It’s pretty obvious that the number of agents who work for your IT service desk should have a direct positive or negative impact on its performance. And yes, adding more agents may be one solution to correcting unsatisfactory service issues. But it often isn’t. Before making that expensive investment, your hospital’s leadership should examine several operational variables that are likely to be more significant than a simple head count.  In this post, we explore some pattern-based factors you should consider in calculating ideal service desk staffing levels within your hospital.  

An obvious first step in your analysis is a review of call volume over several months; the second is a drill down into call volume patterns from day to day. From there, it will be much easier to understand the meaning of the metrics that have collected on your key performance indicators (KPIs).

I regret that the following may not read like your favorite novel, but the ideas and concepts here may very well save your hospital money and increase user satisfaction with your service desk.

Call volume:

• Begin your analysis by calculating daily and weekly volume of calls over a time sample of at least three or four months that overlap seasons and/or yearly quarters. Why such long-term samples? In-patient, out-patient, surgery and emergency room activity can vary greatly between these periods, and not necessarily in the same way. For example, surgeries may decrease in the last quarter of the year due to holiday schedules, but then go back up in the first quarter. On the other hand, ER traffic may go up in the last quarter due to increased accidents, flu, etc, and then return to lower levels after the first of the year.
  These are examples only; your location,  community demographics, climate, hospital specialties and other factors are likely to differentiate your hospital’s patient care volume patterns from others’.
  Why does all this matter? Patient volume = treatment volume, which affects levels of EHR and other systems’ usage. Volume of systems usage may not exactly equate to call volume in your service desk, but is a key factor.
• Implementation of new systems is another major component of call volume with which to reckon. Hospitals are continuing to implement new EHRs and other systems; many already are replacements of systems that were deployed to meet Meaningful Use / MIPS incentive requirements. Any organizational plan to install new systems should be considered in predicting changes — usually increases — in user demand for IT support.
  As we have seen, these predictions are tough calls. Of course, user support needs will increase for a few months with any major new implementation, but your best prediction of upcoming call volume must also consider the complexity of the new system, the quality of user training, the availability of supportive super-users and even the attitudes of your physicians towards adapting the system or asking for help in using it. For example, I recently talked with a group of hospital physicians who stated flatly that they never call their service desk. As the chief resident said,” If we can’t figure it out, we just find a work-around.”

Short term call volume patterns:

Any CIO knows that staffing the service desk is not as simple as calculating call volume over months.  Move deeper into the weeds — look at  call volume patterns over hours and days, i.e. workflow. Observe those patterns over months. Short term call volume patterns have a unique relevance to staffing level requirements and can help explain your KPI numbers —  particularly speed to answer, call abandonment and first call resolution metrics. These are critical elements in service desk quality management, and if they are unsatisfactory, they point to a need for a change. (If your hospital is not measuring these KPIs consistently, your service desk needs more than a change — it needs an overhaul!)

The following graph identifies call patterns for one of our service desk clients, as reported in our ticket tracking system. The pattern shown is typical for a healthcare service desk – peaks of call volume at the beginning of shifts, primarily from nurses as they arrive for work, with calls from physicians tending to occur just before and after their office hours when they are rounding on patients, and a dip in calls during the lunch period.

Such a call pattern represents the single largest challenge hospitals have in staffing their service desks; if you staff enough full time resources to efficiently handle the beginning and end of shift peaks, the agents experience significant idle time during the middle of the shifts. On the other hand, if you staff primarily for the middle of the shifts, agents are overloaded at peak periods.

What happens when peak call volumes exceed your capacity during these peaks?  

