What is scheduling?
WFM is about having the right number of people in the right places at the right times, doing the right things. Scheduling is the part of the WFM cycle where agents are assigned to shifts and tasks so that workload fit is maximized.
Why does scheduling matter?
In a perfect world, the supply of agents always matches the demand for agents. There would always be exactly the right number of people needed to service customers within the desired service level. Schedule efficiency, also known as ‘workload fit’, is a measure of the extent to which the ‘supply’ of agents matches the ‘demand’ for agents. The goal is perfect coverage, i.e. zero under-staffing and zero over-staffing at all times.
The unfortunate fact is that unless you schedule your agents properly, you’ll be over- or under- staffed most of the time. Forecasting and staffing calculation are important, but if you don’t schedule agents to match the calculated demand, all your effort so far was wasted.
How does scheduling work?
- Rostering: The process of creating schedules for employees without regard for the underlying staffing demand.
- Rotas: A rota is a repeating pattern of shifts, typically used to rotate working hours so that each employee is asked to work an equal number of ‘popular’ and ‘unpopular’ shifts. For example, in a business which is open from 8am to 10pm, an employee may have a rota consisting of 8am to 4pm in week one, 11am to 7pm in week two and 2pm to 10pm in week three, repeating from week four onwards.
- Optimized scheduling: A scheduling method that aims to consistently minimize under- and over-staffing by matching the number of scheduled agents with the number of agents required, as closely as possible. This entails selecting the optimum start times, finish times and (optionally) break times, for all employees. Optimized scheduling maximizes schedule efficiency. It enables the consistent achievement of service level goals, while also minimizing employment costs and avoiding agent stress and burnout.
Optimized scheduling: the challenge
Schedule optimization is not a trivial task. Let’s look at some scenarios.
Scenario 1 (simplified)
- Scheduling 25 employees for 5 days
- 1 activity
- Possible starting times for employees: 8:00, 9:00 or 10:00
- With shifts of equal duration
That is 325 or about 847 billion possible schedules. That’s about 100 times the size of the Earth’s entire population. Or twice the number of stars in the Milky Way galaxy.
Scenario 2 (enhanced flexibility)
- Scheduling 25 employees for 5 days
- 1 activity
- Possible starting times for employees: Between 8:00 and 10:00 in 15-minute intervals
- Employees can work 4, 5, 6, 7 or 8 hours
That is 4525 possible schedules, i.e. a number with 41 digits. For comparison, the number of atoms in a typical human body has 27 digits.
Scenario 3 (realistic)
- Scheduling 100 employees for 5 days
- 10 different activities (including breaks)
- Starting times for employees are at 15-minute intervals between 8:00 and 16:00
- The length of shifts varies between 4 and 8 hours in 15-minute intervals
- Each employee can perform 5 different activities per day
That is 33,126,489100 possible schedules. That’s an almost unimaginably large ‘solution space’ and finding the optimum is like finding a ‘needle in a haystack’. But the challenge doesn’t end there.
To build a viable schedule, planners need to take into consideration all the relevant constraints, including:
- The working times specified in agents’ employment contracts
- Labor laws, e.g. minimum time off between shifts, maximum working hours
- Agent availability or preferences - if this is part of your agent engagement strategy - see Employee Engagement
- Agents must be scheduled for activities for which they have the necessary skills. Multi-skilled agents may be scheduled to perform different activities at different times; this is known as ‘block scheduling’. Alternatively, agents may be scheduled on a blended basis, as described below
It is possible to allow agents to state their preferences or ‘bid’ for shifts by breaking down the process of optimized scheduling into four steps:
- Create schedules that are aligned with demand but are not yet assigned to agents.
- Give agents the opportunity to state their preferences within the offered shifts, or ‘bid’ for shifts
- Allocate agents to shifts, respecting their bids as far as possible
- Allocate the remaining shifts to agents, while respecting the constraints
The goal of shift bidding is to combine optimization with agent engagement. It does however have some downsides. It requires much more effort by the planner. Three-stage optimization typically results in slightly inferior workload fit to that which one-stage optimization achieves. Agents need to be educated about how the bidding process works, to avoid the perception of lack of fairness. For example, if 10 agents bid for an attractive shift of which 5 are available, how are the ‘winners’ and ‘losers’ selected? Realistically, shift bidding is not possible without professional WFM software.
Multi-skill blended scheduling
As explained in multi-skill staffing calculation, employees with more than one skill will naturally spend less time waiting for work that they are qualified to handle. The greater the extent of multi-skilling, the fewer employees are required to handle the workload. This effect is known as ‘pooling efficiency. To unlock the pooling efficiency of multi-skilled agents in a contact center, three components are required:
- A communications platform that routes contacts to agents in real-time according to specified rules. A typical automatic call distributor (ACD) is configured to route each incoming call to the available agent with the required skill who has been idle longest
- Blended schedules, in which agents are scheduled to be present to handle any contact for which they have the necessary skills
- A scheduling method that generates optimized schedules, taking into account all the constraints and the multi-skill pooling efficiency. This is extremely complex and requires algorithms that are only available in powerful WFM applications
Breaks, lunches, and meetings
It is good practice to schedule breaks during the working day. Letting agents choose their own break times is empowering, but it does have potential drawbacks. If every agent went on break at the same time, service level and customer experience would be terrible.
Introducing optimized break and lunch times has a significant positive impact on workload fit, even in contact centers where agents are working on fixed rosters or simple rotas. By the same token, workload fit benefits from the optimized scheduling of meetings, training sessions, and 1:1s.
In a perfect world, every agent would be fully multi-skilled and be willing to work any shift assigned to them. In the real world, staff turnover means that universal multi-skilling can never be achieved. And it’s rare to find a contact center with no agents working on legacy fixed-hours contracts. The good news is that even if just a subset of your employees can be optimized, there will be a disproportionate improvement in efficiency. Employing a ‘mix and match’ of fixed shifts, rotas and optimized scheduling is a smart and practical move.
What impact does scheduling have?
Scheduling is where most of the magic of WFM happens: It’s the part of the WFM cycle where agents are assigned to shifts and tasks so that the right number of people are in the right places at the right times, doing the right things. Scheduling is not a trivial task. The number of permutations and the complexity of the constraints mean that building optimized schedules is practically impossible without professional software. But without effective and efficient scheduling, the goals of WFM cannot be achieved.