From “Bird Box” to contact center.
If we are to believe the numbers that have been published recently, if you are a US Netflix subscriber, then you probably have seen the movie “Bird Box”. For those who have not, this recent Netflix original drama-thriller is about a force, entity, or beings (that’s up for debate) that if you look upon them will cause you to — well let’s just say, really bad things happen to you. I’m not going to critique the movie here, as I will leave that to other forums and others more eloquent than I. But, I do want to share with you a scene in the movie that impressed me and is relevant for contact centers.
In this particular scene, some of our protagonists have to go to get supplies from a grocery store near the house in which they are taking refuge from the deadly forces outside. However, they can’t allow themselves to look outside while they drive to the store lest they risk the deadly consequences. Their solution? Conveniently, there is a car in the garage, so, they black-out ALL the windows of the car, including the front and back windshields, with newspaper and paint, they plan the route before leaving (exit garage, one left, two rights, straight, then left . . that sort of thing) and then use only the GPS to guide them while driving there (well, in addition to adjusting when they hit or run over things). Somehow, they actually get to the grocery store and back, most of them anyway . . .
The daily crapshoot.
Now imagine this is your contact center. You plan your call volumes, you plan your agent schedules, and then you hit “play” on the operation and hope you hit your service level targets (while meeting FCR, customer satisfaction and employee engagement targets in the process I might add). Then, the next day, you look at how you did the prior day, adjust the next day’s plan, and hit “play” again. How do you think this will work out? Like our cinematic heroes, you might or might not get close to where you want to go and back, but definitely not in the most graceful or efficient way possible. Surprisingly, as my experience consulting with contact centers has shown me, quite a few contact center operations operate not too far away from this scenario. Is yours one of them? Sub-optimal workforce management can mean you are at the very least leaving money on the table or at best missing opportunities to improve your results.
Workforce management is a must in any contact center operation. Whether you leverage an advanced tool or plug numbers into a giant Excel worksheet, almost all contact centers have to do this function one way or another. However, as a favorite former boss of mine used to say, in between black and white there are lots of shades of grey, and so too with workforce management, there are levels of sophistication that determine how much return you are actually getting for your operation.
Defining workforce management.
First, let’s define workforce management in a contact center context so we are all on the same page.
Workforce Management, or WFM for short, is an integration of technology, processes, and management that allows a company to optimally predict and allocate human and technological resources to handle contacts or other work tasks that are received.
There is definitely math and science behind workforce management, but truly successful programs treat WFM more like art. First, WFM is an integrated approach. There is nothing here about it just being a technology tool or a role in the contact center. Best practice says that systems, process, and people need to be aligned. This is especially the case for WFM. Next, it implies optimal prediction of demand, and subsequently, allocation of resources (agents, voice lines, IVR capacity, etc.) to handle that demand at any given point in time. Underestimate the demand or overestimate your resources, and you start getting a backlog of calls or work, which translates under most circumstances into unhappy customers.
Admittedly, WFM most directly impacts accessibility (how easy customers perceive it is to get in touch with customer service), which is not the end-all-be-all of contact center management; your agents still need to resolve the customer’s request in a timely and effective manner, for example. Nevertheless, since it also deals with agent scheduling, and matching the demand for service to the supply of resources to deliver the service, WFM is the metaphorical grease that keeps the engine running smoothly and efficiently, ultimately allowing you to do what your customers ask of you in a timely and efficient manner.
The 7 ways.
The following is a list of seven (7) practices that will help you raise your workforce management game. If you already do these things, then you are most likely already achieving the most tangible benefits from your WFM program. If not, implementing them should bring you a noticeable improvement to your operational KPIs. In one extreme case I actually worked on, in an operation of about a thousand seats, implementing proper WFM methodology resulted in an increase in service levels from 60% to 90% (answered within 30 seconds in this case), greater agent satisfaction (less stress, more sense of control), and greater customer satisfaction (customer interactions with agents were no longer doomed before they even began with angry customers who had been impatiently waiting to talk to someone) – all of this without an increase headcount!
1. Put in place workforce management KPIs.
Not to be confused with KPIs for your contact center operation, these KPIs tell you how well your WFM processes are running. What is your forecast accuracy? What percentage of your forecast deviation is due to unexplained factors vs explained factors? Do you track agent adherence before or after schedule adjustments? With what frequency are there deviations to your demand planning calendar? What percent of agents do schedule trades on a regular basis? What does it mean if they do? How close are you on a daily basis to your overall service level targets?
2. Review staffing parameters and rules to optimize agent engagement while maintaining flexibility for business needs.
What are the rules that govern how shifts and tours (groups of shifts over a period of time) are allocated to your agents? Are tour lengths so long and inflexible that your management team has a hard time adjusting schedules to meet demand? Or are your tour lengths so short that agents are complaining about not having any way to plan their personal lives? Have you taken into account agent preferences in the scheduling process in a formal manner? Can agents opt to swap shifts with someone else? If so, what are the rules? How much agent self-service is available in the schedule management process?
