The entire Windows 8 roll-out was a palpably desperate effort to leverage Microsoft's existing desktop monopoly to gain a hold on what was/is seen as "the future of computing" tablets and phones.

Microsoft did very badly. But this kind of thing is hard to do and is not a one product or an incremental-product-improvements affair.

When Microsoft was fighting for it desktop monopoly, it did a good job of leveraging it's strengths to produce products that were "bad" by many measures but which satisfied a lot of user requirements and offered the end user good value. Surface is a logical product in this long lineage (and in answer to the parent's question, you can't create a Frankenstein product like Surface with focus group because Surface had to be a "quantum leap", an ad-hoc product of bucket chemistry). The problem is that apparently either the tablet market is not fluid enough that a product can win by just being a bucket of feature or Google already muscled into the feature-side and Microsoft isn't getting that part.

And equally, until now, Microsoft didn't actually sh-t all over their basic UI. But the approach of removing the start button subtracted more existing users than "Metro" could hope to gain.

You could frame this affair by saying that not only did MS' greed not abate over the years, it grew large enough to strangle the beast. Couldn't have happened to nicer people.

Microsoft isn't really leveraging their existing desktop monopoly; if they were doing that they might have had a chance. Instead they competing with Apple and Google on their competitors terms. They created an entirely new locked down API and a store; that isn't leveraging anything.

When Windows-on-ARM was demoed running full Windows 7 with all the fixings I was really interested. But then they crippled it.

There isn't anything wrong with Metro. Having a single OS across every platform is good too. But sacrificing the desktop for WinRT/Metro is the opposite of leveraging their monopoly.

If I was in charge of Microsoft, I'd put the Windows 7 start menu back in Win8. I'd make it so metro applications (and settings) could run windowed in the desktop. I'd focus on simplifying the desktop experience to make it more metro-like without forcing people to switch modes. Windows RT would run all ARM Win32 apps unrestricted.

Also when microsot was fighting for desktop it made one wonderful thing: brought in the platform a shitload of developers ("... developers, developers!"). That was the main advantage and that is why they succeed (if we ignore the "forced monopoly" part).

But now with windows 8 and metro ui with winRT(/JS) it seems the problem is mainly that, simply, the are not developers who want to join the ship. I personally do not know any company or indipendent developer who is betting himself in it or even just exploring it. Anyone! And, infact, if you look at the win8 market (that is also the RT market and only way to get apps for that os) is almost empty or, if you want to see it differently, full of crap-copy-useless-apps.

At least with windows phone 8 the situation is a bit different, but here who we have to thank? Microsoft or Nokia? I bet in the second one.

> My simple, genuine question is: how did this happen? I mean, Microsoft has huge resources, and presumably does testing with focus groups, etc.

I'm sure they do, and I'm sure they focus-grouped the tablets extensively.

But it's very hard to convince someone of something that they really, really don't want to be true. There's a Dilbert cartoon (http://dilbert.com/strips/comic/2012-11-06/) where the Ugly Truth is that there's nothing Microsoft can do to stop its decline into irrelevance and that producing a tablet is their attempt to hide that truth from themselves.

And maybe Microsoft's internal culture makes it hard for underlings to tell their bosses bad news, so any bad news gets a layer of amelioration and obfuscation for every layer of management.

Put all that together and its easy to see how a big failure could happen.

Re: "a layer of amelioration and obfuscation for every layer of management":

http://ogun.stanford.edu/~bnayfeh/plan.html

Haha, that's some great stuff. Gonna have to keep that in mind every time i see plans.

I think it's classic "if you ask horse & buggy owners what they want, it's a faster horse" type thinking. People say they want "real windows" on a 10" screen, or at least some group of people say that. But that's not what they want.

I'm pretty much Microsoft's target demographic with Surface: I use Office all day at work, and often need to work on the go because things pop up. I might say I want Windows on a tablet, but that's not really what I want. I want perfect office document compatibility, so I can review markups and changes on the go. I want to be able to log into my company's domain and browse my company's DMS. But that's what I really want, even if I say I want Windows on a tablet. I don't really mean I want full-blown Windows with its mouse-and-keyboard centric UI on a tablet...

