Six reasons why Microsoft struggles with innovation

Microsoft invents a lot of great stuff but finds itself stuck when it comes to implementation.

When I think of Exchange 2010 and its hybrid approach to cloud computing, it reminds me that Microsoft can be innovative. With Exchange 2010, users can keep some e-mail accounts on premises while sending others to the cloud. It strikes a good balance between maintaining what customers want in an e-mail server product while gently leading them into next-generation cloud e-mail. But Exchange's hybrid approach is the exception. Overall, Microsoft struggles mightily with innovation for these six hard-to-fix reasons.

No. 1: Me-too thinking. The company spends billions annually on R&D and while it does have a whole bunch of interesting technologies in its labs, it focuses heavily on bringing me-too wares to market. It then tries to peel away customers who are basically happy with the original. Here is a list: MP3 players, video game consoles, Webcams, mobile platforms/devices, cloud-based Office applications, multimedia Web development, and Internet search and advertising. Arguably, Hyper-V can also be added to this list although Hyper-V is a good hypervisor at a great price that has a shot of overturning VMware's market-leading position.

What should Microsoft be doing instead the me-too game? Solving problems that others cannot. For instance, no other company knows more about developing operating systems than Microsoft. How about using that knowledge to help the world get off the hack/patch/fix/test/hack/patch cycle with say an entirely new operating system concept? We've heard snippets of such an operating system, called Singularity (formerly code-named Midori). It's reportedly been in development since 2003. 'Nuf said.

No. 2: Microsoft's customers don't like change: Microsoft is hamstrung by its own success. So says Microsoft Subnet's newest blogger, Michael Surkan. Surkan discussed with me the Windows Filtering Platform in Vista, and the complete overhaul of the network stack. He spent months reaching out to third parties that could be affected by the change, such as firewall makers. But with as many ISVs as Microsoft has, changes always tends to break something, somewhere, and Microsoft gets blamed, even if it did all it could to prevent such situations. Microsoft learned its lessons with the Windows 7 process of releasing many beta versions. But Windows 7 is basically an improved version of Vista. Application incompatibility will be kept to a minimum, but so, too, is innovation. Enterprises shoulder some responsibility. They stay with Microsoft because they know the technology. Change too many things and Microsoft opens the door for competitors. The enterprise that pays high prices and signs long-term contracts does so with the demand that Microsoft keep changes to a minimal. (Some enterprises are still clinging to Internet Explorer 6, for goodness sake.) Microsoft is trapped by a circumstance where innovation is at odds with stability.

No. 3: An inability to partner for great technology. Microsoft has a "built-it-here" mentality that does not serve it well. Why did Microsoft choose to create its own hypervisor instead of buying VMware when it was a start-up, or partner to embed VMware in with Windows Server? Why did Microsoft spend what some estimate to be a $1 billion developing the Xbox? It could have sold plenty of high-profit games on other wildly popular hardware platforms. When it comes to the server/PC and mobile market, Microsoft has put itself in another no-win situation. On the other hand, it doesn't own the hardware. Yet, it dictates hardware requirements so severely, its partners are eager to bring their innovative ideas to alternative platforms ... such as Acer's Android netbook. To be fair, no large-scale technology company does partnering especially well.

No. 4: Running counter to where everyone else is going. The folks at Microsoft do generate some great ideas, some of which land in products. But sometimes those ideas never take off because Microsoft's vision is out-of-sync with the rest of the technology world's direction. Surkan notes that Microsoft spent a lot of time implementing innovative uses of IPSec. In Microsoft products, you can create and manage IPsec tunnels, set policies and so on. A company with a Microsoft-centric network can, with relative ease, set up a network so that traffic is authenticated and encrypted and the edge firewall is all-but-unnecessary. Unfortunately, this didn't fit with Cisco's business plans. Microsoft's IPsec encrypted traffic can't be easily managed with traffic management tools. Microsoft has a feature in Windows Server 2008 R2 called DirectAccess that might be more of the same. It is a cool concept, as it acts as a VPN between the server and client. But it requires Windows 7. Organizations that aren't operating exclusively on R2/W7 (and that's all of them) will have to keep their VPNs for a while longer, and could have little reason to add DirectAccess support to the mix.

No. 5: Tying new ideas too tightly with its established wares. Microsoft has done some amazingly innovative work in identity management and cloud operating systems. But it typically attaches its best innovations to its Microsoft-centric world. If you want to use some of its more innovative technology, you've got to be running Windows or .Net, etc. It historically relies on third parties to sell products and services that will link its wares to others. That works fine for the Microsoft shop, but gives few reasons for non-Microsoft shops to buy its wares, no matter how innovative they are. It creates a self-fulfilling prophecy that leads back to reason No. 2.

No. 6: Offering too many versions of the same basic product: Microsoft's strategy to make money is all about upselling. That's fine until the many versions it has created has zapped its resources dry. Each version offers more features than the last, which should lead to increased innovation. But if you have one product team working on multiple releases, with multiple feature sets on each, you have spread yourself pretty thin. A tiny company with three engineers working on one product can be more innovative than a 20-member team working on six versions and 1,000 features.

By the way, I promised that I would include a pointer to Michael Surkan's latest research project, a survey on the state of IT spending in this rough economy. Please participate!

Like this post? Check out these others.
  • Virtualization Day: Virtual Machine Manager R2 RTM and Windows 7 XP Mode
  • Windows 7: the grown-up Vista
  • 5 Reasons To Use SharePoint Instead Of Emailing Excel Files
  • Microsoft's unified modeling platform, Oslo, faces another overhaul
  • Sam Ramji: 'Open source as an unstoppable phenomenon'
  • The Move-Mailbox cmdlet, a Cross-Forest move, and a complex password
  • Fine-Grained Password Policies, Part I
  • August giveaways: we're giving away SCCM books and free training

Plus, visit the Microsoft Subnet web site for more news, blogs, podcasts. Subscribe to all Microsoft Subnet bloggers. Sign up for the bi-weekly Microsoft newsletter. (Click on News/Microsoft News Alert.) Follow All Microsoft Subnet bloggers on Twitter

Follow Julie Bort on Twitter

Copyright © 2009 IDG Communications, Inc.

The 10 most powerful companies in enterprise networking 2022