If you haven’t already, review your scheduling criteria both from a demand standpoint (do you have enough flexibility in your schedules to adapt to changing demand patterns) and from a human standpoint (what is the work-life balance being offered to your agents, and more importantly, what are the established rules that agents understand and inherently agree to regarding schedule flexibility, assignments, changes, priorities, preferences, etc.) and update them with clear, transparent and intentional rules that will facilitate flexibility while maintaining agent engagement.
3. Optimize your predictive demand model with input from other parts of the business.
Using historical patterns to predict future patterns of contact demand is a good place to start, but it does not cover the whole story. Marketing campaign messages and timings, media ad schedules, product launches, billing cycles, weather predictions, and even earning announcements can impact both contact volumes and the distribution of those volumes over the course of a month, week, day or hour. Work with your partners in other areas of the business to formalize a process to obtain relevant information on demand-impacting activities on an annual, quarterly, monthly, weekly and even daily basis. Doing this will improve your forecast accuracy and reduce to a minimum the unexplained variances in the demand forecasts which in turn drives fewer last-minute adjustments in staffing plans.
4. Give agents and supervisors tools that provide visibility of what’s going on with the queues.
You are probably familiar with those moveable electronic signs on the side of the road that flash as you drive by while telling you are going 40mph in a 25mph zone. The immediate response of most people is to at least lift their foot off the gas pedal as they check their actual speed. Mission accomplished. In the contact center, it is no different. Agents and managers alike need the visual indicators that tell them how the overall center is doing as well as how individually they are doing vs. targets. Wall displays are one way, queue and agent information on the phones is another. Some new technologies are emerging that allow these indicators in a visually easy-to-read way on agent desktops, allowing agents the opportunity to self-manage their time (and their contribution to the team queue) and their supervisors the ability to manage larger spans of teams. This does not end all behavioral issues, but it does help those individuals who are acting in good faith to self-manage their efficiency as well as to adapt to dynamic changes in the needs of the team.
5. Create standard protocols that guide the priorities and activities of the center under different conditions.
While “Red alert” may conjure up images of Star Trek for some, it may do well for contact center life to imitate art. In Star Trek, when a red alert is declared, not only does it signal a dangerous situation, but it also triggers a set of protocols and responses from the crew: engage the weapons systems, begin charging the defensive systems, take your predefined posts. Let’s extend this concept to the real world in a contact center context. A few of the centers I have worked with have created precisely these conditions, some examples (names of the states have been changed):
- Condition blue for systems problems (affected agents use pre-determined alternative “systems” or processes to provide continuity of service),
- Condition red signaling a severe overflow state. In this case, all non-essential off-line work is ceased, and every available agent is called to open up to answer calls or chats, or join “the line” to serve whatever workload is currently overflowing,
- Condition purple which means that actual demand is far below predicted. In this case, training, back-office tasks and even voluntary time off can be prioritized over agents being “open” to calls.
Having predefined scenarios coupled with a set of protocols to follow if that scenario were declared lessen the response time of the various moving parts in the center allowing it to adapt to changing conditions more rapidly.
6. Create and maintain a centralized force desk or command center.
The more seats you have in your operation the more important it is to have a centralized place where workforce management is coordinated and monitored in real time. While there is a cost of creating and staffing such a function, the payback in team efficiency, especially for larger centers with more dynamic demand conditions, is in terms of weeks not months. While there are different approaches to staffing the force desk, using actual supervisors to spend part of their time in a rotation for that role helps keep them “on the front line” so to speak, as well as keep them in tune of what’s happening in the whole center as opposed to just their team or queue. It is also a great training ground for agents who are looking to ascend into management.
7. “Drive” in real time.
Surprisingly, one of the biggest missed opportunities for workforce management is actually managing the center in real time. With the force desk as the centralized eyes and ears of what is going on in any given moment with both demand and supply on all channels, force desk staff should have both the criteria and the authority to decide and execute on changes to staffing, shifts in staffing among channels, declaration of blue, red or purple states, and prioritization of off-line activities. Using our driving analogy yet again, you wouldn’t think of driving a car with a blindfold on, yet running a center without real-time management is doing just that. Driving a center in real time allows you to tweak out optimizations in the allocation of resources to demand in hourly or half-hourly periods which adds up along the course of a day or week to quite a few person-hours for training, back-office work, coaching time and other activities.
While most of what has been discussed here are most applicable to inbound activities, many concepts are applicable to outbound activities as well. Instead of planning around peak hours of demand, you might plan for peak hours when agents have the best contact rate or close rate for example.
Let me know your thoughts or some of your workforce management tricks and tips. Happy driving!