Indiana University
Brian D. Voss
Leveraged Support Model

April 1998

A. Introduction:

Indiana University has long recognized that the advent of distributed computing would lead to the need for a distributed support structure. In the late 1980's, IU began to develop (rather than distribute) support resources to campus departments. However as these distributed support programs progressed, it was seen that having distributed support resources, alone, would not avert the support crisis; that there was a need for a strong central support function. And in that the growth in the use of computing and information technology remained staggering, the large "untapped" support resource found in the users themselves needed to be drawn out.

By the mid-1990's, IU had crafted a support model hinging upon these three principals: 1)Local support resident in departments on campus; 2) Education and tools that help the end user support themselves; and 3) A strong centralized support function acting as a lever to facilitate the first two elements.

As we near the end of the 90's, the Leveraged Support Model is through its implementation and we have evaluated both the successes and failings of the model. We are in an 'operational' phase, and we are striving for continual improvement in the various programs and elements. The LSM is a living thing, and it needs to evolve over time to meet the needs of IT users, in a fashion consistent with budget realities facing IU and other higher education institutions today.

This paper is authored by:

Brian D. Voss, Director, Teaching and Learning Information Technologies Division
University Information Technology Services
Indiana University

With the support, guidance, and input of:

Gary R. Alspaugh, Manager, Local Support Provider Services Team, TLIT Division
Sue B. Workman, Manager, UITS Support Center, TLIT Division
Mark Lynch, Manager, Departmental Advising and Support, TLIT Division
Diane Jung, Manager, UITS Education Programs, TLIT Division
Duane J. Schau, Director of Technology, Political Science Department

B. Problem Statement

B1. The Institution

Indiana University is a public university composed of a residential campus in Bloomington, a major urban campus in Indianapolis, and six other campuses located in Gary, South Bend, Fort Wayne, Kokomo, Richmond, and New Albany Indiana. IU currently serves approximately 94,000 students, employs nearly 17,000 faculty and staff, and has a budget approaching $2-Billion annually. It is one of the largest institutions of higher learning in the United States.

The computing and information technology environment is rich and diverse. Specifically on the Bloomington campus, there are over 2,000 microcomputers available for students, and approximately 6,000 in service for faculty, staff, and graduate students. Platform breakdowns are: Roughly 80% are Intel; 18% Macintosh, and 2% unix.

University Computing Services (UCS) provided IT services to the IU system for mission critical systems and services (Administrative Computing), and to the Bloomington campus for academic computing services. Formed by a merger between administrative and academic computing organizations in 1988, UCS's organization structure no longer differentiated between those service sets. Rather, the organization was structured by services provided to the institution without regard to who they were, or where they were located. The organization consisted of approximately 240 FTE appointed staff, and over 200 non-appointed hourly and part-time staff. The annual UCS budget was approximately $26-million.

With the recent appointment of a full University Vice President for Information Technology, University information Technology Services (UITS) was formed out of a consolidation of the IT organizations on the two main campuses - IUB and IUPUI. As a result, a consistent service set across the two core campuses now exists. This organization has over 400 FTE staff and a budget of over $60-million; the telephone operations on both campuses are included in the organization (see http://www.indiana.edu/~uits/). The smaller regional campuses retain their own computing centers and structures, as they have always acted as "local points of service presence" for information technology support on the outlying campuses.

B2. The Situation

Let's break this down and look at some very factual observations and also some as-objective-as-we-can-be opinions on the state of support for IT.

Some Observations:

• IT use is growing -- the employment of computing technology in learning, teaching, and the operation of the institution has grown at a tremendous rate since the early 1980's. While growth has been most explosive in distributed fashion (desktops and departments), centrally served computing has also grown tremendously.

• IT use is distributed -- the use of computers and applications is no longer centralized, or at least not solely centralized.

• IT support is no longer completely centralized in the computing center -- Local Support Providers (LSPs) exist and are functioning well in delivering support to users of information technology and systems. Users, as well, are becoming more involved in their own use of IT and are undertaking supporting themselves.

• IT and IT support are linked -- Computing use continues to grow rapidly within the institution, driven not only by internal forces but primarily by external ones. As it grows, so does the demand for support.

And some opinions:

• IT's growth and the requisite demand for support can not be effectively contained by the central information technology organization (CITO). Attempts to ration support, or control the growth and expansion of computing and IT use have proven useless, contentious, and frustrating for all involved.
  

