Workforce planning in the contact centre: the basics

#Why contact centre planning is different

In most departments, the rota decides when work happens. In a contact centre, customers decide when work arrives. Monday morning differs from Thursday afternoon, the start of the month from the middle, and an outage changes everything within minutes.

Workforce planning, often called workforce management, brings these two worlds together. It makes sure enough people with the right skills are there at the right time, without teams being permanently over- or under-loaded. This guide explains the basics and shows where to start.

#The cycle at a glance

Workforce planning is not a one-off plan but a cycle. Each round produces data for the next. Once you see it that way, you stop looking for the perfect plan and look for one that gets a little better every week.

1. Forecast. How many contacts arrive per channel, skill and interval?
2. Requirement. How many people does that take, measured against the service level?
3. Schedule. Who works when, within the law, agreements and preferences?
4. Intraday management. How does the day deviate from the plan, and what do we do about it?
5. Review. What do we learn for the next forecast?

Weekly schedule

CoverageService level ✓

MonTueWedThuFriSatSun

Early

Late

Flex

Shift swap

Thursday late for Friday early

Approve

#Roles in planning

In small contact centres, the team lead plans on the side. Beyond a certain size it pays to separate the tasks, because they need different skills. What matters is that every task has a named owner.

• Planning: builds forecasts and schedules and keeps the rules current.
• Real-time management: watches the day, spots deviations and proposes measures.
• Team lead: puts measures into practice, plans coaching and knows the people behind the figures.
• Employees: state preferences, swap shifts and report absences through self-service.

Planning is a service for the teams, not their counterpart. Where that attitude is missing, the schedule is experienced as an order from above, even when it is professionally sound.

#The key terms

Planning often fails not on the maths but on terms that different people understand differently. Agree on definitions before you compare figures.

The glossary entry on workforce management explains how these connect.

#Forecasting: how much work is coming

The forecast is the foundation. It estimates contact volume and handling time for every interval, split by channel and skill. Intervals of 15 or 30 minutes are usual; daily totals are too coarse for scheduling.

History is the starting point. It shows patterns across the day, the week and the year. At least a year of data helps make seasonal swings visible. Clean out outliers such as outages, or the forecast will plan the next failure in for you.

#Planning in events

History alone is not enough. Public holidays move, campaigns are scheduled, invoices go out on fixed days. Such events should enter the forecast as factors in their own right, not as manual corrections after the fact.

Ask other departments about planned events regularly: marketing about campaigns, billing about mailing dates, IT about maintenance windows. The post Forecasting with events shows how that works day to day.

#From volume to staffing requirement

Volume and handling time give the workload of an interval. But if you schedule exactly as many people as the workload requires on paper, customers wait a long time. Contacts do not arrive evenly; they arrive at random intervals.

That is why telephony has used queueing models for decades, the best known being Erlang C. It calculates how many people are needed to reach a service level at a given volume. Planning software does the calculation for you.

The example shows two things. First, a good service level needs headroom above the raw workload. Second, shrinkage is one of the strongest levers in the whole calculation. Set it too low and you are short-staffed every single day.

#Digital channels work differently

Email and tasks follow a different logic. They wait until someone has time. Instead of a service level in seconds, there is a response deadline in hours. Planning therefore asks how much work must be done by the deadline.

Chat and web messaging sit in between. Employees often handle several conversations at once. How many is a quality decision. Plan it deliberately rather than leaving it to the system.

#Multiple skills and sites

The example above assumes one queue. In practice, contact centres have many queues, and employees master several of them. Someone who can handle billing and technical requests helps wherever demand is.

That flexibility is valuable, but it makes planning more complex. One person cannot cover two queues at once. Planning software therefore considers each person’s skills and distributes demand so that multi-skilled employees are used sensibly.

Sites and home working matter too. Different public holidays by region, different opening hours and teams in several cities belong in planning as rules. The customer on the phone does not care where someone sits; the plan does.

#Scheduling: who works when

The requirement per interval becomes the schedule. It distributes shifts, breaks, training and team meetings so that demand is covered as well as possible. A schedule is good when it meets demand and is also liveable for people.

#Legal framework

In Germany, the Working Hours Act sets the frame. Working time is generally eight hours per working day and can be extended to ten if the average over six months does not exceed eight (Section 3 ArbZG).

More than six hours of work requires a break of at least 30 minutes, more than nine hours at least 45 minutes (Section 4 ArbZG). Between two working days there must be at least eleven hours of rest (Section 5 ArbZG).

