TechTalks Transcript
Analyzing and Costing IT Services
 Judith Boettcher [JB] |
 Howard Strauss [HS] |
 Christopher Peebles [CP] |
March 2, 2000
Audio
• Streaming
MP3
• Download
MP3 (Download
Tips)
JB: Welcome to the CREN TechTalk series for spring of the new millennium, and to this session on Analyzing and Costing IT Services. You are here because it's time to discuss the core technologies for your future campus.
This is Judith Boettcher, your CREN host for today, and I'm going to welcome the technology anchor for TechTalk, Howard Strauss of Princeton. Howard, as you know, is a well-known Web and now portal expert as well. And Howard, I'll let everyone know that you're doing this with just a bit of a cold today, but welcome.
HS: Okay, thank you, Judith.
I'm Howard Strauss with a cold here. You can probably tell that! I'm the technology anchor for the TechTalk series of CREN Webcasts. As technology anchor, I'll engage our guest experts -- in this case, one guest expert -- in a lively technical dialogue that will answer the questions you'd like answered, and ask those very important follow-up questions. You can ask our guest expert, Christopher S. Peebles, your own questions by sending e-mail to expert@cren.net at any time during this Webcast. If we don't get to your questions during the Webcast, we'll provide answers in the Webcast archive.
New York's former mayor, Ed Koch, would often ask, "How'm I doing?" This is a question that IT folks often ask, but rarely seriously answer. Things are changing so fast, though, it often seems overwhelming just to try to offer services that are up to date. Measuring how well we offer those services is usually at best an afterthought. But if we don't know how we're doing, how can we ever get any better?
Well, we probably can't! What we often do is just listen for squeaky wheels. If some VP, dean or vocal faculty member complains to someone important in our IT organization, then we know we're not doing so well. Normally, we just fix enough of the problem to quiet down the complainer. Even if we use appropriate techniques to measure the quality and value of services we offer, how do we know we are really measuring the right things?
Suppose we get zillions of questions -- you know, call-ins, e-mail, knowledge base, walk-ins, all kinds of things like that -- about creating PDF files for MS Word, and every one is answered courteously, quickly and correctly. This may be a sign of excellent service, but it may miss the point that the current twelve step procedure for doing this needs to be greatly simplified. How do we avoid providing our users with the equivalent of the Web's best vacuum tube when they will be better served by having us invent the transistor?
Once we know how we're doing, we need to know what it costs to do it. Our users have infinite demands, but we have finite budgets. We have to prioritize and control the costs of everything we do. What does it cost for you to answer a walk-in question compared to a call-in, and how does user satisfaction vary between the two?
If you had answers to questions such as that, you could better allocate your limited resources. Offering excellent service at the lowest possible cost cannot be done without the involvement and commitment of everyone in your IT organization. To get this kind of involvement and commitment, there has to be an appropriate IT culture. This culture must be appropriate for IT folks without scaring off your users. In fact, the culture should be welcoming to your users.
Few companies have a stronger culture of service than Walt Disney. Its fourteen-word mission statement that defines its culture is, "We create happiness by providing the finest in entertainment for people of all ages." Is your mission statement that clear? Spending time on these costing and assessment issues may seem a little silly to you when there are so many pressing technical issues. But Disney has made these issues the hallmark of their company's success -- not the technical details of the rides or animation.
Chris, our expert today, is an anthropologist who knows the ways of mice and men. He'll give us some insight about how we might better imitate Disney without feeling Goofy on today's Webcast of TechTalk.
Judith?
JB: Thank you very much, Howard, and I'll move right on to introducing our guest expert today, Chris Peebles from Indiana. As Howard mentioned, Chris is an anthropologist by training, and (as he likes to say) an information technologist by happenstance. Chris currently serves Indiana University in many capacities that I will leave you to search out on the Web. For now, the most important for our talk today is that he serves as Associate Vice President for Research in Academic Computing, and Dean for Information Technology -- and those are only two of some of his many titles.
Welcome, Chris, so glad that you can be here! Your training as an anthropologist must bring an interesting perspective to information technology.
CP: Judith, it does. I'm interested in the efficacy of organizations, both current and past, and I got into information technology 45 years ago in the Air Force, courtesy of, I guess, avoiding going into the Army! And --
HS: Sorry about that!
CP: Oh, it was the Cold War and all that.
HS: There must be folks out there that have been in the Army.
CP: Four years was longer than two, but better than the army.