• The growth of IT support in departments is not a phenomenon resulting from the shifting of "chores" from the CITO to the departments. Rather, it is a phenomenon resulting from the exponential growth in the demand for IT support -- it consumes more resources now both internally to the CITO and externally to campus entities.
  

• CITOs are not capable of providing all the support needed by this ever-growing demand for IT and support of that growth. CITOs have seen their support functions grow from 10% to 25% of their resource (mostly human resource) in the past decade. Further growth in the percentage of the CITO functioning as a support unit will mean that other vital and valued services will need to be eliminated or transferred to departments (networking, application development, research and student computing services, etc.)
  

• User responsibility is increasingly a necessity. Users have integrated IT into their job functions -- teaching, research, and administration. Computers can no longer be viewed by users as "things" which get in the way of them doing their jobs. Rather, they are part of the job now, and as a result, individual responsibility for using IT is on the increase.

B3. Strategic Significance

It is a given that information technology is a part of today's higher education environment. What is not always seen is that the costs of supporting the use of information technology are greatly more significant and long-standing than the costs of acquiring it.

If the need for a strong and structured technology support environment is not recognized, an institution can waste it's capital investment in technology; acquiring and placing hardware and software which is not used effectively (if at all). And if this need is not addressed properly, the institution can get caught in a spending spiral for technology support that exceeds the spending spiral for technology itself.

C. Objectives: Desired Outcomes

The objective of the Leveraged Support Model is to allow the institution to continue to grow its information technology environment without co-requisite growth in resources devoted to supporting that environment. As of this writing, IU has made a substantial investment in support resources (human); so the desired outcome was (and continues to be) to maximize the benefit of that investment without growing it much further.

D. Approach

D1. Financial Resources

The financial resources required to develop and implement the LSM were a mix between one-time investments, intended to 'jump-start' the process, and base funding investments on the part of the CITO to maintain the LSM environment, and on the part of schools and departments to provide for local support.

• One Time Initiative Funding - Seeding the Base

  In the late 80's, when Indiana embarked on its endeavor to establish a distributed support environment, the first step was creating a pool of one-time money, designed to help departments better understand the benefits to their operations of having computing support resident within their organizations. This program -- Distributed Support Assistants (or DSA) -- was initially funded with enough money to start eight (8) partnerships. The partnerships consisted of a two-year 'contract' between the CITO and individual departments, the terms of which follow:

  1. CITO would provide for full funding in the first year of a 20-hour/week DSA position (approximately $10,000 each at the time), and would hire, train, and support that person as 'one of its own.' This included the provision of a current-technology workstation and software for that person, so that they could perform their duties.

  2. The department would provide a supervisor for that person, to deal with service-related aspects of the job. That person would work closely with the DSA Program coordinator in the CITO, so as to learn how to supervise IT people.

  3. In the second year of the partnership, CITO funding would cover half of the costs, with the other half picked up by the department.

  4. After the second year, if the department felt there was value in having resident local support, they would assume all costs of providing the position. This would then entail a base-budget commitment for wages/salaries, training, equipment, etc.

  The DSA Program ran for 5 years. Approximately 27 partnering departments were served over the programs lifespan. All departments that completed the full two years decided to carry on with funding on their own. This 'try before you buy' approach, proved successful in not only establishing the paradigm of local support, but also in establishing base-funding for IT support in units outside the CITO. Essentially, participating departments realized the value of having 'their own' IT support, and were able to translate the benefits to their faculty and staff into a real commitment on a local basis for funding that support.
  

  Another area where one-time funding was used to seed future base funding was in the area of training/educating local support providers in departments. Here, the initial concept was as a complement to the DSA Program, designed to train people already in departments, who were assuming IT support functions on their own, or through initiatives of departmental leadership. Approximately $80,000 in one-time money was invested over the five years to provide for CITO-taught courses for these people. Eventually, this translated into a $20,000 base commitment within the CITO to fund on-going education and certification efforts. This money provides for a training facility within the CITO, allowing us to upgrade equipment every three years to keep pace with changing technology.
  