All three of the KPIs noted above are affected:

• Speed to answer: This is measured by the average number of seconds it takes a service desk agent to connect with a caller. A typical metric is the percentage of calls answered within a specified time period, say 30 seconds. If all agents are already on the phone, no one is available to answer the next call, and the time to answer increases.
• Call abandonment: This is the percentage of all service desk calls in which the caller either hangs up or is disconnected. The slower the speed to answer, the greater likelihood that callers will hang up instead of waiting for service. This metric is less affected than the other two, primarily because callers with access issues for key systems like the electronic medical records tend to remain on the phone if they cannot perform their jobs until the issue is resolved.
• First call resolution: This occurs when the service issue is resolved by the service desk agent who received the call without a hand-off to another resource, e.g. a hospital clinical analyst. First call resolution, usually measured as a percentage of all service requests, is affected by volume patterns because, during peak call periods, service desk agents may not have the time to work exceptionally complex tickets. Agents are more likely to document an issue, and return to it after the calls remaining in the call queue are answered.

The graph that follows, based on data provided by Penny Reynolds, in her article Calculating Support Center Staff Requirements, shows how call answer times are impacted by increasing the number of staff in the call center.

In the table below, call answer times are correlated to performance metrics. See our report on service desk key performance indicators for more information. In this example of a 300 bed hospital, we assumed typical Service Desk performance for a staff of four full-time employees.

Number of Staff vs. Performance Metrics

You can clearly see the improvement in going from an understaffed service desk with only three agents, to a generously staffed service desk with six. If this hospital with four agents and a first call resolution of 80 percent wishes to improve first call performance to 90 percent, one solution would be to add another agent with the attendant salary costs at $45K per year plus fringe benefits, and other costs.

However, unless this is other work for one or more agents during your “valley” call periods, you will have lost service desk cost-effectiveness, and not really eliminated the original call pattern problem described above. For more on how to estimate total service desk costs, read our  report describing key factors to be considered when estimating Service Desk costs.

Other Options for Right-sizing Your Service Desk Staff

There are various ways to build flexibility into your service desk staffing to address irregular workflow without paying full time staff for significant idle-time.

• Use other staff to help cover the peak call volume. This idea can fix the staffing problem, but it does so at a cost. The “other staff” is usually systems analysts —  typically more highly paid staff to support areas in which they are not as well trained as full-time agents. These employees must drop whatever they are working on — often projects that have deadlines and valuable deliverables.  Ironically, user satisfaction and first call resolution may actually suffer, while costs may increase.
• Hire more service desk staff and provide “off-phone” work for some of them during volume valleys. An example would be to assign certain agents routine network support or admin tasks that are not overly time-dependent.This is usually a less expensive staffing strategy than the above, but can create unexpected roadblocks. Your facility must provide space and additional workstations to accommodate the extra staff.  These agents also will need extra training and different management for the “off-phone” work, which can become cumbersome.
• Arrange shift overlaps to cover the beginning of shifts with staff from the previous shift. This can be effective, but again requires extra space and work stations. Sometimes this solution does not provide sufficient flexibility to totally resolve the problem without incurring overtime expenses and staff dissatisfaction.
• Use part-time staff to cover the peak call times. This is a potential option, but the part-time staff, again, requires additional space, as well as the same licenses and training as the full-time staff. Also we have seen that part-time service desk staff experiences greater turnover that full-time employees. Generally, extensive use of part-time agents is not cost effective.

One other option:

Consider outsourcing to a specialized onshore healthcare IT service desk that spreads its operations over multiple time zones, thus eliminating the “peak and valley” service effect described above.

By considering outsourcing service desk functions, you are in good company. The 2015 Black Book Survey on provider use of outsourcing indicated that over 75% percent of hospitals are currently outsourcing IT functions, and service desk is a strong element in the mix.

At the Phoenix 24 x 7 on-shore service desk, we provide outsourcing of service desk operations to hospitals in five different U.S. time zones, giving us the advantage of having the “start of shift” peaks spread out across the entire shift. This allows us to provide the appropriate level of staffing to our clients at a significant cost savings. We pass that cost savings on to our clients.

If you’re interested in exploring outsourcing your service desk, let us know. We would be happy to do an assessment of the cost-effectiveness of outsourcing your service desk to help you determine if it’s the right solution for your hospital. Find out more about our services here or contact us.