Q&A — State of INFORMS: Strategic & Strong

President-Elect Terry Harrison upbeat on the eve of his term in the top spot.

Incoming INFORMS President Terry Harrison (right), shown with INFORMS Executive Director Melissa Moore, sees a strategically strong future for INFORMS.

Incoming INFORMS President Terry Harrison (right), shown with INFORMS Executive Director Melissa Moore, sees a strategically strong future for INFORMS.

By Peter Horner

Life is full of twists and turns. Just ask incoming INFORMS President Terry Harrison, who officially takes over the Institute’s reins as president on Jan. 1, 2012. Flash back 35 years, to 1977, when Harrison – happily married and a freshly minted graduate of Penn State University with a bachelor’s degree in Forest Science – landed a job as a forester/analyst with Pittsburgh-based Koppers Co. Harrison, then 21, seemed to have everything going for him, yet he felt something was missing.

“It was September, I was on loan to the Tennessee Valley Authority, we had just moved to Knoxville and for some reason I had this bad feeling about myself,” Harrison recalls. “It was a small depression, something I couldn’t quite put my finger on. I was talking to my wife, and I said, ‘You know what? I miss learning things. I miss being in school.’ It was September and I needed to be back in school.”

Harrison talked to his employer in Pittsburgh as well as the folks he worked with in Knoxville, and he told them he wanted to start taking courses in his spare time under an employee benefits program. “I was so desperate,” Harrison continues, “I went to the University of Tennessee and took an accounting course just to learn something.”
As an undergraduate, Harrison had studied linear programming and optimization as part of his forestry major, and the concepts captured his imagination. At Tennessee, Harrison wandered into the Management Science Department, shopping around for some graduate work but not sure what he really wanted to do, when he happened to cross paths with a young professor named Rick Rosenthal. The chance meeting would change Harrison’s life and career in ways he could not have imagined.

“Evidently, Rick saw something in me that I did not see in myself,” Harrison says. “He encouraged me to get into the Ph.D. program. Within a few months, he had me working on an exciting project with him. He made me a co-author on a paper he was going to deliver, and then he invited me to go along for the presentation. Rick was just a superb friend and mentor, and I had the good fortune to meet him at a critical point in my life.”

Harrison went on to earn his Ph.D. in management science in five years while simultaneously working full-time for Koppers. In 1982, he returned to his alma mater to teach management science, and he’s been at Penn State ever since. Today, Harrison is a professor of Supply Chain and Information Systems at Penn State’s Smeal College of Business.

Rick Rosenthal, an active and acclaimed member of the O.R. community who taught for nearly 25 years at the Naval Postgraduate School following his stint at Tennessee, passed away in 2008, leaving behind a long legacy of grateful students, including Harrison. “I would not be where I am today if not for Rick,” Harrison adds. “He was the kind of professor I’ve always hoped to be.”

Speaking of twists and turns, when he’s not teaching, conducting research, consulting or trying to help move INFORMS forward, Harrison enjoys riding his Yamaha V Star 1300 motorcycle on the twisty, hilly, picturesque roads near his central Pennsylvania home. Debbie, his wife of 34 years, often goes along for the ride. “The thing I like most about riding a motorcycle is if you’re going to be safe, you have to pay attention,” Harrison says. “When I’m riding a motorcycle I am concentrating probably as deeply as I concentrate on anything.”

At the INFORMS Annual Meeting in Charlotte in November, Harrison was concentrating on the Institute’s business, its ambitious goals and the board’s strategic plan to achieve them when he sat down with OR/MS Today Editor Peter Horner for the following interview.

Let’s start by talking about strategic planning, a focus of the INFORMS board and a top priority of yours. Why is it so important and where does it stand right now?

I’ve been on the board for the past six years. I served five years as vice president of publications prior to becoming president-elect. During that period, strategic planning was always of interest to the board, but there were also other issues that seemed to keep us from strategic planning in a meaningful way. When we had a recent change of leadership, the board used that as an opportunity to say that, Yes, as part of this we need to find a new leader, a new executive director, who will have experience in strategic planning and lead us through a strategic plan. We have done that, starting this summer with a board meeting that was focused very heavily on strategic planning.