  One final aspect of the use of one-time funds had to do with another base-establishment exercise. A third program was funded within the CITO designed to provide on-going support for DSAs and 'home-grown' local support providers (LSPs) in departments. The Partners In Computing Support Program -- or PICS -- was designed to provide a central cache of tools, utilities, and equipment in the CITO, made available for use by LSPs. This funding -- about $100,000 over three years -- provided for servers that housed frequently used utilities and vendor support information, as well as a software distribution server (SoftServe) which would hold institutionally licensed products which could be accessed by users and LSPs for installation of products on local desktops.
  

  The base funding for this service was eventually set at nearly $100,000, and includes a variety of tools for all platforms, and computer-based training modules for use by LSPs preparing for higher-level education/certification offerings.
  

  In a very real sense, Indiana University leveraged one-time funds to help not only departments, but our own central IT organization realize the benefits of leveraged support. This 'seeding' philosophy worked very well in helping us all understand the benefits and costs of a distributed support environment, and allowed us to gather momentum for establishing this new paradigm of support.
  

• Base Funding

  The base funding component of the LSM is substantial. When you look at the overall budget allocations within the CITO and also within campus departments and schools, IU has committed a rather sizable allocation of budget toward IT support. There are approximately 200 positions on the two main campuses (IUB and IUPUI), supplying resident support services to departments. Assuming each averages about $35,000 in total budget impact, then $7-million dollars goes toward this annually.
  

  Within the CITO, there are approximately 25 positions allocated toward supporting the LSPs and other aspects of the LSM community. Again, using $35,000 as an average and then adding in the costs of maintain education/certification and the PICS program, the CITO is investing about $1.4-million annually in this effort.
  

  An important note on this investment, though, is that it is not "new" money. There have been no allocations to departments or the CITO to fund this new set of services and resources. Internal reallocation has been the method used to achieve the base funding.
  

  Within departments, this translates into them replacing other types of positions -- clerical or even faculty -- to this new purpose. And within the CITO, the funding has been achieved by a mix of evolution of previously existing departmental services, and reallocation of funds from other support services. As an example, the overall CITO budget for support has remained relatively constant since 1990, in fact becoming slightly smaller. This reallocation has occurred through both better efficiencies in offering traditional support services (walk-in consulting, telephone support, etc.) and elimination of some no-longer-vital services (such as departmental IT planning, due to the fact that LSPs now do most of this planning with our guidance, rather than direct involvement.)
  

  Only today, when the overall nature of IT support is becoming of greater importance to the institution, is a general move afoot to increase base funding by means other than reallocation. A special request for IT funding to our state legislature has resulted in a one-time allocation of funds to IU for Information Technology. Our hope is that this one-time investment will eventually become base -- following the pattern established at Indiana as described above.

  
  

D2. Human Resources

As touched on in section D1, human resources are the key to bringing about the LSM. Briefly, then, let's examine two aspects of human resources investment.

• Within the Central Computing Organization

  As mentioned, there are about 25 positions within the CITO focused on support of the LSM and local support providers in departments. The people perform a varied set of activities that yield important services to LSPs in departments. Including:

  1. Departmental Advising and Technology Assessment
     This function provides the forward-looking vision for our LSPs. Designed to help them in their busy lives by taking the burden of 'watching ahead' onto the CITO, this area investigates new and emerging technologies that impact the desktop and workgroup environments, and seeks to find ways to provide analysis and recommendations to LSPs on the implications of changes. We have a Departmental Support Lab, equipped with new hardware and software, giving a place where LSPs can congregate to chat with CITO technical staff and find ways to employ new technologies in addressing current, local needs.

  2. LSP Services and Support
     Designed as a complement to the Advising and Assessment role, we think of this group as providing support to LSPs for services and technologies already in place. Our PICS program is housed in this functional area, as are many front-end service points like our LAN Lab. Our Education/Certification program is also housed in this area, as we use our own front-line technical staff to provide that training.

  3. Knowledge Base and Online Help Services
     Although not specifically cited as a LSP service, the staff who develop, maintain, and enhance our online support resources -- the Knowledge Base, or KB -- are also in many ways part of maintaining the LSM. Our logic is that if we build an excellent online support tool, not only is it available to LSPs as another way to get support from the CITO, but more importantly it becomes a tool for end-users themselves. As the end users of IT are better able to support their own needs, they rely less on LSPs for pedestrian support issues and thus, free LSPs for more detailed and technical activities related to improving the IT environment within their departments.

