Windows 7: First Large Deployment and Trial Feedback

Rob Enderle

At the Windows 7 launch, I interviewed two companies in depth about their deployment plans. Given the ITIC survey results indicating that 30 percent of businesses would upgrade within six months of launch, and given that Vista will never get to 30 percent, I thought it would be interesting to see what was different.


One thing clearly different was that a reasonably large company had deployed the offering. This is the first time -- and I've covered every launch since Windows 95 -- that there has been a full deployment reference account available at launch. Trials aren't unusual, but Baker Tilly had rolled out 2,300 new Windows 7 desktops prior to the actual launch date. Typically the first real deployment comes months after a new operating system launches. David Hilland, deputy director of IT was my contact, and I've never seen any IT manager more excited about a new desktop OS.


I also met with Jonathan B. Wynn, manager of advanced technology and collaboration services, and David W. Glenn, director of enterprise operations for Del Monte Foods, who were undergoing a reasonably large and more common trial. Let's talk about what I found out.


Baker Tilly: Clean and Locked Down

Baker Tilly rolled out Windows 7 the way I would if I were going to go with a new OS early. Its managers put it on nearly everything and locked it down so users could not make changes. The first is so you can get everyone quickly on the same code base and focus on problems related to the new software rather than interoperability, and the second is so users don't install anything that hasn't been tested by IT yet.


Typically numbers surrounding savings are unreliable, but Baker Tilly is an accounting and business process company, so you should be able to trust its numbers. While it didn't do this to save money -- later I'll get to why it did this -- it is tracking nearly $200 in annual savings from each desktop, which translates to nearly a half-million dollars in bottom-line benefit (before tax impact). The company did run into a critical compatibility problem with an old Sun financial application that it's having trouble getting converted so about 2 percent of its systems remain on XP until that can be resolved by updating the product or hosting it.


Dell did the installation and did discount significantly because that company was treating it as a training exercise to build its Windows 7 deployment competency. I can't tell you the number of times a big vendor has learned on the job and not offered a discount. According to Hilland, Microsoft did not co-fund this.


The nearly $200-per-machine savings came from energy conservation, a sharp reduction in support costs and reduced image-management costs. There were no staff reductions, but existing staff was freed up to do projects that had been languishing. It took four weeks to roll out all 2,300 Windows 7 systems. The only user complaints were connected to being unable to load their own applications. Otherwise, according to Hilland, the employees loved the new operating system.

Del Monte: 65-System Trial, Deployment Months Off

Del Monte was trialing Windows 7 on 65 systems and had run into a number of expected problems with a large percentage for its vendor-supplied and custom applications. In getting ready for the trial, the IT staff discovered that a lot of the company's production applications had been poorly written and they were able to go back to the vendors and request that problems be corrected. This will take some time, but should result in better applications along with the expected Windows 7 improvements.


Wynn and Glenn indicated that the trial was going exceptionally well and actually gushed about Windows 7. I have to say having IT folks gush about anything is kind of a new experience for me.


Del Monte also is 90 percent mobile (laptops) and Wynn and Glenn appreciated the performance improvements that Windows 7 was showcasing on laptops, including better battery life, lower heat and overall better reliability. They even said their trial users were reporting better productivity, which is interesting given how often increased productivity is promised and how often IT says it doesn't see it. The most useful feature for Del Monte was branch caching or the ability to provide increased performance by caching files on servers located in branch offices. Wynn and Glenn said they thought this would significantly improve productivity with Office 2010.


It is interesting to note that no one was tying Windows 7 deployments to Office 2010 deployments. A friend of mine over at Gartner argued that the two systems were no longer tied together, and these interviews appeared to confirm that. While I have yet to run Office 2010 myself, if there was a product they were more excited about than Windows 7, it was Office 2010, especially Outlook. Wynn and Glenn indicated that, in the initial controlled beta, they were seeing some crashing, but that it was so much better that no one cared much. They had tested and assured me that the upgrade to Office 2010 was very simple once users were on Office 2007, so waiting for Office 2010 no longer made business sense. I look forward to trying that product and writing about it later.

Unexpected Benefits

It is interesting to note that both companies indicated that their aggressive use of technology was both a competitive advantage and a source of pride with employees, who often were deprived of newer technologies at other companies. Having newer hardware and software appeared to improve morale and employee retention which, rather than productivity or cost savings, were the primary reasons both companies were aggressive on products like this.


I have to admit that when I changed jobs one year and went to older hardware and software, I was really upset, and at a previous job, we had engineer turnover of 200 percent that, based on subsequent analysis, was tied back to the aging hardware they had to use, so it is important.

Wrapping Up

Windows 7 is no Vista. Companies really like it -- IT people seem to love it -- and with Vista, they ran screaming from the new version. That's often the nice thing about a maintenance release like Windows 98, XP and now Windows 7. It has nice improvements and none of the risks of a whole new platform. IT people who hated Vista the most may love Windows 7 the most. Now that's a change I can believe in.

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.


Add Comment      Leave a comment on this blog post
Oct 26, 2009 6:50 PM a. asdf a. asdf  says:

---"The nearly $200-per-machine savings came from energy conservation, "

Was that compared to XP or Vista? I can understand if it was compared to Vista, but that number seems a stretch compared to XP. Do you have a breakdown of the savings? What percentage is energy, etc.?

Oct 26, 2009 7:28 PM Rob Enderle Rob Enderle  says: in response to a. asdf

XP shop, I think much of the savings is actually coming from the new hardware they rolled out along with the fact they now centrally administer energy policy (ensure rapid suspend).   They've locked the users out of most of the settings with this roll out and can now assure systems unattended don't remain powered up. 

Oct 27, 2009 7:59 PM a. asdf a. asdf  says: in response to Rob Enderle

I just read a PCWorld article that puts the Win7 upgrade price from XP at $1930 per user for a 2.5k user organization,


which I think is a little overestimation. There is also a link in the article to a InfoWorld 21 page analysis of Win7;


which was very informative. I especially agreed with their conclusion on performance;

"Windows 7 is slightly faster than Vista on identical hardware. It's also still significantly slower than Windows XP, while generating almost twice as many threads and consuming nearly three times as much RAM as XP to run the same application load. The numbers speak for themselves."

When I run XP in Virtual PC, I am amazed at the responsiveness and the memory footprint. I have yet to figure out what Vista or Win7 does with all my memory and CPU cycles.


Post a comment





(Maximum characters: 1200). You have 1200 characters left.




Subscribe Daily Edge Newsletters

Sign up now and get the best business technology insights direct to your inbox.

Subscribe Daily Edge Newsletters

Sign up now and get the best business technology insights direct to your inbox.