And as I became an anthropologist, I kept up my computing, but my anthropology was devoted to the development of organizations, formal and informal, primitive and whatever we can call ourselves -- other than primitive or something. And when I came back into IT full time, I started looking at the relationships between organizational structure and strategy on the one hand, and the pair of those in relation to organizational culture on the other. And that led me to contemplating problems of quality and problems of value, and eventually problems of cost. And that's how I got sort of where I am.
JB: Well, that sounds great, and I'm also very intrigued with Howard's comment about Disney creating happiness. That certainly isn't exactly what we try and do with IT, with our focus on service, but it certainly is --
HS: I think we actually do, Judith?
JB: Is that creating happiness?
HS: I think happy users are a very good sign.
JB: Actually, let's go with that.
HS: And I think that, Chris, maybe because we brought this up -- I was going to ask you a question first about costs, but let's talk really about how in an IT organization do we measure user satisfaction or user happiness?
CP: There are a couple of ways. Let me back up and say that our smallest campus is in Richmond, Indiana. It has a new CIO. He has a new motto for the computing organization, a new vision that says "We make computers scream so you don't have to!" That's a Disney-esque kind of approach to it.
But in surveying customer satisfaction, it is basically working with people who know how to ask other people questions to get an answer that, in fact, is truly measuring their satisfaction with your services.
HS: So you use professional people --
CP: Absolutely.
HS: -- to put together these surveys and tally the results. What kind of professional people? Are these people in your sociology department? Where do you get these folks from?
CP: At Indiana University on the Indianapolis and Bloomington campuses (the two largest campuses), we here in Bloomington use the Institute for Social Research, and we have over the last ten years. In Indianapolis, they use a part of their assessment organization that is in Vice Chancellor Trudy Dance's office. She is, in fact, an assessment expert. But they have a survey organization up there as well, and we use them both to make sure we are measuring what we think we're measuring, and we also use them to assure anonymity on the part of our respondents. And we, this year, look like we will get a 50% return on our survey, which is unheard of in most surveys that are conducted.
HS: When you say "your survey," I thought you measured all kinds of things and did all kinds of surveys. Could you tell us something about the things that you measure and the kind of surveys that you do?
CP: We have a yearly survey that has about 100 questions on it, some of them demographics, including, "Do you own a computer? Is it at home? And do you have a modem and are you going to drive us crazy dialing in?" And then we ask a number of questions about the services that we offer and they rate them on a one to five scale, one being "It sucks beyond words" ("sucks" being a technical term), and five being, "It's wonderful!" So we have the yearly survey.
JB: So the 50% return is on that particular yearly survey? And is that across your faculty, staff and students?
CP: That's separate surveys, slightly different in each case. Because there are some things that are relevant for graduate students that are not generally relative for undergraduate, and there are some things relevant for staff.
HS: And you send this out to all your users?
CP: We send it out to a sample of 2,000 on the Bloomington campus and 2,000 on the Indianapolis campus each year.
JB: And you have some historical data now, since you've been doing that. Did you say you were doing it for ten years?
CP: Ten years on Bloomington and four years at Indianapolis, when we combined the organizations on the two campuses.
HS: Are the questions on that survey available to other people who would like to see them?
CP: Yes, on the Web. The questions are there, the answers are there over ten years, and even the written answers are there over ten years, and that's on our Website and that's referenced, I hope, on your Web page.
JB: You know, I'm not certain that we have that particular one on our Web page but we will add it because I know I've got it in the other document you gave me, Chris.
HS: And you're confident enough so that you put out all the answers that people have provided.
CP: Yeah, as I said, there is one guy that usually -- or it could be a woman, I'm sorry. I'm being sexist! There's one that starts off every year, the written answer, "We think you should be neutered and locked away forever!" And then proceeds to detail our sins of the past year. Yeah, we publish not only the statistical answers but also publish the written answers.
HS: And these written answers, are you also asking your users to suggest ways you could do things better or new services you could provide, things like that?
CP: Absolutely. And these are our best leading indicators of what we ought to be doing, some of which we have not conceived of. So in those written answers, we analyze those, we look at them, we read them, we digest them very carefully and very fully because every year, there will be one or two themes that will suggest to us that there's a whole area of service that we ought to try to offer and do not now. So that's our yearly survey.
We have a daily survey.
HS: Daily?
CP: Daily. Several hundred -- several thousand on some days -- people call the Support Center, and hopefully, it is not (as we pointed out for the reasons you started with) that we have a bad process. We try to have the best processes in place we can, but people do call the Support Center.
It's a general Support Center, and there's seven lines of business going into that Support Center -- Intel, Macintosh, e-mail, various administrative processes, so you get right in to the right person. It's not a deep phone tree, it's only one layer deep. We choose five people from each of those seven lines of business each day, and we send them an e-mail and ask three questions:
- "Did you receive an answer to your question or a solution to your problem? Yes/No."