  4. General Education
     Similar to the logic mentioned about the KB, we feel that education is a key component in a user's ability to support themselves. People who understand the applications and technology on their desktops -- even the very basics of function -- are better equipped to utilize that technology and to also interact with the KB and their LSPs.

  These four main areas comprise the CITO's contribution to the LSM in terms of human resources. Other areas, such as 'targeted areas of technology support', wherein we supply very detailed and in-depth support for 'hot' or complicated technologies, are also a part. The CITO at IU has several such 'hotbeds' of deep technology support, covering areas such as data management (databases, GIS, desktop application development tools, etc.), statistical and mathematical computing, unix workstation support, high-performance computing, and library-related text resources to name just a few.

• In Departments

  Departmental resources, as mentioned comprise about 200 positions on the IUPUI and IUB campuses (which together account for about two-thirds of the demographic population(s) of Indiana University). These positions run a very complete range, from entry level to senior professional, to technical management in some units with large local support operations.

D3. Policies & Practices

The Leveraged Support Model was brought about without much in the way of policies. It evolved in a natural, though encouraged way over many years. It is our perception that policies regarding the move to distributed support are more applicable where an institution wants to make a less gradual transition to a leveraged environment. Such is the case at Brigham Young, where institutional mandate around 1990 resulted in one of the best implementations of a distributed environment in rather short order (1-2 years). Likewise, our colleagues at University of California - Davis, where the central IT organization realized the need for a fast implementation relied more on policies. But even in their case, encouragement rather than enforcement was the key to success.

At IU, there today still is no real policy set covering the LSM environment. About the only 'enforcement' mechanism is that the CITO is no longer really in the direct support business when it comes to departmental issues. We have a very limited 'office calls' service, and charge for it. Our logic being that the vast majority of the institution has made local budgetary commitments to providing local support, and the CITO should not undermine that by providing services to departments who haven't made such an investment, allowing them to be 'subsidized' by the investments of their colleagues.

At IU, we also have not made an effort to limit the support environment by the use of 'supported equipment/software lists.' We don't refuse support to anyone on the basis of what hardware or application we use. However, this is not to say that we support everything. We provide high levels of support for the hardware and software used by the preponderance of our user community, and make available information on a wide array of competing products, including directing users to other sources of support for products not richly supported by either local or CITO resources. Our policy -- if such a word applies-- is of support abundance, rather than rationing. Albeit, not all support is available at no additional charge, or from local resources.

We've instead opted for an approach that encourages 'supportable behavior' on the part of our user community. We encourage strong local support by providing education/certification training at no additional cost to departments. We encourage the use of certain pervasive software by acquiring site-licenses for it and then making it freely available. We encourage the use of 'standard' hardware by working with Purchasing to find volume discounts on leading vendor products, and then provide very competitive acquisition costing of that 'standard' technology, as compared to less supportable options. And we encourage responsible and informed use of pervasive technology by providing 'free' education to our end users.

Taking a moment to focus in on one aspect, let me describe a bit about a very successful practice we at IU have put into place -- that of site licensing application software products. IU has been a pioneer at working out licensing agreements which provides the most popular and pervasive applications to our user community. In 1995, we pioneered licensing agreements with Novell (and later with Corel) and Sybase. In 1998, we worked with Microsoft to establish a first-of-its-kind enterprise licensing agreement that provides for use of a wide variety of Microsoft products for our entire university community.

This licensing approach allows us to make it easy for our user community to not only get the products they desire, but also makes it possible for everyone to have the complete set of products, latest versions, and access to support and educational resources. By leveraging the demands of the broad university population, we are able to negotiate agreements which allow us to translate that leverage to all within IU. While we also encourage diversity in software use by negotiating for competing products, in so doing we are encouraging standardization without a policy to enforce it.

Generally, then, our method is one of encouragement of responsible behavior with regard to IT support, rather than one of policy governance which restricts behavior to a standard.

D4. Technology Resources

As mentioned above, the CITO at IU provides an abundance of support and does so by leveraging many technologies and practices.

A key way we bring technology to bear is in the delivery of both products and support to our user community. We have an elaborate online mechanism for distribution of software products. SoftServe is a networked server that contains not only the software, but custom written applications that make installation and configuration easy. We also use CD-ROMs to distribute software, so that people not connected to the high-speed network can install software on their own machines. Soon we will make this service available from the web, which will increase dramatically the population that can take advantage of it.

