Measure of Alignment Predicts Success

By Edward H. Baker  |  Posted 10-15-2005 Print Email
Kay Lewis Redditt's reasons for getting into the business of measuring IT alignment were directly connected with her frustration with misaligned information technology.
Kay Lewis Redditt's reasons for getting into the business of measuring IT alignment were directly connected with her frustration with misaligned information technology. "I've been in the private sector most of my career," she says (she held positions at both American Express Co. and Macmillan Publishing). "I figured out early on that my business success was absolutely linked to my ability to get what I needed out of IT."

In 1983, Redditt joined forces with Thomas M. Lodahl, a professor of management at Cornell University and the Massachusetts Institute of Technology, and a vice president at the Diebold Group Inc., to found CogniTech Services Corp., a Warrenton, Va., IT consultancy, on the basis of two related assumptions: that businesses were going to become more and more dependent on IT, and that IT was being managed incorrectly.

"IT was being managed on the basis of cost, and it needed to be managed on the basis of what it could contribute to the business," says Redditt, who is currently president of CogniTech (Lodahl is chairman). "That really started us thinking that we needed to work on a concept of IT effectiveness rather than IT efficiency."

Early on, the two decided they needed to be able to quantify how effectively IT was being employed at any particular company. After several years of research, they worked out a metric they call "I.S. Contribution," which directly measures how well IT is aligned with business goals.

It places companies on a scale of 1 to 100—although, in the real world, companies range from the 30s (the poorly aligned companies) to the 70s (the well-aligned ones). Their database now includes more than 160 organizations from around the world, both public and private, for-profit and not-for-profit.

Recently, Ellen Pearlman and Edward Baker of CIO Insight asked Lodahl and Redditt how they measure alignment, and what they've learned about how to get and stay aligned from the hundreds of organizations they've studied. An edited transcript of their conversation follows.

CIO Insight: Your alignment improvement model includes four areas: functionality and quality of IT, customer satisfaction, the organization's culture, and planning and interaction between IT and business. Are these four components equally weighted?

Redditt: No. The most important is planning interaction. That's how I.S. Contribution is improved. The correlation between planning interaction and I.S. Contribution is .71. So you've got to get IT and the business guys to the table, so they can jointly determine IT projects, and the order in which they will be funded and resourced.

But in order to get them to the table—and this is why it's called an improvement model—the No. 1 factor is credibility. If your credibility is good, then there's no real block to implementing planning interaction—getting the two parties together and using goals as criteria to help prioritize resource allocation.

If credibility is low, then you've got to fix credibility first, and the place you go to fix credibility is customer satisfaction. We have six factors within customer satisfaction, and they all have a very high correlation with credibility.

However, a few of those factors are more important than others. No. 1 on the hit parade is what we call reliability of service. It basically means that IT has to be like the telephone company and maintain a dial tone. Unless there is an acceptable standard of service on uptime, response time, security—all those basic operational factors—IT credibility goes absolutely into the tank.

The next most important is quality of output—how good is the information that comes out of the systems? When that's low, then credibility is low, simply because the business guys aren't getting the information they need.

That's where the next step, functionality and quality, comes in. When quality of output scores low in customer satisfaction, the functionality and quality metric will help the IT organization determine where the problems are for each business unit. It might be different for marketing than it is for finance, for instance. But it's the systems that are causing the problem.

The last thing, and the least important, is the organizational culture. Sometimes, even when IT credibility is good and customer satisfaction is good, the culture within the organization is one that promotes a very clannish kind of behavior. Business units deal only with their little set of people when it comes to decision-making, and they don't want to let IT in.

So when you have cultural issues, you're not going to change the culture per se. What you want to do is to identify where the problems are and build big bridges across the cultural divide.

That's how the model works together. You're attempting to improve I.S. Contribution, because I.S. Contribution, when you improve it, improves profit margin. And the only way to do that is by implementing planning interaction. So you have to determine if there are problems precluding the implementation of planning interaction—and if so, what kind of problems they are, and then address them directly.