The sad thing is that I know that there are brilliant enough people at Microsoft to ask the right questions, to drill down in their focus groups and figure out that the horse & buggy owners could actually really use a buggy, not just a faster horse. As startup people, we try to do this every single time we talk to someone! But somehow the institution doesn't allow those people to ask the right questions and act upon the answers - rather, it seems to encourage the short-sighted, iterative questions that would lead to the "faster horse," or in this case a whole bunch of tiny features that miss the big picture.

Its because the people who makes the decisions are too far removed from the people doing the actual work.

In an industrial/manufacturing setting, this might work because the people doing the work isn't doing anything creative (just labour). In a research and development setting, this cannot work because a decision needs to be made using information obtained from "the trenches", and management often are not in the trenches.

This is why small startups are always more agile and moves fast, can come up with products that suite their target better than a corp. I dont think this can change, unless the corp change their method of operation from a hierarchical one to a very flat, self-managed groups of small people, and trim the fat out. This has to start with "trust", and now-a-days, trust is hard to come by.

Well, its CEO refuses to even let rival products in his home, for starters. Talk about a well-insulated bubble, that's a great way to lose touch with customers.

Wasn't that Gates, not Ballmer?

No, that's Ballmer: http://money.cnn.com/magazines/fortune/fortune_archive/2006/...

7 years ago... in a light-hearted response to a stupid interview question...

I'm sure they did all of these things. The only place I can reasonably see it breaking down is in the reaction of upper management. All the advance planning in the world won't save you if the CEO insists that the product will be a hit even when your focus groups indicate that it will be a flop. I can only imagine that's what happened.

Microsoft has always relied on "you have no choice" as its major marketing tool. (X-Box is the big exception.) This is the price of relying on it in a world where it turns out people do have a choice.

Remember how awful Vista was? How did needing to confirm your WiFi password ever get through any kind of review process?

as i said below, in my opinion, ballmer bet to much on the "we have office and you need it" part.

Also he was totally confident of success of windows 8, and with that success it was confident in the metro ui and then in the ease of switching to a rt tablet.

If you see these as a chain of events, the conclusion is really simple.

It was less about Office than Windows getting 400m installs and becoming a platform that devs would be dumb to ignore.

The logic was: - Windows automatically gets a shitload of installs a year (from new PC sales) - Devs will flock to the platform because of the install numbers - Users will buy Win8 tablets because of vibrant app ecosystem.

The issue is that despite the revamped start screen, win8 has not led to massive demand for modern apps. So they've stalled on step 2.

Make no mistake though, Win8 is still being sold on a lot of machines. At some point the majority of PCs will have access to the Windows store.

But we are strongly moving in a world in which the main "platform" for the common consumers are not more pcs, but tablet and (super)smartphones. Already this year will be sold more tablets than pcs.

This is why for microsoft was really important to get marketshare in the tablet market. Now they seriously risk to be stalled in step2 for a long time cause:

1 - no one is using winRT or willing to use it

2 - no one who is using win8 use mainly the metro ui or will to have a system in which that is the main ui

3 - for reason 1 and 2 the developers are not going to spend their time in making app for that ecosystem, preferring still win32 and the desktop/web functionalities or other os more strong in the tablet market (ios and android atm).

Also the fact that windows phone is using another group of libraries and the market is totally independent in relation of winRT is another problem.

It's like for the first time microsoft failed each important decision they had to make

There were plenty of things that doomed Surface RT's initial rollout to failure. It may have done well with focus groups, and I'm not convinced that it was an objectively poor product to the extent that it was guaranteed to fail. But right off the bat they had poor marketing combined with poor distribution (Surfaces were only available online or in Microsoft stores, of which there aren't very many, so very few people could try before buying). Even once they started allowing Surfaces to be sold through other physical retailers, they did a poor job training salespeople and getting attractive placement in stores. And the RT is also just a hard product to sell - there are lots of aspects that are really nice, but it's lacking a simple elevator pitch.

I'll add to that. I tried very hard to buy a Surface Pro from their online store. It was literally impossible. They did not allow me to give them money, no matter how much I tried. They had separate systems for everything (the online store, the physical store had different systems and databases), needed a live login to make a purchase (why?!), didn't accept any of my cards with the helpful error message "error: 0". When on the phone with them it took over an hour to order it. I had to be a very persistent customer in order to finally get one. So far I'm very satisfied with it, the Wacom pen and it being a laptop/tablet replacement are the killer features for me. But after dealing with their online store and phone support, and add to that their image problems with the public, and I'm not surprised it's not selling well.

Hubris