When Microsoft CEO Steve Ballmer announced he'd be stepping down back in August, I had a chance to ask him about his biggest regret during his CEO tenure.
His short answer was "Longhorn," the repeatedly delayed and ultimately recast operating system that finally was released as Windows Vista in January 2007.
When Ballmer and I talked last month about defining moments of his 13-year CEO tenure for a piece I wrote for Fortune, Longhorn came up again. Ballmer said he considered it one of four defining moments of a good part of his CEOship. (The other three, for those who are curious, were the resolution of outstanding legal issues; the new business-unit structure, which Microsoft only just recently undid with its "One Microsoft" reorg in July 2013; and the reconfiguration of his working relationship with founder Bill Gates once Gates became Chief Software Architect in 2000.)
"When I look at it and I say, okay, what's the thing that I did that I feel -- that I regret the most, not just in my CEOship but my whole time here, it's absolutely 'Longhorn becomes Vista.' That was the single biggest mistake I made," Ballmer told me during a sit-down interview in his office in Redmond at the end of November 2013.
"Why?" he continued. "Not only because the product wasn't a great product, but remember it took us five or six years to ship it. Then we had to sort of fix it. That was what I might call Windows 7.
"And what we wound up with (was) a period of let's say seven or eight years where we had the A-team -- not all of the A-team but a bunch of our best people -- tied up not driving. We did not make years progress in eight years, and there were other things those people could have been working on, (like) phones," he conceded.
Ballmer said he personally takes responsibility for the situation.
Ballmer: 'We tried to fight it off'
"The mistake wasn't just an executional mistake. It was a technical strategy mistake. We tried to fight it off," he said.
"The big things are the important things to get right in this industry, but then you've got to execute with cadence," he continued. "People think it's about changing strategy every three seconds, because that's what people say. 'Oh, the industry changes so fast.'"
But in reality, Ballmer said, "a few big bets really pay off." He noted that Apple's "bet on touch and low power" ended up working out well for the company. Google made a big bet on search. Microsoft made a bet on PCs and software, and more recently, the data center with Windows Azure.
With Longhorn, Microsoft started out with the wrong technology approach, the wrong focus by the company's tech leadership and the wrong cadence.
Longhorn kind of fell through the cracks. "It wasn't Bill (Gates') thing and it wasn't (former Windows Chief) Jim (Allchin)'s thing and I didn't get it," Ballmer admitted.
"When I look back and I say it was sort of a focus issue, because we weren't focusing on what we needed for engineering cadence, bite-size approach, what was the big bet that that represented," he said.
"I'm willing to admit when I first started as CEO is probably when I made my biggest mistake. And a lot of what we've been doing is just the last five or six years is really catching back up from the mistake that really you could say I made with Bill and Jim Allchin earlier in my CEO days," Ballmer said.
I asked Ballmer whether either almost buying Yahoo in 2008 -- or ultimately backing away from the acquisition -- was something he regretted or considered a mistake.
"I'd call it mostly -- in the grand scheme of things -- not monumental to the company's fate either way," he said.
"I think it's important for us to build up the capabilities that we're building up in machine learning and search and the like. And we were headed down that path, and the scale and volume of Yahoo (with whom Microsoft has a search partnership deal) has mattered. There's no question about that.
The decision to bid on Yahoo "had a lot of economic synergy, a lot of product creation," Ballmer said. But "in the annals of history, it will be really smart we didn't buy it, because the market collapsed, of course. It would have really looked silly to have bought it."
Ballmer noted that, after Microsoft withdrew its bid, Yahoo attempted to forge a deal with Google. But the antitrust regulators weren't going to allow that. If they had, "you would say, hmm, who knows what position we would find ourselves in," he said.
While the Yahoo deal was big because of its $45 billion size, Ballmer said, "in the grand scheme of the shape of the company I don't think it will be in the top 10."
Another regret Ballmer did have, however, was how long it took the company to make money with Xbox, he acknowledged. Microsoft launched the first Xbox in 2001 and the second, the Xbox 360, in 2005. Up until 2010 or so, Microsoft lost money on every console sold, according to estimates.
Ballmer said he still believes Microsoft made the right move in getting into the living room with Xbox. He isn't sorry he gave the team a lot of freedom in creating and building the product.
"But what you're trying to do is make money for the long run, not the short run," Ballmer said. "So I feel bad about how we got here on Xbox, but we've built a heck of an asset. And could we have built it a little cheaper, yes. But we built it. We weren't swayed from building an important asset."
Ballmer said he's feeling good about where the company is right now with the Xbox One.
"We have almost a full Windows System now back in it," he noted, referring to the Windows 8 and Hyper-V underpinnings of the platform. " And I'm glad we built a base. I'm glad we built some enthusiasm. I kind of wish we were a little closer to the Xbox One architecture sooner in the process."