Best of ICMI - #8: It’s Time to Revisit Service Level

Best of ICMI in 2021 - #8

Can your customers reach the services they need, when they need them?

Many organizations are grappling with severe labor shortages—a combination of the waning pandemic and economic resurgence. Contact centers have not been immune. In fact, as I write I am in a virtual queue for a callback from an airline (the expected wait is 6 hours.) The work from home (WFH) arrangements that are still largely in place have helped, and some contact centers have fared far better than others, but we expect challenges to continue for some months.

This is a great time to revisit an age-old topic in contact centers: service level. Providing access to the services customers need through the channels they choose goes to the heart of what contact centers do. Service level—ensuring contacts get to the right place in a reasonable amount of time—is central to that effort. Yes, the quality of service delivered is what ultimately matters most; but, you can’t even get started until contacts get to the right places at the right times.

The following are important principles of managing contact center service levels. Revisiting and building them into both strategic and day-to-day operational decisions can help you navigate this challenging season. They’re key to ensuring your center is both efficient and accessible:

Categorize contacts correctly

Service level has a specific definition: “X percent of contacts answered in Y seconds,” e.g., 90 percent answer within 20 seconds. It is a concrete and stable objective for contacts that must be handled when they arrive. Response time is the related objective for contacts that don't have to be handled at a specific time (e.g., email, social posts that do not have to be handled immediately, and others).

Differentiating between service level and response time is essential for two reasons. First, base staff calculations vary for these two major categories of contacts. Second, deferring work that doesn’t have to be handled as it arrives can provide much-needed breathing room. Yes, you have to manage it well and within customer expectations, but categorizing work and being smart about when you handle it can lead to much better results.

Apply appropriate staff calculations

Because of random arrival, base staff requirements for contacts that must be handled when they arrive must be predicted by using either a queuing formula that takes random arrival into account or computer simulation. To calculate the staff required for contacts that do not have to be handled when they arrive, you can generally use a more traditional approach to planning. For example, if you have 60 customer e-mail messages to process, and they require an average five minutes handling time, you have 300 minutes of workload to fit into staffing requirements.

Ensure service levels are in parity across contact channels

Being in parity in this context doesn't necessarily mean being equal—e.g., it doesn't mean you need to respond to email as fast as you answer calls. It does mean, however, operating within general customer expectations across contact channels. A customer who expects a reply to an email or social media post within a few hours but doesn't get it may call. Similarly, if a customer ends up in an endless telephone queue, they may send an email or contact you through another channel. Assessing and meeting expectations across contact channels will keep your center in balance.

Manage service levels by increment (interval)

Consistent performance by interval (e.g., half hour), is one of the telltale signs of a well-run contact center. Daily, weekly, and monthly summary reports often conceal problem areas. Teach your team to think, plan, report, and manage in terms of what’s happening throughout the day. That is key to consistent performance.

Don't force occupancy to unrealistic levels

A well-worn principle is that when service level gets better, occupancy (the time agents spend handling contacts versus waiting for them to arrive) goes down. Therefore, average contacts handled per individual also will go down. But remember, this "unproductive" time is sliced into 12 seconds there, two seconds there, and so on - the result of random arrival. Don't try to force occupancy within an increment to be higher than what base staffing calculations predict it will be, or you will jeopardize service level.

Ensure that budgets reflect workload dynamics

Today's customers demand user-friendly self-service systems, as well as the means to reach well-informed and capable representatives when and as they need them. Yes, it's important to do everything possible to provide and encourage customers to use automated support alternatives where they make sense, and, even better, prevent contacts all together through better, simpler, products and communication. However, it is important to be realistic about demand for agent-assisted services.

Support service levels with effective workload planning

Effective planning involves the totality of forecasting, staff and system calculations, scheduling and real-time management. The most successful contact centers have an established, systematic planning process; they revisit scheduling options often, involving their agents in the process. Also, good planning provides benefits far beyond schedules—it is a catalyst for collaboration across the organization that effective customer services require.

---

Service level does not guarantee a great customer experience, but it is an important enabler to delivering services your customers need and expect.