I wrote in the article I spent 25+ years at Microsoft, and Windows is getting the care it deserves again about how Windows evolved over the years, and that Microsoft said it’s putting more effort into it again; CEO Satya Nadella said it’s getting a new focus. From the comments, it’s clear that at least some readers don’t believe them.
Microsoft would not exist today if it hadn’t been for Windows, but it’s no surprise that the relative importance of such a core product has dimmed as other parts of the company grew.
Nobody buys Windows these days, for the most part. The spectacle of computer stores opening at midnight to sell Windows 95 to queues of expectant buyers hasn’t happened since. If you buy a new PC, it will likely come with Windows 11 pre-installed and inevitably the OEM will be paying some amount back to Microsoft to license the OS.
If you’re building your own PC from scratch, maybe you’d decide to shell out £120 (in the UK) for the Home edition, or £220 if you need some of the extra Pro goodness (assuming you’re not just going to use Linux; while that might be increasingly viable for techies, it’s a small percentage of the overall market). Maybe there are good numbers of people paying to upgrade up to Pro from the Home edition shipped with a new PC, but I bet the number of people buying a full, discrete license of Windows 11 is vanishingly small.
Traditionally, Microsoft made most of its Windows revenue from those OEMs selling it pre-installed on new PCs, or from enterprises who will license the Windows Pro/Enterprise desktop as part of a larger package. Many companies will have Microsoft 365 which includes Office apps and services, security suites and Windows client licenses, all built in. They might even buy Windows 365, meaning the whole thing is running in Microsoft’s datacenters, and they remotely access the machines.
“Which of your children do you love the most?”
In a large company like Microsoft (or Apple, Alphabet/Google, Amazon etc), investment in product and attention from executives tends to go on where they see growth opportunities and where they fear competitive threats. Throughout the company’s growth, the darling products will change – from Apple being the Mac company to the iPhone, Google Cloud becoming almost a fifth of the company, and Microsoft moving on from its Windows & Office duopoly to other areas. Some products or services will receive out-of-proportion investment because of the feeling of strategic importance – see all the $billions being poured into AI datacenters in the hope that there will be a payback in coming years.
Money that’s classed as R&D will be spent on current products, but is really just to improve current products with new features. Most technology companies will also have investment into research that one day may pay off, such as quantum computing.
Microsoft Research has a variety of long-term projects, pursuits that don’t actively develop a current product, but they’re there for the future. Technologies like fonts (remember ClearType?), handwriting and speech recognition have been developed in labs for decades but weren’t really about building a specific product. Measuring RoI on any kind of “pure research” is hard, as is the nearer term stuff that will just make things better.
Putting (say) an extra $1Bn into Windows development isn’t going to double its user base or meaningfully prevent existing users from replacing their PC with something else. Pouring money into Surface devices over the years might have helped Microsoft show the OEMs what they expect PCs to be like, and to try to compete with Apple doing arguably a better job on making more desirable Macbooks and the like.
Not putting $1Bn into fixing the broken windows and leaking roof of the old building that is Windows, could be more costly to Microsoft in the long run.
Windows’ wilderness years
Cartoonist and software engineer Manu Cornet published a joke Organization Chart which showed how different companies were structured…
Microsoft was famous for having warring factions competing against each other for investment and power. When Satya Nadella took over the reins at Microsoft, he even referenced this depiction in the opening pages of his book, “Hit Refresh”, and pledged to change the culture.
Everybody involved in developing a product wants it to be a good one – even if the Corporate Hunger Games mean that some groups get more influence than others. What I mean is that nobody is deliberately writing and releasing a bad product but if that does happen, it’s because of lack of attention, or infighting got in the way and maybe a few poor decisions were taken.
Windows client used to be the big dog, but its strategic importance to Microsoft as a whole has been waning for years. Even if a customer has a Mac on their desk and an iPhone in their hand, Microsoft will have something else in the ecosystem to sell them.
You won’t get anyone in Microsoft publicly admitting that they had neglected Windows in favour of something else altogether more shiny, but some mis-steps might have made that happen. A reorganization in 2018 split Windows into the core OS that sat in the Cloud & AI group (essentially Azure and all its friends, including Windows Server, under Scott Guthrie) and the UI and Apps belonging to the Experiences & Devices group, led by Rajesh Jha. Windows no longer reported to the CEO.
Panos Panay took charge in 2020, reuniting the Windows client teams and joining together with Surface hardware, launching Windows 11 the following year. The combo of OS and hardware all under one roof continued in the hands of Pavan Davuluri after Panos left for Amazon to head up their devices business. Initially, Pavan reported into Rajesh Jha, but since Rajesh announced his retirement in March 2026, Pavan has directly reported into Satya. So having split things apart and moved them around, Windows’ org has been rebuilt and it appears has been tasked with improving quality.
Perhaps there has been a realization that a bad product is in danger of having customers lose faith in the whole offering. If Windows has become, through lack of attention, a poor quality product that makes people unhappy, then if enough enterprises didn’t like what was going on with the desktop OS, they might be less keen on buying all the other stuff that’s sold to them as a follow on.
Learning from past mistakes
Take the furore around viruses and worms which blew up in the early 2000s: there was Code:Red, Nimda, ILOVEYOU and many more. Real reputational harm was being done to Microsoft because of the attacks on Windows, SQL, Office, and management was forced to act.
Microsoft halted all development and forced their developers to switch to undertaking training into writing secure code, then doing drains-up security reviews of existing code. Feature work was delayed unless it was building new capabilities to improve security: that’s where XP SP2 came from. This exercise was called the “Trustworthy Computing” (TwC) Initiative, with Bill Gates declaring, “When we face a choice between adding features and resolving security issues, we need to choose security.”
The recent mood music from Microsoft about fixing some of the issues with Windows 11 and making sure the quality bar for updates is higher, sounds like a similar move to TwC – the risk if they don’t do it, is to lose support from key customers. If Windows is the cornerstone for the rest of their Microsoft estate, they may move components away to competitors – identity, security, productivity – and eventually maybe even replace Windows with something else.
TwC was as much about being seen to do the right thing as it was about doing the right things. This supposed new focus on Windows quality feels like it could be in the same vein, and it’s about time.
The post After 25+ years at Microsoft, I can tell you why Windows was neglected appeared first on Windows Latest
Log in
