November 2002

Volume 18
Number 8


Reducing Costs

The downturn in the national economy has brought pressure on all information technology organizations to reduce costs. Lower revenues from taxes, among other things, have prompted state legislatures to scale back funding to public institutions of higher education. Private colleges and universities have seen lower returns on their endowments with the plunge in the stock market. Even a rapid return to more prosperous times is unlikely to lead to immediate budget relief. Amid the overall effect on higher education, IT organizations have not been exempt from budget retrenchment and instructions to prepare contingency plans to get by with less.

Everyone’s hope is to ride out the recession, knowing that economic cycles come and go. For the short term, making cost reductions without cutting into essential services, replacement cycles, programs, and staffing is the focus of attention. But there is also a challenge and opportunity here to establish ways to moderate costs, permanently for some categories, and periodically (in the future) for others.

Savings in ordinary operating costs are the obvious place to start. Institutions that are not currently charging students for printing from computers are now thinking about how to recover those costs. Computers and monitors that have been running day and night are now being turned off at the end of the day. These improvements in operating efficiency are bound to have only moderate value at best. Some of them, such as increased recovery of printing costs, entail new expenditures on the technology to enable that to happen and continuing costs in administration. But overall, these efficiencies are limited by being tied to expense types that amount to only a small part of the total IT budget.

More substantial savings can only be had by targeting the sources of the largest expenses. The analysis of costs and search for savings usually begins with a look at the IT budget.

At most institutions, sixty percent of the outlay is in personnel costs. The next biggest fraction, at twenty-five percent, is hardware. The sheer size of the personnel budget generally leads to staff reductions in extreme budget crises and hiring freezes in circumstances that are expected to be temporary.

The current down cycle is the first big one to hit since most colleges and universities established regular replacement cycles for equipment, and so there is really no history for how budget austerity applies to this category of cost.

But adequate staffing and timely refreshes of equipment are two of the hardest won achievements of IT in the past decade. Although the temptation to retrench in these categories is undeniably strong given that they account for eighty-five percent of the operating budget, imposing percentage cuts on those budget lines is not the best way to proceed.

Those new initiatives
Some part of the equipment and personnel budgets (e.g., contracted services) are tied to new projects. Sometimes funding for new initiatives is in fact outside the IT budget per se. Postponing or scaling back new initiatives is one way to reduce IT-related costs without cutting into the established base of services, systems, and facilities.

New initiatives that increase the scope of support obligations for IT deserve especially close attention when funding freezes or reductions are anticipated. Not only do new IT assets cost money to establish, they also build in continuing costs, whether assessed and charged or simply absorbed. If the IT organization faces lean times, the added load can often turn out to be harder to bear than was forecast.

Equipment life cycles
Hardware replacement cycles have been an important achievement in IT planning and budgeting in the past decade. Quality of computer and network services and the effectiveness of help desks have benefitted enormously from the periodic sweeping away of obsolete systems. But this progress has also introduced higher costs in equipment acquisitions, as the tendency has been to buy a single standard of computer for everyone, figuring that differences in usage requirements were less important than the benefits of uniformity. Having fewer models of devices has seemed to correlate well to better support.

But this might be the right time to revisit the balance of costs in computer replacements. Most lines offer a price spread that is keyed to processor speed, hard drive size, RAM, and communications devices. The volume of these devices purchased annually for routine upgrade or replacements campus-wide is certainly big enough to warrant trying to save a hundred or two hundred dollars per unit, particularly if some of the machines will only run a modest spread of applications. Administrative workstations might not miss the higher clock speed needed for running calculations or graphics.

This reasoning carries the stigma of having been used in the past to justify what became an unwieldy mix of equipment on campus. But the differences in configuration and performance within a model type are now relatively insignificant in supporting computers.

The timing of replacements can also be a source of some savings. A year might be added to the cycle, particularly for machines whose operating system will not change during that added year of life.

Somewhat more problematic in net savings – but worth consideration – is the cascading of replaced computers as a way to stretch the replacement cycle. In this process, the oldest computers are retired, to be replaced by the next oldest cohort, moved from one user to another. Two-year old computers from the math lab, for example, could be replaced by new machines, and the old ones moved to offices. But when this migration of computers is undertaken on a large scale, labor costs and slowed delivery rates brought on by the need to refurbish and reconfigure the computers has to be taken into account. Swapping new hard disks into these computers can speed this process and give the new owner better confidence that the machine is less likely to encounter the most dreaded of breakdowns – hard disk failure.

Rationalizing software
Software standardization, too, has brought undeniable improvements in the quality of IT support. Nobody would want to return to the seemingly endless variations in configurations that were once very common. But some workstations could have smaller suites, omitting software the owner never uses. Key-serving licenses can also be used to reduce the number of license copies needed for software not used frequently.

Too often, software packages are purchased, installed and then really never used. HTML editors, project management software, and photo/ graphics editors look appealing but turn out to be more detail-intensive than needed by casual users. Buying fewer copies of infrequently used software and making it available via a server can reduce the waste of software abandoned in place on desktop hard drives.

Trimming facilities
As the percentage of students owning computers approaches one hundred percent(though not there yet), colleges and universities need to rethink the number of computers they feel they need to make available for general access. It is true that students tend to use computers at several places during the course of a day on campus, but a close watch should be kept on usage levels throughout the semester, with an eye to reducing the total number of workstations needed to meet the need.