Now why is it important? It’s important because we need to be looking at what’s happening now and planning for two and five and 10 years down the road. Without it, you end up solving lots of little tactical problems of where to spend money this year and maybe next year, but without thinking about how to spend money in a way that it enhances the organization for the long term. How do we catch the big trends and satisfy the needs of the membership well beyond the time that we will be on the board.

Where is INFORMS now in terms of having a strategic plan? Do you ever complete a strategic plan or is it an ongoing exercise?

It’s a living document. We went through a strategic planning exercise. From that exercise, it was quite surprising, and pleasantly so, to see that the board coalesced around a couple of key issues. There really wasn’t much dissention about what we needed to do long term. Again, Melissa [Melissa Moore, who was named executive director last spring] led us through a strategic planning exercise this summer, and it ended up that we developed a couple of strategic directions. One, of course, is analytics, something that we had been working on, and another dealt with how we work internally, how we face the outside world, how we connect with social media. Just a general collection of things with IT at the core, but things that touch everyone in INFORMS in various sorts of ways.

I understand the strategic plan includes four specific categories: certification, continuing education, outreach and the IT initiative you just mentioned. Does that sound right?

That is exactly right.

Let’s address those four topics one at a time, starting with certification, which is potentially a huge initiative in terms of both risk and reward.

We started thinking about certification, primarily led by Jack Levis, Anne Robinson and Gary Bennett, about two years ago. We contracted with Capgemini to do a study about analytics in general, and certification was one of the action items we could take out of it. Certification is a large program; for us it will be at least a three-year endeavor to get up and running and be fully functional. It will cost some money, too. What it does, though, is establish us as the premier organization for analytics.

There is no standard certification group out there right now. Like most certification programs, the market leader declares what the standards will be. As of now, we’ve studied certification, we’ve gone through a couple of rounds of review of cost, and we’ve hired a consultant to help us understand what the various aspects are that we need to do to do it well. We’ve formed a committee that’s worked with that consultant. We are now to the point of identifying subject-matter experts so we can create a body of knowledge. Part of a certification program is the development of a body of knowledge that is used as part of a certification process. We aim in the next year or two to start offering certification, subject to some additional board scrutiny. The plan is that we will start with a certification called “certified analytics professional.”

What would it mean? What knowledge would that person need to demonstrate?

A certified analytics professional, as it’s developing right now, is a person who can demonstrate an amount of knowledge relative to analytics and its use and would have an understanding of the contents of the body of knowledge that is being developed. In this case, the person would come with a resume showing that they have been working in this area for a certain period of years and that they are already a professional of sorts. The certification would then just validate that they are a person who has a certain set of skills in this area.

Will there be a test?

There will be a test. It will be a two-part process. In the first part, the person will submit a resume that says here’s my background and here’s the things I have done. That resume will then be screened. In addition, folks who go through that screening process will take a test. Initially, it will be offered at a limited number of locations. The locations will increase along with the roll out of the program. Those passing the test will be a certified analytics professional.

How does certification benefit those members of INFORMS who choose to go through the process?

We have talked to a number of firms as did Capgemini. You can think of certification here just like you would in any other venue: It puts a mark on a person that says there’s a certain amount of experience and knowledge that they bring when they become part of a project. That’s attractive and over time, as that certification becomes better known, it becomes much more valuable in terms of identifying here’s someone who can do the kind of things that I’m looking for.

In the long run I expect that we will have multiple certifications assuming this one goes well. This first one would be a general certification as an analytics professional. Then we might adopt, for example, more specialized certifications that would build upon this general certification. So you might be certified in Web analytics or you might be certified in visualization or prescriptive analytics or predictive analytics. But for now it’s a very across-the-board generalized certification.

It’s a big program that will require considerable initial investment.

Yes. At the same time, the market seems to be quite attractive, so as a result, we expect that it will do well financially and not take too long to reach a point where it’s paid for all the costs of development.

I can see some dovetailing between certification and continuing education.

