Staffing Ratios for Support: Not the Best Way!

Quite often, HDI gets some variation on the question, “What’s the right ratio of support analysts to employees for internal support?” It’s a fair question, and one that goes back as far as there have been support centers, but it’s not the right question.

Consider the following scenarios:

Organization 1 has 3,500 employees in a highly complex and proprietary technical environment. On average, each employee contacts the support center 10 times per year (.83 times per month). The average handle time per contact is 23 minutes, by phone or email.

Organization 2 has 3,500 employees in a straightforward environment, using commercial off-the-shelf (COTS) software. On average, each employee contacts support twice per year (.16 times per month) and average handle time is 7 minutes, by phone or email.

Each organization has the same number of employees. Do they need the same number of support staff? It’s pretty obvious that they don’t, so a ratio is useless in this case (and in many real-world cases).

So, what’s a better way?

HDI recommends a Gross Staffing Model based on some simple calculations. Here’s what you need to know:

• Your total contact volume
• Your average handle time
• The number of hours (on average) a support center analyst is actually available to work with customers

HDI members have access to the Gross Staffing Calculator, which does the math behind the scenes based on some simple input from you.

In Organization 1, we can say that the total contact volume per year is about 35,000 (3,500 x 10). Each contact, on average, lasts 23 minutes, so the total amount of work for the analysts is 805,000 minutes (23 x 35000), or 13,416.7 hours.

In Organization 2, we can say that the total contact volume per year is about 7,000 (3,500 x 2). Each contact, on average, lasts 7 minutes, so the total amount of work for the analysts is 49,000 minutes (7 x 7000), or 816.7 hours.

Given an 8 hour workday (yes, that includes all breaks and lunches—we’ll subtract them out later) for 5 days a week and 52 weeks a year, each person has a 2080-hour work year. You can already see that 1 analyst should easily be able to handle the workload at Organization 2, with plenty of time for projects and other tasks, while at Organization 1, it will take something closer to 8 analysts.

To get a more exact figure, take out lunches (30 minutes per day, or 130 hours per year), breaks (2 x15 minutes per day, or 130 hours per year). Already, our 2080 hours is down to 1,820. Then take out sick days (let’s say 5 days, or 40 hours averaged over the 260 day work year, or .154 hours per day) and the same for vacation, holidays, training days, project work, and so on. What we’ll wind up with is roughly a 6-hour day—that’s a 1,560-hour work-year—for each analyst to be in the chair and conversing with customers via their assigned support channel.

Now we can say that Organization 1 needs about 9 (8.6 to be exact) analysts to handle the contact volume.

Which ratio is right? 1 analyst for 3,500 users/customers, or 9 analysts for 3,500 users/customers? Answer: They are both right. That's why ratios are not a good way to calculate staffing levels.

You can stop thinking about ratios and start asking How many people it will take to get the work done? That’s a much better question.