- "Did you receive this solution or answer in a timely manner? Yes/No. (You define what timely is.)"
- And third, "Were you treated with courtesy? Yes/No."
JB: Um-hum. Now, I was wondering when we were talking about this before, Chris. Who is the person and just how do you operationalize that type of daily survey? Do you have someone from the professional survey folks that manage that process?
CP: We do complete problem tracking to resolution in historical files, and we use that either to improve processes like don't have a 12-step PDF creation or to write a script for our knowledge base for areas that we can't, in fact, affect ourselves. And so from each day's problem tracking system, it just automatically chooses 35 -- it randomly chooses 35 people.
JB: So you've got this all automated, then?
CP: Absolutely.
HS: You're talking about measuring the quality of your Support Center. Are you measuring the quality of other things out there? I mean, a lot of universities, for example, do programming or Web page development for faculty and things.
CP: Yep.
HS: Are you doing that as well?
CP: We do it for all our services. We have narrowly defined services. Things like account administration. That's a real service. We either do things to your account or we establish it. I know what it costs and I know that 91.5% of the people who have accounts are satisfied with it.
We have database management support. That's a very specialized Support Center. Last year, they were contacted 12,020 times by users to help with database work. This is general users in the general community, not internal users who run the big Oracle and enterprise wide databases. We track that, we track that satisfaction.
We advice departmental support providers, so that's a service, and we measure that. We measure all kinds of weird things. We have a number of listserves. We measure satisfaction with that. Everybody used to love newsgroups but doesn't anymore, but then the total costs for newsgroups last year was only $57,000.
HS: Only! Okay.
CP: Web research computing -- we measured that. That's different from just general ordinary computing. Support center, telephone services, various telephone services, we measure satisfaction with that.
HS: Okay, I'm convinced! You do it!
CP: Yeah, of course, we measure quality, cost and try to measure value across every discrete service we own.
HS: It sounds like you come close to defining, you know, measuring how clean the buildings are and things.
JB: Yeah, I'm almost afraid to ask the question, Chris. Have you identified just how many of those discrete services have you, A) identified, and B) do you measure?
CP: We measure at the aggregate level 44 suites of services. And within some of them, we measure from one -- well, we measure at least one, but in some, we measure up to six services.
So let me give you an example. Student computing. We have student technology centers out there that my organization runs. It has about 1,500 machines. We measure cost and quality of hardware, software, printing and student consultants. So four services within the student technology centers.
HS: Okay. In some of the other areas, I don't know if this was on one of your Websites or one of your papers (I read a lot of stuff), but you said that IT organizations should offer as many clearly defined, highly visible entry points for user support and education as can be designed and maintained. A lot of people might say you should really only offer one. You want to talk about these clearly defined, highly visible entry points?
CP: Absolutely. This is one that may be more relevant to a university the size of Indiana University. We run a general support center. It takes several hundred thousand calls a year; the knowledge base which they write and maintain gets millions of hits a year. That is the first line of defense, and they can refer people.
But if, for example, you have a humanities computing problem, don't call them! You'd know enough to call the Letters Project in the library, which is one of our support organizations. If you're a departmental support provider -- and there are 219 of those on the Bloomington campus alone, who are paid for by departments or by schools or by centers. They're the technical support for the department of History, for example. You don't want them calling the support center. They know as much as the support center does. You want them calling directly into an organization that supports.
We have a statistics and mathematics center that we do in partnership with the graduate school and the Department of Mathematics. It works with graduate students and with faculty on dissertations and on the creation of courses. Their questions about SPSS and S+ and Mathematica couldn't be answered by the support center. Why bother the support center to roll the call over when everybody knows that if you've got those kinds of problems, you call the Stat-Math center.
Teaching and learning technology laboratories. That's a joint project -- a partnership between our office and the Provost here called the Dean of the Faculties -- to work with faculty and to work with students in the creation of courses. The technology organization takes care of the applications and the Dean of the Faculties' experts take care of the pedagogy and content end.
So multiple points of entry, not for confusion, but in fact, for effectiveness.
JB: Okay. In terms of the next question that comes, then, Chris, is how do people know where to go if you've got that many different points of entry? How have you addressed that problem?
CP: Well, lots of education, but we do tell everybody, your first line of defense if you're on campus is dial 5-6789, and that gets you the general support center. But we do work with incoming students. We have multiple ways of broadcasting news via the Web, via listserves, via print media. Most all of us in the organization spend a lot of our time facing outward, meeting with people, groups, students, faculty, staff.
