We’ve all been there. The forecast looked accurate, the staffing plan put the right people in the right place at the right time, and yet you still missed the mark on your service level agreement (SLA).
As a workforce management (WFM) professional, your leadership team expects you to trade in your crystal ball for a magnifying glass — it’s your job to put the moment under a microscope to figure out what the heck happened.
When everything looks right but somehow turns out wrong, it’s usually due to one of three things. In this article, we’ll discuss what those things are and how you can isolate to identify the root cause.
Your forecast was wrong
Predicting the future is a tricky endeavor, and even the most well-calibrated of models will occasionally miss the mark. If the base forecast you’re relying on for your staffing plan is off, your corresponding staffing plan will be off to a similar degree.
Common culprits for forecast inaccuracy include:
- Outlier events, such as site outages, product recalls, or new product launches. These events can cause inaccuracy in the moment by driving contact volume much higher or lower than expected. They can also have a longer-range impact on your forecast (depending on the method) if you fail to account for the outlier in future planning (i.e., you probably won’t have a product recall on the same day and time next year).
- Not tweaking your forecast for meaningful variables. Meaningful variables will, of course, vary (it’s in the name, after all) by team. But common items include growth curves, seasonality trends, contact to [revenue unit] ratio, etc.
Variables in your staffing model need to be tweaked
While forecasting accounts for the demand side of the supply-and-demand equation, there are a lot of nuances to consider on the supply side. It's not easy to take into account human behavior across a large number of people, so your staffing model will change constantly — but that's part of the fun, right?
Common culprits for variable-based inaccuracy include:
- An unrealistic shrinkage assumption. If you find that you consistently fail to meet your SLA goal when your staffing plan indicates you should, that could be an indication that your staffing plan isn’t accurately accounting for the call-outs that happen the day of. An unexpected deluge of PTO requests is a familiar bane of existence to WFM professionals everywhere — but if you find yourself feeling the pain too frequently, you may need to pivot your shrinkage assumption.
- An inaccurate average handle time variable. Things that can strongly influence this include 1) If a large amount of your team is training on a new process, you can expect a higher average handle time (AHT), and 2) If your usual ticket increases in complexity, you can expect a higher AHT.
It’s important to note that forecasting outliers often happen when something unexpected — like a site outage, recall, or controversy — happens. Teams generally aren’t as trained to handle these rare events, making them inherently more complex. And again, as complexity rises, so too does AHT.
Your agents need coaching
WFM is all about getting the right people in the right places at the right times to deliver stellar customer experiences. But that’s not the whole story. What WFM is really about is getting the right people in the right places at the right times doing the right things. At the end of the day, all the effort that goes into WFM means absolutely nothing if employees aren’t doing right by customers.
Common culprits for agent behavior that can have a big impact on SLA include:
- Poor adherence. When an agent is scheduled to be doing the right thing but they aren’t actually there, it really throws a wrench into the works. You can have the perfect plan, but if the plan isn’t followed then you’re basically gambling on your customers’ satisfaction.
- Low solves history per unit of time. While adherence offers a great bird's eye view of the issue, it’s certainly possible — depending on how your team has operationalized adherence — that an agent could also contribute to SLA misses while still showing high adherence on the surface. This is why it’s important to look at both solve history per unit of time and adherence to get a more holistic picture.
Like most things related to WFM, the answer to why you’re missing SLA is probably not going to be that straightforward. In most cases, it’s not just one thing that derails your operation. Rather, it’s the result of several factors converging together to create an overall failure.
From that vantage point, investigating the exact cause can feel like a daunting task. Keep in mind that every team misses SLA on occasion and that sometimes things are out of your control. But there are also things within your control. By digging into past misses, you can reduce the chance of missing SLA again in the future.