Our online help -- the UITS Knowledge Base or KB -- is another technology-enabled application. Recognized in 1997 by Yahoo! Internet Life as the most outstanding support resource on the internet, the KB is a vital part of the IU IT support environment. Via the web, the KB is repository of thousands of publications and answers to technology-related questions. Users point their web browser to the KB, type in a subject area, or even a text question, and receive back documents and answers containing references to keywords in their queries. If users do not find an answer to their question, the KB is then mechanism for sending that question to the CITO support center, where an answer is both emailed back and added to the KB for use by others.

We've also taken steps to use technology in the operation of our support centers, and in our interactions with LSPs. Using a vendor tool -- Apriori -- and a home-written interface called 'WorkBench', we've automated the tracking process of problems reported to our support centers. Questions not readily answered by front-line resources are forwarded internally to technical staff, who can keep track of open problems with a quick glance at their desktops -- hence the name WorkBench. Problems are resolved, reported back to frontline staff and users, and also queued for entry into the KB. Another tool -- Eclipse -- provides for delegation of Apriori queues to LSPs, so that they can utilize this problem-tracking tool locally. They can track open problems and also escalate more challenging ones to the CITO for resolution. The technology allows us to better track problems and communicate resolutions to all parties - thus better serving our users.

D5. Mix And Balance of Resources

There is, therefore, a requirement of both human and technology resources in order to bring about the LSM at IU. As detailed above, technology helps us better leverage our human resources. It is our belief that pursuing a single course -- human resource or technology -- will not address the needs of our users. Only by carefully balancing and using both can we achieve the success we have at IU.

One aspect not to be lost is leveraging human relationships. Often, what we lose sight of is that it is not only people who make this all work, but the relationships they establish with others. Our distributed support efforts -- from the earliest programs to today's ongoing services -- are successful because we have established good working relationships with our LSP partners and the users. We like to call this 'leveraging the good ol' people network.' And what it means is that we can spend our collective time, together, enriching the environment rather than engaging in non-productive contention. This is why we're so into the encouragement approach, rather than the control approach.

We take great care to maintain this "GOP" network. Our LSP partners are important to us, and we ask their advice, seek their review, and listen to their needs. Every year, we take a representative group of them off-site to listen to what they have to say about our services. We use this to try to better explain our reasons for providing certain services, so as to ensure that there isn't a lack of understanding between 'the partners.' We also use this to find ways we can make improvements, refine services, drop ones no longer needed and add ones that are going to be valued by the LSP community. We cherish -- and that's not too strong a word! -- our relationship with the LSP community, as they are now the ones on the frontlines. We view it as key to our success to make them successful. As you can see by our services offered, we try to translate that sentiment into services that make it a reality.

E. Results

We feel we've achieved a tremendous success with the design, articulation, and implementation of the Leveraged Support Model environment at IU. IU is recognized nationally as a leader in support services. But more importantly to us, the IU support environment is strong for the benefit of our users -- students, faculty, and staff.

E1. Benefits/Successes

• Growth of LSP Community

  Since 1990, when we embarked on establishing a local support environment, we've seen a growth in positions within departments to today's level of over 200 such LSPs resident in departments. This was not achieved through the acquisition of new funds. Rather, through departments realizing the benefits of local support in meeting the ever-increasing demand for IT support by their faculty and staff, they have chosen to reallocate other resources. For a CITO to accomplish this in a centralized model, a great deal of negotiating or policy-level enforcement would have been required. And we feel the service would not have been as good, had we tried to do this centrally.

  The LSP environment continues to grow, as the role of IT and its requisite support requirements increase. But central mandate or CITO requests are not a part of this. Departments are now empowered to analyze their own needs and are better positioned to understand the benefits of support in meeting the demands of their own users. They rely upon the CITO for directions for the future, but in a very real sense control their own destiny and do so from a position of knowledge and experience.

• Ed/Cert and General Education Successes

  Today our education/certification training is no longer local in nature (specific only to IU). We offer Microsoft and other vendor education and arrange for certification from those vendors. Our LSPs can take advantage of this education to not only improve their ability to deliver services to their users, but also to benefit their own careers, as the certifications are recognized beyond IU. This has led to a problem,which we'll describe in section E3. But this tangential problem of 'broadening their horizons' is not so bad as to lead us to believe we shouldn't have embarked on this effort, nor that we should abandon it now. Everything worth having has a price -- and a qualified LSP community is definitely worth having.