JB: Okay, so basically, however -- people when they're fairly new to the campus have one place to go. And then once they've become a little bit more sophisticated, they kind of know where to go.
CP: Yes.
HS: And that one place will redirect you if that's the only thing you can think of.
CP: Yes. 5-6789 works, and I call it occasionally when I'm confused!
JB: We do have a question coming in that's closely related to what we were just talking about here, Chris, and it goes back to the fact of all the different suites of services that you are measuring. And the question is from David Lazner out in Hawaii, and he's asking, "Just what is the cost in terms of people and money of the measurement and assessment processes that you are running?"
CP: Let me talk just about the Bloomington campus. The Bloomington campus, 300 FTE; 300 part time; 36,000 students; 1,600 faculty; 2,500-3,000 staff. The activity-based costing you can probably do for that size enterprise, which is about a 35 million dollar enterprise with one FTE or less. The survey, from start to finish, spread across the Bloomington part of the organization will cost less than an FTE in its construction each year and less than half an FTE in its final analysis in our statistics and mathematics center. And we pay the survey research organization for the Bloomington campus a little less than $40,000 to create and administer the survey. So these are not backbreaking costs. I hope that's a full answer.
HS: Chris, you mentioned activity-based costing. Did I get that right?
CP: Yeah.
HS: Could you tell us something about that? What's activity-based costing? If you could give us an answer not designed for accountants.
CP: Yeah. Well, fortunately, activity-based costing is not designed for accountants because it goes way beyond the generally accepted accounting procedures, GAP, which is the dullest book in the world.
Basically, if you look at a service -- let's define our service as account administration. These are the people who do all the DNS and all the security and all that other stuff to keep your account safe, to make changes in it, to establish it initially. The costs of that are the people and the technology, which may be costed or may be depreciated -- things like the phones they have, things like supplies, applications that go on the hardware.
You basically take all of the resources that go into creating that service, whether costed as year over year or depreciated, add it all up and then divide it by your output. How many user IDs did you support? How many actions did you take on the accounts?
Last year, just for your own sort of amusement, almost 300,000 user IDs at Bloomington and about 60,000 human intervention actions on the account for a total cost of $218,000. Cost us about 77 cents a year to administer an account, and it costs us about $4.00 to change it.
Does that make sense to you? Take all the resources, add them up, and then divide by what it produced. And that's what's so important about this is what it produced. You know, I can say I spent "X" dollars per student per year, and that is meaningless unless I tell you what services I produced for those students.
HS: Another issue is what value those services are.
CP: That's a toughie!
HS: What?
CP: That is a tough one.
HS: Do you try to deal with that?
CP: Yeah, let's first argue that the quality is in the eyes of the beholder always. And we can measure it or define it as fitness for use and freedom from defect. But it's still in the eyes of the beholder -- rating it in our case on a five point scale.
Value, we ask about, but I don't consider that a measure of true value. It is a measure of how much that individual values that service. They like it or don't like it, it's important or not important.
Howard, I think the question you're trying to get at --
HS: Well, I'm really trying to get at the question, how do you clarify services? We can't do everything, so we ought to offer the services, I think, that offer the most value.
CP: Agreed.
HS: And so you know, if we could somehow figure that out, then we might know where we should put our resources.
CP: But I was going to go on and say, value in this case is in part of the eye of the faculty or the researcher.
HS: Or the student or whoever.
CP: Or the student or whoever. But the value, of course, for researchers, we can measure. Because you and I work with them every day. We know that there are things we can do with information technology that could not be done otherwise -- calculating the relativistic mass of the proton. That could not have been done other than with information technology. The doing the bioinformatics stuff, the sequencing of large genetic chains, DNA chains -- that can't be done otherwise.
The place where it is hard to measure value is, does information technology add to teaching and learning in ways that we can measure and understand and place in the context of how we see the mission of the university? That is an extremely hard question to answer.
HS: But even at a lower level, if we're trying to decide, for example, should we offer support to Linux or something, how do we decide if that's a good bet? I mean, we could decide we're going to do it because enough people are screaming about it.
CP: That's why we do it!
JB: Or enough people --
CP: In the case you gave, Howard, in the case of Linux, we do it because people are using it.
JB: And you get some indications of that trend by your yearly survey as well.
CP: Absolutely. And we have been supporting Linux for three years, by the way.
HS: Right. But given that you support it because some people are fussing about it or enough people are fussing about it, but then the question becomes, how much of your resources do you put into that? It seems like you have to decide what the value of that service is to make that decision.
CP: We don't consider that a service so much as a support issue. Do you see where I'm going with that?
HS: Yeah.
