Workforce management basics
Good workforce management begins with good forecasting. This means predicting volume for every support channel by day and time interval (often down to fifteen minute increments). Best practices are to use at least two years of historical data and run it through algorithms that predict the future volume. Workforce managers can use spreadsheets (though cumbersome and not ideal) or sophisticated workforce management software to support this complex task.
In addition to history, forecasts also need to reflect known business plans that will affect contact volume, such as promotions, catalog drops, upcoming holidays, and infomercials. Once everything is accounted for and the forecast is complete, it's used as an input to hiring plans and agent schedules. Workforce management staff revisit volume forecasts often and update them to reflect trends that are different from projections.
Despite everyone's best efforts, sometimes forecasts are widely different from actual volume. Volume could be over forecast if the company website goes down causing an influx of service calls, or it could be under forecast if the big promotion is a dud. Additionally, events like snowstorms can wreak havoc on agent attendance, rendering the existing schedules mostly useless. This is where intraday management comes in to try to salvage the plan.
In a scenario where a gap between actual volume and forecast is affecting contact center performance, workforce management staff would reforecast volume based on what they're experiencing and then adjust schedules accordingly. Intraday management can also include activities like calling more agents to come in, sending agents home, and assigning agents to skill sets they don't typically work. Contact centers are dynamic places and strong intraday management practices – coupled with automated intraday capabilities from workforce management software - give them the flexibility needed to respond to changing conditions.
The forecast will help workforce management teams determine how to schedule agents, but there are other factors that need to be considered. For example, the forecast may show that 245 calls are expected during a specific day interval, but whether they're expected to last 2 or 10 minutes will have a big impact on the number of agents that need to be scheduled for phone support. Therefore, average handle time needs to be factored into the scheduling.
Additionally, workforce management analysts need to consider skill sets when scheduling agents. For instance, when that infomercial runs, agents with a "sales" skill set will likely convert more callers than those with a "service" skill set, so more sales agents should be scheduled. As if scheduling isn't complex enough, digital channels further complicate the process. For example, each agent will have a different proficiency regarding the number of emails and chats they can have in process at one time. This will affect throughput and needs to be accommodated by the schedules.
And of course, we haven’t even considered the human factor yet - individual agent schedule requirements and preferences. For example, one agent may need to leave by 4 pm to do school pick-up, while another agent may be unavailable to work Tuesdays because they have college courses.
Luckily, like forecasting, scheduling can be automated and simplified with workforce management software. The best tools will also help with calculating how well agents are adhering to their schedules, which is a necessary input to determining the effectiveness of agent schedules.