  Our general education classes have also seen an increase of use by our community. In fact, even faculty -- who historically have been reluctant to make use of this training -- are beginning to attend classes and are gaining basic skills which help them better use technology, and better apply it to their pedagogical activities. Our hope for the future is to broaden this effort, and to come up with even more methods conducive to faculty involvement.

• Use of Online Help and the Knowledge Base

  Use of the online help tool -- the KB -- has increased dramatically in the past two years. Today, almost 40,000 'hits' on the KB occur each week! {It was 25,000 hits per week in July 1997.} Demand on traditional support mechanisms -- telephone, email, and walk-in -- while not decreasing due to KB use, have also not increased substantially. They are holding level in many regards. In that we accept that the demand for support is increasing, this leads us to believe that this growth is being satisfied in a large part by the online tools. And they are less expensive to support per-contact than traditional mechanisms.

E2. Failures

These successes have not been achieved without some failures. The LSM is an evolving thing, not a pre-designed-perfect solution. In each case, we are trying to adapt the implementation as these failures arise.

• Expense to the Institution -- loss of economy of scale

  An argument can be made that collectively, IU is spending more for IT support in a distributed environment than it would with a totally centralized approach. In fact, there is evidence that this is true. Anecdotally, there is redundance of service provided individually in departments that, if gathered together, would yield financial efficiencies in a totally centralized approach.

  But what is lost in this convenient logic is that the IU community is diverse and decentralized in itself. A totally centralized environment would in many ways hinder the diversity and pervasiveness of support, as one organization can not be omniscient in the IT support environment. Specialization at the departmental level is necessary, as is the redundancy of service which must exist to ensure that local users have access to support they need.

  It is true that a distributed model is more costly. But it also delivers better service. Hence, we believe the added benefit has value. In response to this, however, IU is investigating ways that consortium-like sharing of talent and services can be provided in localized centers. One example is in IU's large College of Arts and Sciences. Many entities are too small to afford local support, or perhaps have sources of funding which allow them to 'over-provide' resource. The College is looking at ways to more evenly spread support to these smaller units, creating decentralized centers of support which serve a collective of smaller units. We will embark on this initiative in 1998, and have results of this modification to the LSM by 1999.

• Moved too fast to pull traditional support mechanisms

  In our enthusiasm to advance the LSM, we too quickly made trade-offs in our budget to provide for distributed services at the expense of traditional support services. In essence, we moved resources to building the sprinkler system while fires were still burning too widely! As a result, we took a 'hit' in the eyes of our users, with regard to their perception of our service attitude. And worse still, the perception grew that this reduction in service was due to our efforts to build an environment which relied more heavily on user and LSP contributions to addressing the problem.

  Once we realized this, we made a move to 'patch things' in traditional services until such time that our LSM services were better in place and the environment in the user community better situated to sustain their role. This was another one-time infusion of funding -- $250,000 over two years. It has since expired. However, we also learned that we had inadvertently reduced the overall support budget in the process, and are taking steps now to return it to its 1994 level across all of support services.

• Lack of Consistency in LSP job functions/descriptions and compensation

  A failure that still is haunting us is that there is no consistency outside of the CITO with regard to LSP job descriptions nor compensation. Essentially, in our 'free competition' environment, departments are free to create positions at whatever level they can afford. Many departments are weary of the highly fluid nature of the LSP environment, and are deciding it is better to 'highly compensate' someone and keep them, rather than periodically engage in a search process for a replacement. This has served to drive up the market for LSPs, and foster perhaps unhealthy competition between campus departments.

  It has also created a situation where these departments are able to better compensate talent than the CITO. The CITO has to worry more about comparative consistency, as we have a large group of technical staff. We can not make 'by-position' decisions with regard to classification and compensation, as to do so creates internal equity problems. The result of which is that departments frequently can afford to 'out-bid' the CITO for staff talent.

  The CITO at IU has had a philosophy to actually encourage our staff to move to departments, as we've found that having folks with experience within the CITO make better LSPs, and also better partners for the CITO. For this reason, we do not now, nor plan in the future, to pay 'top dollar' for staff. However, we must get within 10%, in order to allow staff to make good career decisions without too much emphasis on compensation.

