January 24, 2010 — The Upgrade Treadmill, Part 1


The apparent success of Microsoft’s Windows 7 in the marketplace is now forcing us to make those unpleasant upgrade decisions. The extraordinary long life of Windows XP has brought a sense of complacency, which in hindsight we will declare to have been dangerous and somewhat stupid. The end users in the organization (otherwise known as the Madding Crowd) are beginning to agitate for an upgrade. So we’re going to have to do it, I guess.

The simple statement “we’re going to have to do it” masks consideration, financial analysis, technical evaluation and cultural impact to the organization. The stability of the XP installed base allowed us to succumb to the temptation to defer hardware refreshes and channel the investment in infrastructure upgrades and new applications. While there always comes the time to pay the piper, I would argue that we came out ahead by skipping a generation of refreshes.

But now I have an installed base of about 3,000 systems, a majority of which are not really capable of running Windows 7 reasonably. On the cultural side, Windows 7 and its attendant Office upgrade will introduce a substantial change in working environment. So in addition to the hardware and installation costs, we will have substantial training costs, particularly for Office. File formats are another problem – I have dragged my feet on an Office upgrade for years, hoping the industry would settle on a stable, long-lasting file format standard for office suites. The .doc format is on its last legs and OOXML has not really settled down yet. That is the topic for another discussion.

This is an example of one of the worries which besets CIOs. The industry and marketplace are driving us to upgrade away from a platform which has been perfectly adequate for most uses and still is. We are forced to spend vast sums of money for something which exhibits a very weak business case – some would say there is no business value attached. But we simply have to do it — the alternatives are less attractive.

So we put together a project team and a target. Actually, I did this about a year ago. We started with the assumption we will upgrade to Windows 7 and Office 2010. Because the Office platform has imbedded itself so deeply into our ecosystem, any upgrade is non-trivial. It not only impacts the file migration and interoperability, but it also affects other enterprise applications which utilize Microsoft Office for report generation and mail merge. At the very least an upgrade will require extensive testing ahead of the fact, to avoid the kind of nasty surprises that can afflict these activities.

With careful preparation by this team, I anticipate an uneventful upgrade. This, of course, is precisely the goal for any CIO who charters major projects of this sort. It begins consultatively in conversations with the CEO and with other CXOs to ensure they not only are aware of the upgrade, but actively support it. This gives you the firepower to squelch dissent as the project moves on. It also builds consensus among senior management of the value of doing this upgrade.

If you are currently trying to develop your skills as a CIO, projects like this are a wonderful opportunity to build senior management’s confidence in you. Since you are going to have to do the upgrade anyway, you can accustom your team to doing things the right way, and building a set of practices which will useful on other projects.

Contact Marvin at mpreem@gmail.com