My editor brought an article to my attention yesterday that contained some startling news. "Popularity of IT integration and service oriented architecture waning," warned the headline.
But integration? Now that hurts. And it's hard to believe, given how much the IT community talks about integration.
As it turned out, Computer Weekly is based in the UK. The study cited was conducted by a UK-based firm, Reed Business Insight, and I'm guessing those statistics are from surveys of UK-based companies. I'm waiting on e-mail confirmation of that, however.
I'm confident those statistics would look different in the United States and possibly in other European countries. Still, it is possible the survey is simple a harbinger of things to come in the United States, especially since Europe is claimed to be ahead of the United States in SOA adoption.
Aside from questions about the survey's scope, I also have reservations about the article's conclusions. It often seems to compare apples to oranges.
But, criticisms aside, I'm still recommending it, because it's filled with good advice about how you can ensure you're putting the business first -- even ahead of integration, SOA and other high-minded ideals about how IT should operate. True, the business may benefit when IT is most effective and efficient. But, the business has to balance IT's needs with the rest of the company. And let's face hard facts: That argument won't secure funding at budget time.
Here's my take on the advice I found most helpful from the article: Measure, measure, measure. Incredibly, the Reed Business Insight survey found fewer IT decision-makers were measuring the business benefits of IT integration than in 2006. And it's an extreme difference, not just a few percentage points. In 2006, only 5 percent said they had no integration metrics. Now, 33 percent reported they weren't measuring the business benefits of IT integration. Bad, bad, bad move.
You've heard it said, "What gets measured, gets done?" Well, I have my own version of this saying: "What gets measured, gets money."
I'll give you a real-life example. I once had an editor who gave me a meager pay raise, saying she didn't think I was producing as much copy as I was the previous year. She based this on ... a vague feeling. Fortunately for me, print copy can be measured, and I went back through the whole two years, measuring the number, length and variety of the articles I'd written. It turns out, her "feeling" was wrong -- way wrong. I had doubled the amount of copy I produced. I proved it with measurements -- and I got a huge raise. What gets measured, gets money.
Get real about what your business needs. My husband completed a Project Management Institute certification about a year ago. He loves where he works, but his company does not use PMI methodology, and he's never had a chance to put that certification to work. He keeps telling me how PMI could help this or that if the company would just use it.
Now, maybe that's true, but this company believes it's doing very well without PMI, thank you very much. Its core business is digging ditches and maintaining sewage, which it seems to manage very well with its own, engineer-based approach to juggling projects. It isn't going to change just because someone in IT says so, because that's not its culture. I keep telling him, "They do sewage. They listen to engineers. They're not going to do PMI because, frankly, they're getting along fine without it."
Not every company needs the most efficient, effective IT department, and certainly, not every company needs SOA. You might think it would be better if your company integrated this or built a SOA for that, but if its main job is something other than information technology -- like, say, digging ditches -- you're probably wasting your time and energy. Get real about what your business needs and what it doesn't.
How do you do that? Simple. Just ask. It's crazy, I know, but you've got to accept that you're not running the business and you may not know as much as you think you do. And even if you're right about what the company should be doing, nobody likes a know-it-all.
Bola Rotibi, principal analyst at MWD Advisors, said IT leaders shouldn't assume the business needs the latest integration technologies, according to the article:
Rotibi suggests that IT decision-makers speak to users and business managers about their day-to-day pain points and work on integration projects to reduce these. This way, it is easier to measure before and after to assess whether the project has had a net benefit to the organisation.
If that works out well, make sure you publicize the results to the board. And by "publicize," I really mean brag about it constantly and loudly every chance you get. This might help you fund more projects. If it doesn't, well, as the article notes, at least you can say it didn't cost much to try.