Alternatively, multiple small facilities might be consolidated, saving costs in computers, printers, supervision and maintenance. Small public labs with minimal supervision in remote locations have always had a more costly maintenance history and would be reasonable candidates for closure, if usage can be redirected to other facilities.

Because a lot of computing in public facilities is recreational – as a quick walk-through in any of them will show, setting up "cyber cafes" in places where students congregate might be worth trying. These would be equipped with older computers displaced in the refresh cycle and kept for web-oriented use, including e-mail checking. If these proved convenient and succeeded in drawing non-academic usage out of the labs, the number of first-line computers needed to meet overall usage might be reduced.

What not to cut
Some kinds of cost-cutting can have adverse results in quality of service that outweigh the money saved. For example, reducing help desk hours or downgrading the level of staff serving them would erode hard-won standards of support quality. In other instances, a reduction in initial costs might just result in higher costs later on. Buying computers with inadequate amounts of RAM and disk capacity could result in upgrades before the end of the machine’s life cycle. Not only is the hardware upgrade cost deferred in these cases, but labor to install it also needs to be factored in, even if those costs are not assessed and charged. Savings that do not reduce the total cost of ownership over the lifetime of the device are probably not worthwhile.

Fill vacancies
Staff positions left unfilled generally shift work to other staff. This overloading is inevitable and normal in the short run, but when a position is left vacant to save personnel costs, the side effects on those who cover the shortage add up. Staffing level is a perennial concern in IT shops; prolonged under-staffing harms not just morale but also the organization’s standards in quality and timeliness of work.

Institutional hiring freezes in times of financial restraint are usually inflexibly enforced, for fear that exceptions will be difficult to contain – and resented by others. But when key IT services are jeopardized by being pushed to unqualified staff, some remedy has to be found. Hiring help on a temporary basis might be the only recourse while the hiring ban is in effect.

The strongest argument for keeping staff positions filled despite cost pressures is that the demand for IT support continues strong and does not go down when budgets get tight. Coping with growth of services with a constant staff level is hard enough; allowing decline in staff risks damage to the organization’s credibility for service.

Training
Cutting back on technical training for IT staff is another example of false economy. Acquisition of new skills is vitally important to staff productivity. The constant growth of demand for support services can only be met by increases in the efficiency of response from IT staff. Undertrained workers spend more time on difficult tasks; they need to fall back on trial-and-error methods when they encounter problems that are beyond their competence.

Training is also a strong motivator for staff. If the IT department has taken cuts that increase stress on the staff, the removal of training opportunities only adds to morale problems. If the money is simply not there for commercial training, in-house peer training would be preferable to going without. The better course, though, would be to protect this budget line if possible.

Keep moving
Along with training, new projects are important to morale and staff retention. Sparing some of these from the budget ax is a good way to hold on to a sense that the organiz- ation is moving ahead and not giving up on innovation.

Hard times are also the logical occasion to solicit projects that could result in costs savings. These might not otherwise seem worth the priority, but if the benefit outweighs the costs by even just a little, the effort might be worthwhile. Reconfiguring servers to free up disk space needed for other applications would be an example of this kind of project. In normal times, buying new disks is the more cost-effective approach, but foregoing even modest expenditures if the time and skill for a suitable work-around is available would make a worthwhile project when money is tight.

The least harm
The choices among cost-reduction options are important because their impacts can be very different in severity. IT organizations have had to work very hard to reach stability, equipping and networking entire campuses and establishing effective help desks. The communities they serve would be very discouraged to see services slip back to unsatisfactory levels because cost-cutting resulted in reductions in quality. Exploring cutbacks that do relatively less harm is an important tactic for holding on to standards of quality.

Reductions whose consequences can be recouped later are the best choices. Lengthening replacement cycles is a reasonable cut because it does not diminish the quality of the equipment in place. In the past, hardware failures increased dramatically towards the end of a computer’s life cycle. In more recent times, the increased requirements brought on by software upgrades (especially to operating systems) has driven the replacement cycle, at the same time that devices such as hard drives and power supplies became longer-lived. In slowing the replacement cycle, there is only a modest risk of greater support costs – as long as software changes are also delayed.

The national economy will rebound, and after a time budgets will grow again. The challenge for IT managers is to analyze their operations to see which projects, services, and equipment are least damaging to reduce for a period. TW

Top of article


 

The Edutech Report is a monthly publication of Magna Publications

The EDUTECH REPORT is published each month by Magna Publications www.magnapubs.com, 2718 Dryden Drive, Madison, WI 53704; 800-433-0499. President: William Haight whaight@magnapubs.com; Publisher: David Burns dburns@magnapubs.com; Managing internal editor: Rob Kelly robkelly@magnapubs.com. Content provided by contributing editors Linda Fleit lfleit@edutech-int.com and Thomas Warger twarger@edutech-int.com. Subscription Customer Service custserv@magnapubs.com. Copyright 2004. All rights reserved. Authorization to photocopy items for specific clients is granted by Magna Publications for users registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, provided that 50 cents per page is paid directly to CCC, 222 Rosewood Drive, Danvers, MA 09123. Phone: 978-750-8400; www.copyright.com. For those organizations that have been granted a photocopy license by CCC, a separate system of payment has been arranged. One-year subscriptions: $199. Discounts available for multiple subscriptions.