First let me say that continuing education is a little behind certification, and that was by design because certification was driving a lot of our activities. We want to stay focused on that. One can become certified without taking any courses, so it’s not necessary that you go through a sequence of one or more courses to prepare yourself for the certification exam. Many folks who are out there now will be ready to take the test based on their experience.

But we also realize that there is a need for continuing education, so we have a committee that is starting to look at that. In fact, we are going to offer two, mini, one-day courses at the spring meeting. One will be on soft skills; it’s something that we have done before and it’s popular with many practitioners. Another will go right to the sweet spot of analytics. We have not selected the topic or the speaker yet; the board just approved it and allocated the money to do hold it at the spring conference in April.

Can you describe the nature of the course and the sweet spot?

As I said, we haven’t yet decided on the topic, but that will happen soon, certainly before the end of the year. If you think of analytics being broken down into three areas as we have consistently branded it within INFORMS, there’s descriptive, what is happening; predictive, what will happen; and prescriptive, what should happen. The course will probably be at the high-end toward prescriptive analytics because that’s where INFORMS right now has its strengths, but over time, depending on how the board views the success of what we are doing, we may move down that chain and make courses and certifications available in those areas.

When an initiative such as certification or continuing education becomes a part of the strategic plan, does that mean funding is set aside to develop those programs?


Outreach is another big part of the plan and it includes many components. Can you summarize what the board hopes to achieve in that area?

You’re right, many things are in there that affect just about every area of INFORMS. We want to develop a better “inreach”: a better way to reach our current members, a better way to communicate, and that in turn ties into social media and our IT structure.

We also want to reach out to what we think are many potential INFORMS members who just don’t know it yet. They don’t know the value they would gain from becoming affiliated with us. An example of that is the practice meeting last spring; it was rebranded and re-oriented on the theme of analytics and we had success beyond what anyone expected. Attendance set an all-time record – 30 percent higher than the highest we had ever had at any of the practice meetings. I spent some time talking to some of those high-level people – the director of a large investment firm, for example, and some others who had never attended an INFORMS meeting. One person in particular had a Ph.D. in O.R. yet had never come to an INFORMS meeting. Making the conference about analytics, which that person viewed as quite distinct from O.R., and suddenly it went from being ignored to being very relevant.

What we are trying to do with outreach is to make sure that the folks who have the profile that fits what we are doing with analytics know about us and are able to be easily accommodated in INFORMS. What do I mean by that? Well, we have a huge meeting here that attracted well over 4,000 people. It can be hard to navigate. But we have a new analytics section, and one of the things that the new analytics section has done is build a series of tracks that creates a meeting within a meeting for a new analytics professional. They can come in and get value in the analytics tracks, and, if at some point they want to venture into areas beyond analytics, they’ll see amazing things, but they can always come in and experience a meeting within a meeting just tailored for them.

As the editor of Analytics magazine, I’ve encountered analysts from all over the world who have never heard of INFORMS. They’re craving a professional connection and outlet if not a home, and many attended the spring “analytics” conference, but they don’t recognize the value in joining INFORMS.

We are slowly making progress on that front. For example, two of the board members, Anne Robinson and Jack Levis, have been very active in attending the other analytics events out there such as Predictive Analytics World and some others. They note that in the beginning, people would come up to our booth and they didn’t know INFORMS. There was no recognition whatsoever. Now, a year later, because of the many activities that we’ve undertaken, INFORMS is becoming increasingly known. Our goal, our vision, is to become the premier advanced analytics organization in the world. We’ve made steps in that direction. Companies that are highly associated with these and other analytics events now seek us out. We have some cross-promotional agreements where they come to our meetings and we go to theirs. The goal is to make sure that INFORMS becomes increasingly known for advanced analytics practice and theory.

I’ve been to a few of these non-INFORMS analytics conferences myself, and while there’s clearly some common ground, the two worlds have not quite yet meshed.