CP: Linux is not a service. Linux, for us, is a support issue because, in fact, it doesn't fit into the software that we provide.
But let's talk about why we signed an enterprise license agreement and another one with Corel. That was to get everybody on the same release of all the software so we could more easily support it. Not only have we saved the students lots of money in having an enterprise agreement, but we have saved huge support costs by making sure that our expertise is focused on what is being used rather than people using a little bit of everything.
So there's a way of managing costs by reducing diversity.
HS: Diversity of software, of course! Not students here.
JB: I do have another question that might be good to bring in right now, and it's from Mike Berman at Rowan University. And he links back to, I think, a real important part of how you've got your campus set up, Chris. And that is -- well, let me just read the question here. "With over 200 department-based support people, how do you inculcate your service ethic to such a widely distributed number of people who may have objectives and agendas that are not always the same as yours?" So obviously, how do you get that culture out to this large number of support folks?
CP: Okay, first of all, those 219 people in departments do not work for University Information Technology Services.
JB: They work directly for the departments, then?
CP: They work directly for the department.
JB: Okay.
CP: The partial answer to his question is "happy circumstance." Most of those people started off working for UITS as a cluster consultant, 20 hours a week as a student. They decided they liked Bloomington, they graduated, they may have applied for a job inside our organization and then, to move up, they may have applied for a job in the School of Health, Physical Education and Recreation. And after working there for awhile, they may have come back to University Information Technology Services as a manager of GIS systems. So in part, they learned the culture because they've been a part of it, off and on, in and out.
The other thing is that we have a support organization inside UITS -- University Information Technology Services -- that does nothing but support those people. So it helps the service ethic in the relationship between the support organization that supports local support providers and the technology and all of the special things that get done just for those support providers. We have software servers up for productivity software for them that nobody, including me, gets. These are all kinds of tools to help them deliver their service to their constituents.
JB: Well, then, you said that you had a team of folks within your organization that is tasked with that specific service of supporting those folks. People may want to know how big that is.
CP: The distributed support organization is about eight people and it costs about -- you've got the paper there, Judith. I don't have it in front of me. It costs us -- well, I can go down it.
It costs us $2,000 a local support provider for that advising, $319 a provider for the specialized software, and about another thousand dollars a year per department for services we provide directly to them, usually technology assessment. So we do it with about eight people.
JB: Okay. Let me just remind everyone to go ahead, that they can send in e-mail questions to expert@cren.net. And while people are writing questions, Chris, it might be good to go back and answer online that question that came in earlier, asking about the return on investment for IT for faculty work practices.
CP: This was a question that came in to me this morning, and so not thinking that it was coming in to me supposedly this afternoon, I just wrote back a very short -- and I hope informative -- answer.
Basically, the question was, how do we do return on investment? How do we rate IT investment in terms of faculty work and value? I can cost it. I can do the activity-based costing and my colleagues and I can talk about the "investment." But it's a faculty issue about return, and a faculty issue about value.
We try and measure the users' satisfaction with our services. But we do that in a context -- and I want to get some of this out now -- where our customers have no exit strategy. We get all the money up front. We are the Ferenghi, for those of you who are Star Trek fans, of the University. You know, you remember the first Ferenghi rule of acquisition -- once you get their money, never, ever give it back!
So we get it all up front, and part of what we're talking about is how do we deploy those resources to give the greatest gain, the greatest value, the greatest satisfaction. And those three words -- gain, satisfaction, value -- are, in fact, defined by faculty and students, not by the IT organization.
And there's not a market out there to inform us. I mean, we're not selling into or being pulled by a market that, in fact, incrementally rewards us for the excellence of our services. That's all paid for in advance, so these are the only ways we can deal with having a captive audience, a captive group of customers.
JB: It does occur to me that, by your constant measuring, that in itself communicates the message to your users that you do focus on the real results and value that you provide. I'm sorry, I shouldn't have used the word "value," but the real results and services that you do provide.
HS: "Value" seems like an okay word there. In fact, when Chris talked about the IT culture, he said that service, quality and value were at the center of IT culture. Chris, I wonder if you could tell us something about what you think the IT culture is on your campus, and is that what it should be everywhere?
CP: I don't know about everywhere. I'm enough of an anthropologist to admit that there are a variety of ways of doing things, all of which are appropriate and valuable.
We have become, over the last decade, a service organization that we focus on our products, our services. And generally, because we do things like quality surveys on a yearly, weekly or daily basis and we do get good feedback from our customers, that helps create the culture, just like Judith suggested. That doing this, in fact, focuses our organization, the people who make it up, on the quality of their service.
