strong hint that Apple now intends to update the e models annually, just like the rest of the iPhone lineup. People complain about the $600 starting price for the 16e, but that’s $200 lower than the no-adjective iPhone 17. If you’re in the market for a lower price than that, you’re in the market for a refurbished older iPhone.

iOS 26 is Apple’s best implementation of the Liquid Glass concept, by far. I prefer it, in just about every way, to iOS 18. There are some individual apps from Apple in iOS 26 that have poor implementations of Liquid Glass (Music, I’m looking in your direction) but most of them are decided improvements, with more consistency system-wide (like the placement of search fields).

iPad: B (last year: C)

iPad hardware continues to be fine, and “fine”, by iPad standards, means “the best tablets in the industry by far”. The lineup is well-differentiated and spans a larger than ever gamut, ranging from “totally casual user” to “actual pro usage”.

iPadOS 26 is the most exciting release of iPadOS ever. I don’t love all of it. I think the biggest problem is that too much complexity is exposed to very casual users, for whom the main appeal of using an iPad as their main “computer” is its rigorous simplicity. But the course reversal Apple has made for advanced users, from eschewing (often to the point of frustration, sometimes to the point of absurdity) the desktop GUI concept of overlapping windows, to embracing regular old-fashioned GUI windows, was the right call, and a welcome sign of humility.

It’s a new start for iPadOS, and I look forward to seeing where it goes. It’s been a long time since I’ve thought that about iPadOS.

Wearables Overall: B (last year: B)

AirPods Pro 3 are frigging amazing. AirPods, overall, continue to exemplify Apple at its best.

Apple Watch: A (last year: C)

Apple Watch Series 11 and Ultra 3 are solid year-over-year improvements from the Department of If the Design Ain’t Broke Don’t Fix It. Battery life improvements, in particular, are impressive. No one comes close to Apple at making very small, powerful computers that don’t really seem like computers at all. And the best Apple Watch news of the year, by far in my opinion, is the SE 3. The SE 3 is simply an outstanding Apple Watch at very low prices ($249 for 40mm, $279 for 44mm). That’s the price range a lot of people are looking at if they’re thinking about getting themselves “a nice watch”, smart or not.

Vision Pro: C (last year: B)

An M5 speed-bump update to the Vision Pro was nice to see, but only as a sign that Apple is still committed to this new platform. And they’re actually starting to build a nice little library of immersive content that is extremely compelling — including baby steps toward immersive live sports with a limited slate of games, albeit just from one single NBA team (the Lakers). The new version of Personas in VisionOS 26 are amazing, and strikingly improved from the first implementation. That’s another sign that Apple is doing amazing things with this new platform and the concept of spatial computing. But in terms of VisionOS being a productivity platform in its own right (not counting the excellent Mac Virtual Display app), I didn’t see any progress at all. Nor any outreach at all to third-party developers to make VisionOS into a serious productivity platform. Frankly, it’s weird — perhaps even alarming — that some of Apple’s own core apps like Calendar and Reminders are still iPad apps running in compatibility mode, not native VisionOS apps.

Home: D (last year: D)

Why isn’t this platform improving, in drastic groundbreaking ways, with any urgency? I really thought 2025 might be the year, but nope. I can’t think of any area where Apple’s attitude more clearly seems to be that “good enough” is good enough.

Apple TV: C (last year: C)

Same grade, same comment as last year (just replacing the specific year with “[this year]”):

I’m a very happy daily (well, nightly) Apple TV user. But what exactly improved or changed [this year]? Anything? It may well be fair to say the current hardware — Apple TV 4K 3rd-gen, which shipped in November 2022 — is fine, and this is a hardware platform that only needs updates every 3 or 4 years, but we’re grading what happened [this year].

Also: I feel like Apple has never yet made a truly great remote control for this platform. The current one is their best yet, but it has obvious flaws.

I fear complacency has set in. Apple TV 4K really is so much better than any competing set-top box (or built-in smart TV system), but it also still falls so far short of “insanely great”.

Services: B (last year: B)

Quality is high, value is fair (except, still, for iCloud storage), and it’s getting to the point where it’s hard to keep up with all the great series on Apple TV.

Hardware Reliability: A (last year: A)

No news remains great news in this category.

Apple OS Quality: C (last year: B)

Apple Apps: B (last year: B)

For two straight years, I’ve written the same comment for this category: “I have concerns and complaints about aspects of the direction Apple’s software design is headed (or in some ways, has been now for years), but their software reliability has been very good for me.”

The reliability and technical quality remains excellent. While writing this report card, I checked, and my uptime on MacOS 15.7.2 got to 91 days before I got around to restarting, which I only did to upgrade to 15.7.3. At one point I literally had over 1,000 tabs open in Safari, spread across over 50 windows. (I have a problem with tab hoarding.) That is technical excellence.

But years-long growing concerns over the direction of Apple’s software design reached a breaking point with MacOS 26 Tahoe. It’s so bad — or at least, so much worse than MacOS 15 Sequoia — that I’m refusing to install it. That makes it hard to assign a single grade for “OS Quality”.

Developer Relations: D (last year: D)

Fifth year in a row with basically the same comment: Resentment over App Store policies continues to build. Frustrations with the App Store review process seem unresolved. Apple’s goal should be for developer relations to be so good that developers look for excuses to create software exclusively for Apple’s platforms. The opposite is happening.

Social and Societal Impact: F (last year: B)

Tim Cook is in an excruciatingly difficult position regarding the Trump 2.0 administration. But that’s his job. He’s clearly attempting to take the same tack he took with the Trump 1.0 administration from 2017–2020, which, in hindsight, he navigated with aplomb. To wit: staying above the fray, keeping Apple true to its institutional values while keeping it out of President Trump’s wrath.

But the Trump 2.0 administration isn’t anything like the 1.0 administration. Cook, addressing employee concerns back in 2016 regarding his participation in then-President-elect Trump’s “tech summit”, said, “There’s a large number of those issues, and the way that you advance them is to engage. Personally, I’ve never found being on the sideline a successful place to be.”

“Awarding” Donald Trump a 24-karat gold trophy emblazoned with the Apple logo in August 2025 — after seeing eight months of Trump 2.0 in action — wasn’t “engagement” or “getting off the sideline”. It was obsequious complicity with a regime that is clearly destined for historical infamy. Cook’s continued strategy of “engagement” risks not only his personal legacy, but the reputation of the company itself.