Last week, Microsoft introduced the striking new Surface Pro 3, with its unexpected and welcome new form factor, and it didn't introduce its long-awaited Surface mini, after a second last-minute delay. Both of these events are of course related in a number of ways. But the most interesting, I think, is how these products both fit within a subtly-introduced new strategy for the Surface team.

Related: Microsoft Launches Larger Surface Pro 3

This is important, because Surface, so far, has been a complete disaster. Don't get me wrong, the devices have always been thoughtfully designed and well made, and Microsoft has at least proven that it can create hardware that is the equal of, if not superior to, the best PCs and devices in the market. (And yes, I'm including Apple in that list. As I've noted previously, if Apple only made an iPad "Pro" lineup, they might look something like Surface.)

Surface has been unsuccessful in two ways. First, the devices have not sold well, and of course everyone has heard about the nearly $1 billion in write-downs that Microsoft had to take when it overestimated sales of the first Surface RT device and thus overproduced the devices. This kind of problem could accurately be described as a learning experience, and the losses could simply be called an investment as Microsoft becomes a more sophisticated hardware maker.

The second issue with Surface is more troubling and has had wider ramifications for the firm and, indeed for the entire PC ecosystem. Whether they admitted it publicly or not—and many were in fact quite vocal—Microsoft's entry into the PC market two years ago was not welcomed by its PC maker partners. We can debate why Microsoft did this, and the firm certainly can make a compelling case for why these partners had done nothing but undercut the values of both Windows and PCs in general thanks to years of lackluster hardware, crapware bundling, and other unsavory practices. But whatever: It's equally fair to note that Surface competes with directly with the products its own partners sell. And these partners were not happy.

In fact, they were so unhappy that virtually all of these PC marker partners now sell laptops that run the rival Chrome OS from Google, a free, web browser-based operating system that is both simpler and limited compared to Windows, but also can be sold for low, low prices. Chromebooks, as they're called, have sold well enough that PC makers have embraced the platform wholeheartedly in a blind bid to reclaim past glory. It's like netbooks all over again.

And so Microsoft has responded, as it must. At Build in early April, it announced that it would provide Windows for free on phones and tablets that have screens under 9-inches in size and that cost less than $250, a direct attack on the low-end market for both mini-tablets and Chromebooks. And this past week, it announced that it would provide a new Windows offering, called Windows 8.1 with Bing for free (on sub-$250 devices, as noted above) or at a lower price than is offered for the previous low-end offering, Windows 8.1 "Core."

And then there's Surface. How does Microsoft make up for its foray into coopetition with its PC maker partners? Simple: It stops competing with those partners.

This isn't just obvious, it's semi-brilliant. And it explains why new CEO Satya "can't do no wrong" Nadella came to New York this past week for a very brief appearance at the Surface Pro 3 event. He is clearly the right person to communicate this strategy sea change.

Here's what he said.

"We're not interested in competing with our OEMs when it comes to hardware," he said. "In fact, our goal is to create new categories and spark new demand for our entire ecosystem. That's what inspires us and motivates us with what we're doing in our devices and hardware. Today is a major milestone on that journey."

This is a fascinating statement that not coincidentally addresses both of the issues I note above. Surface Pro 3 is the first Surface product designed specifically not to compete with Microsoft's hardware partners. And since Surface, generally, is now seen as a petri dish of sorts for creating new categories of PCs and devices, it doesn't necessarily have to be successful. It only has to inspire partners to do better and, perhaps, follow Microsoft into these new categories. It is an investment in the broader set of Microsoft solutions, not just PCs and Windows that are sold through partners, but cloud services, applications and more.

So Surface Pro 3 is a new category that doesn't directly lineup against the various 2-in-1 and transforming PCs that are currently flooding the market. It is, as Microsoft says, "the tablet that can be a laptop," with a unique 3:2 screen, thin and light form factor, and a set of innovations that is indeed both deep and wide. I've been writing about Surface Pro 3 daily since the announcement, so if you want to learn more, you can discover a ton of information about this device on the SuperSite for Windows.

This new Surface strategy also explains why Microsoft again delayed—yes, delayed, not cancelled—Surface mini at the 11the hour. (This happened less than one week before the event at which it was to be launched alongside the Surface Pro 3.) An executive team that consisted of Mr. Nadella, Devices Chief Stephen Elop and others determined that Surface mini simply wasn't differentiated enough from the existing Windows mini-tablets in the market. And because Microsoft does not want to compete head-to-head with its partners, Surface mini will be retooled again.

I do think Surface mini will make it to market, perhaps as soon as this fall. And that Microsoft's original vision for the third generation of Surface devices—Surface Pro 3 and Surface mini—will eventually wipe the old Surface lineup away for good. Those me-too devices, with their iPad-like 10.6-inch screens and a set of compromises so vast that they were almost universally non-compelling for any usage scenario you care to mention, simply didn't make sense. But Surface Pro 3—and the in limbo Surface mini—make plenty of sense. These are the Surface devices Microsoft should have made all along.

Back in February, I argued in the article "In Windows, Veritas" that Microsoft should focus its efforts on its core strength of productivity, that it should target what I called "the doers," those people who are being productive and getting work done. At the Surface event last week, Mr. Nadella echoed this notion and said that "do more, be more" was the unifying theme behind everything Microsoft creates. In short, they get it. And that theme is now being applied to the firm's hardware efforts. I love that, and think Microsoft is right to make this crucial strategy change.