In turn, that focus creates a culture that serves it. And I don't have it right in front of me, but we do have a vision statement. We do have a guiding vision that does say that we are committed to assisting the university with IT in filling the goals of the university in the areas of scholarship and teaching and learning and research and service. So we are deploying our services in behalf of the classic university kinds of vision. But in doing that, we let the faculty and the students, to a great extent, determine what services they need, if that makes any sense. So that's very much part of our culture.
HS: When you do surveys and things, how do you do that? If I'm either unhappy with a service or I think you should be offering some new service, beyond surveys, how can I get to you folks and tell you that?
CP: Multiple entry points! Which I like, and every one of them gets closed. So if you start with me -- if you're a student who says, "I don't like X," I will work with that student until that "I don't like X" has been resolved in some way or another.
Now, if the student wants unlimited use of NAPSTER, I'm sorry. We've cut it off. You see where I'm going with that.
HS: But what I'm saying is, do you have, for example, the ability for people to make anonymous suggestions?
CP: Yes, ombudsman. And that's well known.
HS: Tell me about this ombudsman. That sounds interesting.
CP: I don't know who it is, but there is one. It's ombuds@indiana.edu, and that's for IT.
HS: That's the IT ombudsman.
CP: Yes. And advisory committees. We actively, as I say, are outwardly focused. We meet, collectively, individually, with faculty groups, with staff groups. I've just gone around this campus and around the university doing strategic plan updates.
HS: So anybody on campus can write to this ombudsman and say, "I don't like this, I think you should be doing that."
CP: Yep.
HS: "Get rid of all the PCs and put Macs in," anything they want to say?
CP: Absolutely.
HS: And you folks, this ombudsman will actually consider the thing, and it's anonymous?
CP: Yes. And as I say, I do not know who the ombudsman is.
JB: Okay, we do have a question coming in that takes us back to the activity-based costing process, and it's linked back into what you've told us, Chris, about totally costing out each of your services. And the question is coming from Scott Schevely at Cornell. And he's asking, do you factor in administrative costs or IT infrastructure such as wiring and electronics into the rates you charge for services? And if so, on what basis do you allocate these costs?
CP: The answer is yes and yes.
I'll go to the administrative costs first. You add up all the people costs and technology costs and other things. My office is a cost, the Dean's office, the Vice President's office. These are allocated to services as well. The only thing that is not in our services.
And you can find it in some of the things that I've written and Laurie and I have written -- Laurie Antelovic and I have written. The only thing that we are not able to estimate is cost of floor space, environmentals like heat, cooling and all that, because the university doesn't charge us for them. The industry average is 13% of total costs for those services, heat, light and floor space. If you want to compare benchmark, our costs -- which we publish against industry -- you have to add 13%.
So yes, we include our administrative costs and no, we don't include all those environmental costs, but in fact, you can estimate them. Now, on the wire plant and other things like that, we have depreciation schedules which we use and for chargeback services, yes, those depreciated costs are worked into the cost of the service every year and that's what we charge on chargeback, plus a small reserve as allowed by the federal government.
JB: There's a couple other related questions to the activity-based costing. One has to do with if we take a look at all the costs involved in supporting students and faculty and staff access to computers, just what percent of the cost is really the hardware and maybe perhaps the software? I mean, can a university reduce dramatically their IT costs if they require students to purchase their own computers?
CP: They'd love it! I'd love it!
JB: You would love to do that?
CP: No, the cost of support is exactly twice the cost of hardware and software. We put the entire university on a life cycle three-year replacement of desktops for faculty and staff --
HS: Why do you single out desktops as opposed to including laptops?
CP: Well, laptops --
HS: Because they have a different --
CP: No, no, but desktop could be a laptop.
JB: Okay, so it's including laptops. Okay.
CP: The cost of each station, be it laptop or stationary, Macintosh or Intel, is basically $1,000 a year, which includes all the peripherals that go with it, printers and things. A thousand a year.
When you look at those distributed support providers, local support providers, just to support those faculty and staff devices, it's over $2,000 per station per year. That squares very well with out costs in the student technology centers.
JB: Okay, so in other words, they might be able to reduce costs somewhat, but --
CP: By a third, if somebody gave you the hardware and software for free.
JB: Okay.
HS: But do you figure that that's about the same costs, then, to support a student-owned machine?
CP: Yes.
HS: So it's costing you, you said, $2,000 a year?
CP: Yep.
HS: So if all the students own their own machines, you still have -- most of your costs are still there.
CP: That's right. And then we can also, at that point, start talking about students owning their machines, the cost of modems, the cost of the Internet. The campus network is included in those costs, but the Internet is not. But the vast majority of the costs are support costs.