What are the traits of the top performers?

Redditt: We have observed that the high performers, our top 10 percent on IT contribution, do not necessarily score the highest in customer satisfaction. Why isn't customer satisfaction as important to them as alignment and I.S. Contribution? What they told us was that they're not in the happiness business. They're in the business of doing what the business needs to do in order for the business to be successful. That's why we established the credibility measure. It is very important for IT departments to make sure customer satisfaction is at least adequate, so that customers know that if they ask for something, IT can do it. In other words, IT has to maintain its credibility, but satisfaction by itself shouldn't be the primary goal. Without credibility, IT doesn't get the money it needs in order to do the projects that support the business goals, because the business managers do not sufficiently believe that they can successfully complete those projects.

So when we measure with our set of metrics, one of the most critical things to look at is scores on IT credibility. If credibility is too low, the business guys aren't going to come to the table to try and work out the problems that exist. They aren't going to communicate with IT about their goals and objectives, or about the projects they're working on, so that IT can get aligned with the business.

Is culture ultimately the toughest hurdle here, the hardest problem to fix?

Redditt: Yes.

What's the worst example of cultural dysfunction you've run across?

Lodahl: One of the principles we've learned over the years is that, as far as governance goes, the most effective organizations are ones in which business basically decides the what and IT decides the how. When that's not happening, there's bound to be trouble.

At one of the old RBOCs, the IT strategic plan was done at great cost and with great effort, but without referring to the business planning process at all. They just did what they thought was the right thing without even talking to the business guys. And their contribution score was so low that we warned the CIO that unless this changed, there was going to be a revolution. And he was gone within six months.

Why is it so hard for companies to stay in alignment?

Lodahl: One big problem is that business goals change much more rapidly than you might imagine. So alignment has to be a continuous process. You can't wait for the yearly scrum to realign. Most of our high performers will realign either semiannually or quarterly—and some of them even monthly—by talking about how business goals are changing, and how IT needs to change its project portfolio to meet those new goals.

How important is it for everyone in an organization to be aligned?

Lodahl: Very important. We find that if you look at senior management's goals—say the top eight guys in the company—versus those of the operating managers, those goals are often different. Sometimes the steering wheel isn't connected to the rudder. So if all you do is interview the top eight guys and find out what their goals are, and align to those, you won't be doing the company any favors. And generally, when that's the case, the I.S. Contribution scores are very low, because IT doesn't know who to serve.

But this gives the CIO an opportunity. He or she can go right to the senior management group and show them how their goals are not the same as those of the operating guys. And whose problem is that? It's not the CIO's problem. It's the senior guys not communicating well with their operating managers.

Redditt: Let me give you an example: We were working with a company called PECO Energy before it merged with Unicom Corp., and we discovered a complete divergence between senior-manager goals and the operating-manager goals in one of their units, called the ventures unit.

The unit had been put together to identify and develop new product lines. But the company's senior managers put a group of old-line power-generation guys in charge of the new division. So it was no real surprise to discover that the goals of the ventures group had nothing to do with actually generating new revenue. Their goals were old-line power generation goals: increase efficiency, decrease costs, that kind of thing.

On the basis of that data, senior management put a whole new set of business managers into that group. And when we remeasured about 14 months later, the I.S. Contribution of that group had improved by 13 points. And the goals of that group and the goals of the senior managers were very much in sync.

Does it matter whether IT is "strategic"?

Lodahl: Our data strongly suggests that big improvements in I.S. Contribution are more likely to come from a whole bunch of

small IT projects that support high-priority goals and get you aligned with day-to-day needs, rather than from the big blockbuster projects that might soak up all the IT resources.

If you're a strategic IT group looking for the big hit—the huge strategic bullet that's going to solve all the company's problems—chances are you will not do as well as if you paid attention to the day-to-day.

Redditt: We're not saying you don't do the big projects. But here's an example of the problem: We measured a U.K.-based company called the BOC Group over a period of years and they improved very nicely. Then one year the bottom dropped out. So we said, "What in the world have you been doing?" And the answer was that they had been told to drop all projects except for a big ERP implementation.

