Is What You're Working on a 'Zune'?

I opened the SmartLife app last week to adjust an air conditioner.

There was a new AI assistant button. I asked it what it could do. It took about 30 seconds to think about it, then told me it could help me with... all the things the app already does.

I closed it and adjusted the air conditioner myself.

This isn't a SmartLife problem. It's a pattern.

The app I find most irritating is the Copilot chat that now greets me on the Office 365 homepage before I've done anything. Don't get me wrong, Copilot is an impressive product and a serious business ($3B+ ARR). But this part of the implementation feels like it's designed around "we need to get you using Copilot" rather than "what were you actually trying to do when you opened this page?" (Open Outlook.)

The bolting-on is the problem. Even good technology can be deployed badly.

History has seen this before. And the companies that got it wrong didn't fail because they were slow. They failed because they were moving — just in the wrong direction.

Microsoft Zune, 2006.

The iPod had been on the market for five years. Apple had sold 100 million of them. Microsoft looked at that and asked: "how do we compete with iPod?" So they re-badged a Toshiba hardware device, added a music store, and launched at roughly the same price as an iPod. Peak market share: 2%. Written off: ~$300 million. Meanwhile, Apple wasn't asking how to improve the iPod. They were asking what comes after it. The answer launched the following year. It was called the iPhone.

Amazon Fire Phone, 2014.

Seven years after the iPhone redefined smartphones, Amazon decided it needed one too. The question they asked was "how do we get into phones?" The answer was a device with four front-facing cameras that created a 3D effect nobody had asked for, locked exclusively to AT&T, priced the same as an iPhone. Thirty-five thousand units sold in the first 25 days. $170 million written off in a single quarter. Discontinued after 13 months. The question Amazon should have been asking — "what problem do our customers have that a phone in our ecosystem could actually solve?" — didn't appear to have been answered before launch.

Google+, 2011.

Facebook was threatening Google's dominance of online advertising. So Google built a social network. The pitch to Larry Page, according to a former executive, was roughly: "Facebook is going to kill us." A fear response, not a customer insight. Google+ reached 90 million sign-ups quickly — and 3 minutes per month of actual engagement. The Wall Street Journal called it a ghost town. Google shut it down in 2019. Eight years of resource, brand distraction, and user goodwill spent answering the wrong question.

The pattern in all three is the same.

The question that launched each product was some version of: "how do we not get left behind?"

That's a valid fear. But it's not a product strategy. And it's not the question your customers are asking.

The companies that actually won in each of those transitions — Apple with the iPod and then the iPhone, Samsung with Android, Netflix across every format shift — weren't primarily asking how to compete. They were asking what people actually needed that the existing solutions weren't delivering.

Right now, a version of this is playing out across almost every industry with AI.

The pressure is real. Boards want to see it. Investors are asking about it. Competitors are announcing it. And so products are being shipped, features are being bolted on, assistants are being added to apps that control air conditioners.

Some of it is genuinely useful. A lot of it is Zune.

The question worth asking isn't "where do we add AI?" It's "what problem do our customers have that we couldn't solve before — and can we now?"

Those aren't the same question. The first one produces features. The second one produces businesses.

Are you moving toward a real problem — or just away from the threat of looking like you're behind?

Part 5 of the "Every Era Has Its Reboot" series. | redebuter.com Read more

Sources: Microsoft Zune market share — NPD Group via Knowledge@Wharton (2009); write-off estimate ~$289–300M, widely reported. Amazon Fire Phone — Amazon Q3 2014 earnings, $170M write-down confirmed. Google+ — WSJ/comScore engagement data (3 min/month); former exec quote from reporting on Google+ origins.