So you know, in a funny way, Ronald Norman is right in The Invisible Computer. We aren't anywhere near it being an appliance. And as we talked, when we did the preparation for this and my musing about being raised in Florida in the 30's and 40's and us having gas ranges and gas refrigerators there because the gaslines were buried and didn't get broken in hurricanes. And they had people, they had a support organization for gas refrigerators. They had classes for people who used gas ranges and gas refrigerators. They were not these sort of appliances that one could intuitively approach, as I found out once trying to light a gas refrigerator, and use easily.
Well, computers, information technology is perhaps at the state of Servel gas refrigerators in Florida in 1939. And what we want is the plug and play with the chip in it that we get now with certain other refrigerators that presumably will call the repairman when they have indigestion.
JB: Now, you've mentioned an author that I don't know is clear. Chris, who is the author that you mentioned?
CP: Oh, it's Ronald Norman's book, The Invisible Computer.
JB: Okay.
CP: Well, I don't want to push it beyond that.
JB: Okay, but in your comment, you kind of implied that an appliance -- if something is called an appliance, that it's easy to use, then.
CP: Absolutely.
HS: Like a toaster.
JB: Like a toaster?
HS: I mean, that's your model of, if it's harder to use than a toaster. If it's as hard to use, for example, as a VCR, it's too hard to use.
CP: Well, I don't find a VCR or a phone that hard, but okay, I'll give you the VCR. But as easy to use as a telephone. We don't need classes on how to use this telephone, although I don't know how to use most of the features.
HS: I was going to say, depends on what kind of phone system you have.
JB: Yeah, I've needed a class, myself, in phone. Over the years, the phones have gotten more difficult to use rather than easier because they've gotten more complex. So we find appliances going in interesting directions.
Let me just mention that we have time for just a couple more questions if folks want to send them in. And meantime, Howard, we've got a couple questions that we haven't gotten to that we planned to.
HS: Yeah. Chris, one of the things, when you're talking about supporting all these folks, obviously you support students, faculty and staff. But what about support for alums, prospective students, their parents, all kinds. I mean, who do you imagine your users really are?
CP: We don't support alums other than working with the alumni association to put up -- you know, how to manage mail, not a reflector but a mail address directory. We do work with parents, we do send out mailings with students who we hope will matriculate. I work with parents all the time. Parents do call me. So I have the same customer responsibility to parents that I have to students. Beyond that, we don't go much farther. We do help support colleagues in other institutions where we have joint research projects.
HS: What about distance education?
CP: We provide the transport. We have an Associate Vice President for Distributed Learning who reports half to my boss and half to basically the Vice President for the Academic Affairs of the university. And so he is our customer. And just like the Bursar/Registrar customers are not ours, the Bursar's our customer and we provide information technology on behalf of the Bursar and the Registrar.
HS: How do you know how well you're doing with these folks, like the prospective students and the distance education folks, these people that seem to be a little bit further away from your control?
CP: With prospective students, we don't. But I would guess that when I retire in three or four years, we will have something in place as a part of assessing the overall experience of dealing with Indiana University that will include an information technology component. Who else did you want me to address besides prospective students?
HS: Even the distance education folks.
CP: Certainly that will be a component of their evaluation of the course, will be the IT delivery.
HS: You say "will be." Is that being done now?
CP: Absolutely.
HS: What?
HS: Okay, another thing that I saw you mention -- and again, I looked at a lot of stuff that you had written or was up on the Web or whatever -- but there was something in there about a thing called a balanced scorecard.
CP: Yes.
HS: Could you say a few words about that? It seemed like an interesting thing, but I wasn't quite sure exactly how this thing fit in.
CP: Well, we don't have a lot of time for me to tell you how Kaplan and Norton start their book, but basically they start by you being a passenger in an airplane, walking by the cockpit, and the pilot is riveted on the altimeter, saying, "Today I'm doing altitude, today I'm doing altitude, today I'm doing altitude!"
HS: That'd scare you, right?
CP: You know, you say, "Don't you want to know about air speed and direction and whether the plane's right side up or upside down?" The balanced scorecard basically says, "Don't get fixated on just customer quality. Don't get fixated just on financial measures." Both are important. Look at your internal business processes. Look at the way you relate to your customer base. Look at your learning and growth as an organization.
And I should point out that we put aside for every full time employee, for toys, travel and training -- and these are allocated at the lowest management level, they're not held in my office or anybody else's office --
HS: Toys?
CP: Toys. Anybody who wants a new computer.
HS: Okay.