Because ERP takes years to put in place, if you do nothing but implement ERP, then your I.S. Contribution and your profitability will drop. That's because the improvements that need to be done, the things that IT needs to be doing in order to support the business goals, need to go on. They can't stop supporting the day-to-day business.

Then is it an issue of how much companies spend on IT, or how much discretionary spending an IT department has?

Redditt: We have found that business units will individually fund projects that are necessary for them to meet their strategic goals. When IT has high credibility, when the business knows that if the money is invested, then IT will successfully deliver on the goals that are critical to their being successful—in those cases, there is typically no problem with discretionary funding.

In fact, we find that highly effective IT organizations don't spend the most on IT within their industries. It's not how much you spend, it's what you spend it on. Of all the organizations we've looked at, we have found only two where we had to tell them they weren't spending enough on IT.

Over the years that you've been doing this, has alignment been getting better overall?

Redditt: I would love to say yes, but the answer is no. I think the problem is that IT projects are usually identified through a system by which the business guys send in a request and the IT guys take all the requests and attempt to prioritize them. And the business guys don't ask for the right things, because they aren't thinking about their goals when they're asking for IT projects.

Lodahl: In other words, the governance process is still broken. Companies where the main method of deciding IT budgets is through the assistance-request process are among the lowest performers. Where it's a collaborative process of planning interaction, it tends to be higher.

Business should not be coming to IT and saying, "I want X system." Business should be coming to IT and saying, "Here are my problems. Here's what I have to accomplish. These are my goals." And IT should be saying, "Okay, here's how we do it." They should sit down together. IT is very good at being creative and innovative about what IT can do to help the business units achieve their goals. That's why it's a joint function.

For the past five years or so, companies have been sharply focused on cost-cutting. And now there's more of a focus on innovation and growth. Does your data show anything about how hard it is to change gears and get IT and the business to refocus on growth strategies?

Redditt: We just ended a cycle where cost was the most important thing. And that was the strategic direction of the business. But now we're entering a cycle where the focus is on revenue growth, competitive advantage and improved quality. It's much harder for IT in that domain, because business doesn't know what to ask for. Consequently, IT keeps going in the same old direction and alignment gets worse and worse.

Lodahl: It's clear that revenue-growth goals are highly under-supported, whereas cost-cutting goals—everybody knows how to use IT to cut costs. Not many companies have figured out how to use IT to increase revenues. I think the reason we have CRM systems is because companies are recognizing this problem. Our research shows that the I.S. Contribution to revenue-growth goals—and to competitive-advantage goals, for that matter—is quite low. The problem is that not many IT resources go to revenue-growth projects, because business guys don't know how to ask for IT projects that are going to help increase revenue.

So again, you come back to the problem of the request mechanism. If the IT department only does the things that business asks it to do, it is limited by what business thinks IT's capabilities are.

Redditt: If the business is moving toward growth, or revenue improvement, and IT is only supporting cost-reduction goals, then I.S. Contribution scores will be low, and profit margin will go down. Either way, the correlation is strong.

Lodahl: The point is that profit margin is a great measure of performance, business performance, because it deals both with revenue growth and cost reduction. We measure both of those in the omnibus measure of I.S. Contribution.

So are we potentially entering an era in which alignment will suffer?

Lodahl: Well, I think that whenever there's a sea change in the business climate or business goals, the energy it takes to change direction is big. Remember that IT portfolios generally run over a two- to three-year period. But business goals can shift within a two-month period. So you have to be willing to dump your favorite projects if they're no longer relevant to where the company's going, as well as invent new ones that are going to use some of that same money.

If you have a planning interaction process in place, one where IT sits down frequently with the business and says, "What's on your mind this week?" or this month or whatever, then you're much more likely to be able to stay abreast of where the business is going. And you'll also be better able to figure out what IT ought to be doing to help it get there.



 

Submit a Comment

Loading Comments...