I think of it like this: INFORMS is working well, our journals are strong, our meetings have been growing in attendance, our membership has increased this year after 10 years of decline. By and large things have been very good at INFORMS, but we’ve been looking inside and growing because people find us. The way I think of it is this: We have an extended family out there that doesn’t really know about us and we don’t know about them, and now we are finding out to our mutual benefit that we ought to get together and have a family reunion so everybody profits from it.

The last piece of the strategic plan is IT. What’s going on there?

Our systems function, but they have been far from state of the art. We have not had a good strategy for how to use social media to improve the member experience at INFORMS. We now have that strategy. We have a VP of IT, Bjarni Kristiansson, who is filled with ideas, a decades-long INFORMS member, who is very much aware of the state of the art. We are in the process of replacing an open position in the IT director role at the INFORMS office, and we will probably have that person in place by the first of the year. I think you’ll see by the end of the first half of 2012 a whole slew of improvements and new features that will be available to members.

You mentioned an uptick in membership this year after a decade-long slow leak.

I’ve studied this both before I ran for office and since. I can tell you this: Since 2000, we lost about a thousand members, and it was fairly consistent at about 100 members per year. The question in my mind was this: We have academic prowess. We have people who are among the best in the world at what they do. We have a strong practitioner base, and yet our message wasn’t getting out to the point where it was being embraced as it could be by others outside of us. Why? To sit on a plane and have someone ask you what you do – “I do operations research” – was always a conversation that took a lot of explaining. At the same time, I attribute a lot of our membership leakage to what was going on in other societies – a widespread loss of membership.

The INFORMS board, with this growing interest in analytics, decided to embrace it and go forward and did the risky thing of rebranding and changing some of the content of the spring practice meeting. That, combined with a couple of other things, pushed up our membership more than 700 persons this year. That’s a 7 percent increase after we had been dropping for a decade. We also have a significant increase in retention. So the picture both ways is very attractive, and we expect it to do nothing but improve as we expand INFORMS’ footprint into the analytics area.

INFORMS wants to drive more membership by creating member-only “value,” for example, putting online content of OR/MS Today behind a password-protected wall. On the other hand, INFORMS wants to boost its outreach efforts and one way to do that is to make content open and accessible to non-members. How do you reconcile these two objectives?

That’s a very timely question and it’s completely accurate. At this board meeting a few days ago we talked through the issue of creating visibility. In order to do that, we’ve had to put a lot of content – very valuable content – out in a way that people who don’t know us see us and recognize this content as having value. Yet we have a new analytics section that has attracted 500 members in less than a year, and they have materials they want to keep behind a password-protected gate so there’s an incentive to become a member.

At this point, we know the problem you mentioned exists. I think it’s healthy trouble to have, and we have just appointed a subgroup of the board to look at developing an overall analytics strategy. We started the analytics movement with a couple of different initiatives in a few areas. Now those initiatives have become successful, every one of them, they are growing, and it’s time for us to think about an overall strategy for INFORMS so that we know how we want to manage growth in each of these areas in a way that complements and enhances INFORMS overall.

It’s complicated because action you take to enhance membership growth could have unwanted consequences on outreach, and vice versa.

That’s absolutely true. At the winter board meeting in January, [President-Elect] Anne Robinson and a number of others are going to present to the board what they see as strategy going forward for managing the overall analytics efforts of INFORMS in a way that satisfies the needs of our internal constituents but still attracts others to INFORMS’ message of analytics.

Surveys and anecdotal evidence suggest the vast majority of INFORMS members support the analytics initiatives, but pockets of skepticism or outright resistance still remain, primarily from older members who want to hold on to their O.R. identity and don’t want to get lost in some nebulous “analytics” movement. Not to beat a dead horse, but …

We have beaten this thing for 25 years. I’ve been in INFORMS 30 years, so I’ve seen a lot of trends, a lot of programs, and I’m more exited about this one than just about anything else we’ve done over that time. Part of it is because the results are immediate and they seem to be very tangible. On the other hand, I’ve also had folks tell me directly, “I like INFORMS the way that it is. It works well for me. Why are we doing these other things?”

INFORMS works particularly well for an academic. We publish journals and provide a place where research can be certified as high quality through our journals and the refereeing process. We provide conferences, a forum to present research and to network with others. INFORMS works very well for an academic, and some worry that that will be lost.