CP: Who wants a specialized computer or really wants one of those palm goodies so that he can do e-mail or she can do e-mail while they're under somebody's desk working on a connection. We allocate almost $4,000 an employee a year, and no reasonable request for training, for travel to professional meetings or to trade association meetings or toys -- that is, computers and other equipment necessary for your job -- are turned down.
And these monies are lodged at the manager level, so basically, you only have to convince the person who is one up from you, that sort of leads your team of people that creates this service, that you need to do this.
HS: That sounds great! Are you trying to recruit people here on this Webcast?
CP: Love to! But this directly addresses the learning and growth perspective of the balanced scorecard.
JB: So that is part of your balance, then, is insuring that your people get developed and supported and trained.
CP: Yes.
HS: That sounds great. I'm going to get the transcript of this part of the thing and kind of paste it on my door.
JB: You know, highlight it and send it on to your boss, right?
HS: I'm going to hang it on my door for awhile.
CP: We worked very hard on that, to convince the university that this had major positive results, and it does.
JB: Well, I can't believe our time is up again for today. Howard, do you have a final question for Chris?
HS: We always have a final question here!
HS: Our final question is, Chris, it sounds like you do a lot more of this than any university I've seen, and I think it's wonderful, really. The ombudsman idea is just fantastic. There's a number of things that I'm going to annoy a lot of people with at Princeton.
But for universities who aren't as far along as you are -- and that's like almost all of them -- how do they get started doing this? What's it take to get this kind of thing going?
CP: There's good guidance out there. There's the McClure and LaPata manual that's available from CNI that talks about assessment in IT. There's stuff the Flashlight Project has done. There's looking at and improving what we've done because we put up all our instruments and things like that. There're all kinds of ways to get started.
There is a learning curve. There is, if you will, some discipline involved with it, some methods that seem to be well-studied and produce the kinds of results that folks want. But this is not cleaning the Augean stables, believe me. Once you get doing it and get the rewards from it, which are understanding your customers and understanding your cost structure and understanding your internal strategy, your internal structure and culture, it is a large payoff. And it's not that hard, honest!
So I would start with various places like CNI, like the Flashlight Project, which does try and address value, by the way. Call me! I mean, you know, CNI, I've been a fellow there off and on for the last several years and I'm sure that they would be happy to loan me to you for a day or so.
JB: Okay, well, we have a lot of different resources up on the Web page, Chris, but I don't think we've got a couple of these that you've mentioned, so maybe we can -- over the course of the next week -- just add a couple more links there on that page.
CP: But please do strip out our cost and quality links and make sure they're in there.
JB: Yes.
CP: At indiana.edu.
JB: All right, we can do that. Okay, Chris, before I have closing notes then, do you have a final comment to close out today?
CP: You know, there's one other thing that I've left out where we get feedback, get advice, and like all university IT organizations, we have some very good advisory committees, both on this and the Indianapolis campus, that comprise faculty, staff and students. Some of them are rather general, others are rather specialized.
And we strike advisory committees -- in the polite sense of the word. We don't hit them, but we create them. We work with people to create them for special purpose ends. We are, like so many other schools, implementing PeopleSoft for the student and HR side. We have advisory committees, we have operational committees, we have implementation committees that advise about business process changes and the implementation of these very large and expensive software packages.
So I didn't want to leave out the fact that you can't do it all with customer surveys and phone surveys and e-mail surveys and casual wandering around. There is something to be said for the structured advisory committees as well.
JB: Okay, so the surveys are really part of a full structure to get feedback from your users..
CP: Yes, absolutely.
JB: Okay, well, great. With that, unless there's anything else, I think we'll go ahead and close for today -- and just remind folks, if they want to send in some notes, we can still answer some questions online after the session.
Many thanks to everyone who helped support these TechTalks, and we invite you and your institution to help support them if you're not already a CREN member. For now, you also want to be sure to set aside time in your calendar two weeks from today, March 16, for a session on Research Computing and Linux Clusters, and our expert that day will be Kevin Marooney from Penn State. And I hope that you all can join us then.
Thanks to everyone else who made this event possible today: to our guest expert, Chris Peebles; to technology anchor, Howard Strauss, in spite of his state of health there; to Terry Calhoun, event page producer; to David Smith and Patty Gaul of CREN; to Julia O'Brien, Jason Russell, Carol Wadsworth and the whole support team at MERIT; to Susie Berneis, audio file transcriber; to Laurel Erickson, transcript editor and indexer; and finally, a thanks to all of you for being here and for asking those good questions. You were here because it's time.
Bye, Chris. Bye, Howard.
HS: Bye, Judith. Bye, Chris.
JB: Take care and get well!
CP: Bye
