informa
/
2 MIN READ
Commentary

Death By Architecture: Eight Steps For Surviving Instead

Perhaps nothing is more drawn out and aggravating for an IT organization that what I call 'death by architecture.' Sound familiar? Getting visions of standing on a very narrow ledge very high above the ground? Fear not - here's an 8-step plan for surviving the ordeal from Cutter Consortium.
"Perhaps nothing is more drawn out and aggravating for an IT organization that what I call 'death by architecture.' " Sound familiar? Getting visions of standing on a very narrow ledge very high above the ground? Fear not - here's an 8-step plan for surviving the ordeal from Cutter Consortium."There are right ways and wrong ways to do everything, and IT architecture has more than its share of wrong ways. Architecture is hard to do well," writes Cutter's Enterprise Architecture Practice director Mike Rosen in a recent blog. Ticking of a list of obstacles such projects usually encounter, Rosen mentions megaplans that are years in the making and, when proposed, no longer are relevant to the business; a lack of buy-in for the overall effort; a lack of clarity of the project's proposed value; changes in management; a shift in priorities; and several others.

But then, after he's got his readers good and depressed and thinking about some of the details in their last wills and testaments, Rosen offers an 8-step survival guide that I'll summarize below - be sure to see his post for full details:

--Quickly create an architectural vision and strategy - two months is the maximum allowed. --Be careful in selecting the starting point for implementation: not too big, not too small. --Tie a small portion of the technical and application architecture to a project with very clear business value, with a focus on showcasing archistectural consistency and flexibility. --Along the way, build out the business and information architecture and use those to guide project design. --Iterate and integrate: after each project, keep the overall effort current by integrating lessons learned into next steps. --Focus like crazy on metrics that prove value across cost, time, and quality. --Strive to be good and on time, rather than struggle to be great while just being late. --"Don't try a big-bang approach. It never works." (That's a direct quote because it's too good to pass up.)

The rewards for doing it right, Rosen emphasizes, are significant, and well worth the effort: "There are many very successful architecture projects and they are a joy to behold - from a technology perspective, from how they enable and transform the business, and from how they have changed the organization's behavior and perception of architecture."

Perhaps I'm being naïve, but as lousy as the economy is right now, it's not going to stay this way forever. And when growth returns, you're going to need some sage advice like this, so be sure to check out Rosen's post and the related comments - during these days when too many of us have spent some time out on that ledge, his 8-point plan offers a far better alternative.

Editor's Choice
Brian T. Horowitz, Contributing Reporter
Samuel Greengard, Contributing Reporter
Nathan Eddy, Freelance Writer
Brandon Taylor, Digital Editorial Program Manager
Jessica Davis, Senior Editor
Cynthia Harvey, Freelance Journalist, InformationWeek
Sara Peters, Editor-in-Chief, InformationWeek / Network Computing