My answer is, “I assure you it will not.” Now and in the future, publications will continue to play the important roll it has always played within INFORMS. In fact, we have a new journal, Service Science, coming online the first of January to add to our collection. When I was VP of Pubs for five years, I had three criteria that we would follow if we were going to adopt a new journal. One was that it had to become an A-level journal within a reasonable period of time. We want to publish the journals that people want to put the best research in. We will continue to do that.

Another way of looking at it is, what if we are so successful with analytics that we become a society of 30,000 people and the majority of the members didn’t join as academic members focused on O.R., but they joined as practitioners focused on analytics? Again, we would be lucky to have this problem. I don’t see that happening so quickly that we can’t grow into it. That would certainly be a question for a board down the road. The real value is the opportunity for INFORMS to be recognized nationally and beyond as the premier advanced analytics association; that we are recognized as the place to go to be certified as an analytics professional; that people will be coming to us in a way that will enhance the experience of both the analytics professional and the traditional O.R. type, particularly the academic O.R. type.

In your candidate statement, you listed making INFORMS “more entrepreneurial” a priority. What do you mean by that?

I think that we should look for ways to see how we can improve across the board and do some experiments. Some of them will work and some won’t. In the process we will find ways to better serve our members, to expand our reach and to present our research more widely. We have a collection of brilliant people, and pooling all those people together – thinking about how can INFORMS change in a way that makes it better – to me is just too good to overlook. I would like to encourage us from the board down to think about ways that we can do things, try some things. The ones that don’t work, so be it. The one’s that do, let’s support them.

INFORMS is often described as a three-legged stool, supported by its journals, conferences and subdivisions. We’ve talked about journals and conferences, but not so much about subdivisions with the notable exception of the new Analytics Section of INFORMS. How important are subdivisions to the success of INFORMS?

Subdivisions are the heart of INFORMS. People who are involved in subdivisions are much, much more likely to be engaged with INFORMS over a longer period of time. They are more likely to be satisfied, they are more likely to contribute, and so the board is looking at ways to encourage more folks to join a subdivision. That’s very important to us.

How is INFORMS doing financially?

In 2008, when the economy started to weaken across the board, INFORMS felt that in a lot of ways. We had a budget that was built on an expectation of an economy that no longer existed. A year ago we went through a very painful process of realigning what we were spending with what was coming in. We had to cut a million dollars out of an $8 million budget. That’s a lot. Jack Levis led us through a series of budgeting exercises. We asked staff to help us find cuts and places we could be more efficient. We made a balanced budget last year, as painful as it was, and we went forward.

Now we are a year hence, and we’re going to run a half-million dollar surplus this year because we have cut our costs. Things are doing better than we had hoped. We’re back to the point where INFORMS was four or five years ago when we were running surpluses every year and we had money to do things. We are in very sound shape. We have a substantial working capital account. We have a permanent reserve that we have saved over the past 10 years that has a substantial amount of money relative to our operating expenses. We are in great shape financially, and as a result we have some flexibility to take on some new initiatives that we think in the long run will make us even more financially sound, improve the organization, improve our outreach and allow us to do the things we are charged to do by our constitution – the dissemination of knowledge around operations research and the management sciences.

Given all that we’ve talked about this morning, can you provide a brief State of INFORMS from your standpoint?

I am delighted to be sitting here talking right now about INFORMS because I feel so good about it. I’m proud of the efforts of so many people that really changed things fundamentally in a positive way over the recent past. The State of INFORMS is strong financially, it’s strong professionally, our journals are strong, our meetings are strong, our membership is increasing. I can’t think of a single major area of INFORMS that hasn’t been addressed and that hasn’t responded positively. When you interview Anne Robinson, the next president, a year from now, hopefully she will be saying the same thing. My bet is she will be telling you an even better story. Right now, I feel really good about what INFORMS is and what it’s going to become.

Peter Horner (horner@lionhrtpub.com) is the editor of OR/MS Today and Analytics magazines.