A planning system should know these rules as constraints.

#Co-determination on the schedule

The start and end of daily working hours, breaks and the distribution across weekdays are subject to co-determination under Section 87(1) no. 2 BetrVG. In many German contact centres, a works agreement therefore governs shift models, notice periods and how preferences are handled.

That agreement belongs in planning as a set of rules. A schedule that is optimal on paper but breaks agreed notice periods is not a good schedule.

#Preferences and swaps

Schedules are accepted more readily when employees can shape them. Preferred shifts, a swap board and leave requests through self-service take load off team leads and make the plan more reliable.

Clear rules are what matter: a swap is allowed if skills and demand stay covered.

The post Shift swaps as self-service describes rules that work for both sides. In Univerx, employees see their schedules, swap shifts and submit requests through self-service and mobile access, described under employee experience.

#Intraday management

No day goes to plan. Volume deviates, someone calls in sick, a conversation runs long. Intraday management compares plan and reality continuously and reacts early.

Typical measures are moving breaks within permitted windows, bringing training forward or postponing it, switching skills and offering callbacks instead of waiting. For larger deviations, a prepared playbook helps, as described in the post When everyone calls at once.

#Putting adherence in its place

Adherence measures how closely actual presence follows the plan: whether someone is logged in at the planned time, on a break or in training. It is a planning metric, not a performance metric.

A deviation can have good reasons, such as a long call just before a break.

Because real-time adherence makes individual behaviour visible, in Germany it is subject to co-determination under Section 87(1) no. 6 BetrVG. Agree who sees deviations, how long they are kept and whether they may feed into performance conversations.

#Assessing the planning itself

Planning needs metrics too, and different ones from the teams. They show whether forecast and schedule do their job. A handful of figures, reviewed together regularly, works best.

Pay particular attention to the service level per interval. A good daily average can hide the fact that customers wait a long time every morning and nobody calls in the afternoon.

#Capacity planning: the long view

The schedule looks weeks ahead; capacity planning looks months to years ahead. It answers questions such as: how many people do we need next winter? What happens when a new product launches? What does a new channel mean for staffing?

Capacity planning works with scenarios. It links expected volume with onboarding time, attrition, leave planning and budget. That shows when new people must be hired so they are trained in time. See capacity planning for more.

#Quality and development in the cycle

Planning does not end with the schedule. Scheduling training, coaching and quality reviews gives employees time to improve, which lowers handling time and repeat contacts over time. Those slots belong firmly in the plan, not in the gaps.

In Univerx, quality management and performance & development close the cycle: recording with consent, evaluation forms, scorecards, coaching and learning modules per person. Reports are designed to be compatible with works council agreements.

#Common mistakes

Many workforce planning problems repeat from one organisation to the next. They are rarely about the software and usually about the assumptions behind the plan.

• Underestimating shrinkage. Leave gets planned; sickness, training and team meetings get forgotten. The plan looks covered and is not, every day.
• Planning daily totals only. A day can add up overall and still be too thin in the morning and too generous in the afternoon.
• Handling time without wrap-up. The time after a call is work. If it is missing from the plan, it is missing from staffing.
• Occupancy as a target. Permanently very high occupancy exhausts teams and drives up absence. It is a result of the plan, not a goal.
• Adherence as a performance measure. Using adherence to rate individuals costs trust and invites conflict with co-determination.

#Where to start

Many contact centres still plan with spreadsheets that have grown over years. Moving to a planning solution works best in small steps. This sequence has proven itself.

1. Review the data. Volume and handling time per interval for at least a year, cleaned of outliers.
2. Fix the terms. Service level, handling time, shrinkage: one definition for everyone.
3. Start with one queue. Build the forecast for one area and compare it with actuals week by week.
4. Capture the rules. Working-time law, works agreement, shift models, skills.
5. Open self-service. Enable preferences and swaps for one team and review the rules together.
6. Expand. More queues and channels, then capacity planning.

#Checklist

#How Univerx Workforce helps

Univerx Workforce joins the parts of this guide into one cycle: forecasting per channel and skill from history and events, scheduling with a swap board and preferred shifts, real-time adherence and occupancy, quality and development.

Quality and performance reports are subject to co-determination, and recordings require consent. Both can be reflected in Univerx. The guide Running a contact centre in Europe shows how to prepare with privacy and the works council. To look at your planning together, talk to us.