  We are trying to examine this issue, both within the CITO and across the institution, in the hopes of providing better guidelines for positions and more equitable and consistent compensation for like-functioning positions across the university. However,it is true that the over-riding market forces outside of IU have a larger part to play in this challenge. The market for staff with these talents is tremendous, and because of the national shortfall, a classic supply-demand situation is driving up compensation for IT staff nationally. We will try to address this problem's many facets -- both internal and external in nature. But it is complex and no readily available, simple, or cost-free solution seems to exist.

E3. Unintended Consequences

• Change in the supply/demand for talent (seller's market); CITO takes new place in "food chain"

  As mentioned in section E2, departments are fully capable to out-bid the CITO for qualified staff. When we first seeded the LSM environment, nearly all IT positions were within the CITO. Now one can see that there are nearly 10-times the number of positions outside the CITO than within it, for staff with the required LSP skills. The CITO does offer a lot of advantages for an employee. But so do departments, especially if a person likes the 'ownership' aspect of a job. We've had to deal with this and it remains a challenge as described above.

• Provided excellent mechanism for rolling out major systems (CITO no longer on the 'hot seat' alone)

  As more and more of our university information systems go to a client-server model, we've found the presence of the LSM and the LSP in departments has given us a great mechanism to roll out these applications and get them into the hands of users. When we implemented our Financial Information System (FIS) we relied heavily upon LSPs in departments to ensure that the desktop client was properly placed on user machines and that special configuration requirements for desktops were put into place. For a CITO to have done this would have been a daunting and contentious task. Our LSP partners handled this with ease, and helped us better position the application by giving us feedback on requirements and changes to the technical function of the client.

  It is also true that we've capitalized on the LSM to help sponsoring application departments -- like Financial Management Services with the FIS, and now Human Resources with the forthcoming HRIS system -- understand the nature of computing support at IU. These application owners have also followed a LSM-like approach, finding local support contacts within departments to provide the functional-area support for their applications. This is not the technical support, but 'how do I use this' support. They have come to not expect the CITO to provide support for their applications at our helpdesks, and have established their own centralized operations along the service lines of our service role in the LSM. And they've sought out local contacts within departments for their application, and sought ways to support them well. In the end, we feel this broadened use of the LSM has increased the effectiveness of IU's ability to use these new application systems, and to use them well.

F. Lessons Learned

We've covered many of the lessons learned in the above sections. Generally, we feel we've done an excellent job in getting the ball rolling, and then working within the LSM day-by-day to help adapt our services to continue its success. We've had our failings -- some which were the result of lack of foresight. But most we do not feel, in hindsight, we could have avoided. Instead, we're trying to move forward and make adjustments as we go along. We're not without advice for others though ....

F1. Advice to Others

Generally, we feel we did the right thing and we are proud of our accomplishments. From our model and experience, we offer the following advice to others.

• Start Early and Build Slowly

  At IU, we had the advantage of getting started early and taking our time to let the model evolve over time. It is not as if we had some Machiavellian insight and plan, and carefully manipulated the situation to yield today's result! If we had, we would have certainly avoided some of the failings! However, I believe we did have the benefit of high-level insight back in the late 80's. We were fortunate to have both first-line leadership and staff who made the vision of others a reality, bringing it into focus those early visionaries never could see. In fact, some speculate that the original intent of the visionaries was to remove IT support from the CITO's portfolio all-together. Instead, what we ended up with was a highly functional three-tiered model where the CITO plays a vital role in the success of the overall environment. To have removed the CITO from this model would have been, in our retrospective view, a terrible mistake.

  Today, we think it may be too late for others to follow the IU 'blueprint' in total, simply because the crisis in support of Information Technology is already upon us in a critical fashion. The overall model is very applicable today. But the deliberate, evolutionary approach probably won't work for many facing the crisis.

• Move intensely with a lot of commitment

  Universities can, however, take the LSM and make an intense commitment to its implementation, including the use of policies. In our view at IU, the absolute best example of this is what has been done and is working very well at University California-Davis. We leave it to our very successful sister institution to describe their success and methods, but as a pioneer in distributed support, IU lauds their accomplishments and encourages others to follow their approach.

• Get Executive Order

  Another approach that has a historical record of success is that of Brigham Young University. Basically, BYU used an executive mandate to implement a distributed support environment in the early 90's. That model continues to be effective today, and should serve as an example for institutions with the will and ability to achieve LSM environment by way of mandate.

  Several years ago, we at IU believed that most institution's CITO functions could bring about a BYU-like implementation. But today, given the state of crisis in the support of IT, perhaps such a solution is the only viable option in order for these institutions to survive in their use of IT. Again, we leave it to our colleagues at BYU to describe the programs and methods that have brought them success.

• Learn from the mistakes of others; but also be willing to accept your own mistakes and adapt, improvise, and overcome

  In short, now is not the time for trying to re-invent the wheel. We advise other institutions to review the work done by IU, UC-Davis, and others in this area. We encourage them to accept some things -- the problem statement and the fundamental concept of a leveraged support model -- as proven facts. And then concentrate more on how to make it happen at their institutions, in the varied and diverse environments in place at each.

  All of us have made mistakes along the way. We're all still alive. Don't be afraid to try things and even to fail. Don't be afraid to keep trying things that worked other places, even if some of them fail at your institution. As any US Marine might tell you, your job is to "adapt, improvise, and overcome." And also as they say: Semper Fi!

F2. Advice to Self

We've covered a great deal of this throughout this paper. But quickly, we'll summarize:

• What would we do differently?

  Very little.

  Except we wouldn't pull traditional services as early as we did. Had we left those in place, our costs would have been about half what it took to remedy the situation. This was the biggest error we made along the way, and while it was a result of a positive zeal, it shows us that zeal is a dangerous thing in any form! The cost to our organization in terms of user perception was great. And worse, because the reduction in services was linked to the LSM implementation, it damaged the perception of the LSM within the IU community. A price we still occasionally have to pay today.

• What changes would we make?

  We might have worked more on the end-user aspects of this sooner. The implementation focused initially on establishing and growing local support positions. Combining this with our early withdrawal of traditional services, we put too much pressure on the LSP community, which is only now starting to ease somewhat. While it is true that the KB function needed time and energy to evolve, and probably could not have happened more quickly, we could have focused more on end-user training. However, as a result of previous leadership strategies, user education was considered tangential to our former mission, and thus, was a charge-back service. This discouraged people from getting skills that would have, in the long-run, eased the pressures on the CITO and LSP operations.

  If there is one thing we could have done differently, we would have "given away" basic training and more strongly encouraged use by faculty and staff. We did it for students -- as their technology fees funded basic training. But we did not do so for faculty and staff due to lack of funding for such a program and the lack of foresight on our parts to find the funding. Today, we are trying to address this situation, and hopefully will do so in the next 12 months. It certainly isn't too late to provide this training; but it would have been grand to have had it in place years ago!

• What changes do we plan to make?

  Again, we've covered most of this throughout the course of this paper. But our main focuses of change include:

  1. Establishing better support within our College of Arts and Sciences to address the lack of pervasive an uniform support service throughout this very diverse academic unit. We're working closely with the Dean's office at COAS, and believe we'll make some progress during the next fiscal year.

  2. Continue to broaden the impact of our education programs, both what we offer to LSPs and to end users. We plan to expand our EdCert efforts beyond the Bloomington campus (though LSPs on all campuses in the past could drive to Bloomington for this training -- it was not convenient, and thus not well utilized). We plan to make our educational offerings to users more consistent across all campuses. And we plan to find better ways to encourage our faculty to gain basic and advanced IT skills, as we work with them to utilize IT in their pedagogical activities, most especially teaching.

  3. Look for ways to utilize the LSM in the growing area of instructional use of technology at Indiana University. The special requirements of instructional technology go beyond the bounds of traditional LSP services. Many larger schools and departments need to establish and grow this aspect of IT support within their own operations, in much the way they did with 'pedestrian and pervasive' IT support functions during the past decade.

  4. Finally, we will continue to look at the LSM as an evolutionary thing. We will continue to look for places where its implementation is not effective, or has become that way as the environment in which it exists evolves and changes. We feel the LSM is conceptually adequate for the next decade and beyond. But the services which bring it about must constantly be assessed and reviewed,and changed where necessary. We intend to keep the LSM successful.

---

	Developed & Maintained by: webmgr@cni.org	© 2008 Coalition for Networked Information ALL RIGHTS RESERVED.
Any comments, or feedback?		Last Update:   Tuesday, 08 October, 2002 - 01:14 PM - EDT