Contact Center Forecasting Fundamentals #2: How to Master Workforce Forecasting

Charles Watson Forecasting Jul 19, 2018 17 min read

In the first blog article of our Contact Center Forecasting Fundamentals series, we covered different methods to forecast workload. That is, in fact, the first step in the process. However, despite workload forecasting, it’s equally important to invest time in forecasting the workforce required to staff against the demand you’ve identified in the first step. Ultimately, the objective is to not determine the mere requirement, but to uncover the gap of actual vs requirement in staffing. The work you do in forecasting the workforce will then drive a more accurate picture of existing staffing gaps and helps you to cover each of them to increase efficiency in Workforce Management.

Let’s start with the top 3 mistakes many planners make when forecasting the workforce in their contact center.


Top 3 Mistakes to avoid in Workforce Forecasting

 #1 Not reconciling headcount

Headcount reconciliation should be done at least monthly, if not more often. Understanding exactly how many people you have gives you a baseline to work with.

For scheduling, you should really reconcile weekly, but as we focus on forecasting, monthly headcount reconciliation will give you what you need to identify the gaps for your hiring training plans.

#2 Not properly accounting for attrition

Attrition is really challenging to forecast in small organization. If you have 50 people in a staff group, and project 1% attrition per month, that’s a half person.

Do you reduce by 1 person or 0? Neither is technically correct, and may impact whether you’re overstaffed or understaffed.

And don't’ forget, you have both voluntary and involuntary attrition.

Do you have insights into how many agents will be terminated due to the attendance policy or through performance management? 

#3 Not having staff groups set up correctly

In contact centers, you have groups of calls and you have groups of agents. Usually, you have a group of calls that go to a primary group of agents, but then you have some overflow, or cross utilization that takes place.  

Ideally, you have higher skilled teammates that can help down with simple call types, for example. When you have complicated overflow and cross-utilization setups, it’s easy to lose sight of what a staff group actually is.  

Is it a group of agents or is it the aggregate of a particular set of calls? Remember, when you forecasted demand, you forecasted for calls.


Forecasting doesn't need to be hard

or time-consuming!

Discover the brand-new injixo ForecastFully-automatic and always up-to-date.

Automated contact center forecasting for calls, chat, email and more. 


Learn more


Now, let’s explore key elements of a workforce forecast:


Key Elements of a Successful Workforce Forecast

As you can guess from the tips above, reconciling headcount is the first step in the process. You have to know exactly how many headcount and FTE (Full Time Equivalents) you have.

To keep things simple, any time I use FTE, I’ll assume a 40-hour work week.  But it varies based on location and company. Leverage your company’s HR system to get the headcount.

That should always be the starting point. Some workforce management teams only track agents taking inbound calls. Others will track coaches, team leaders, etc…

If you track the job types in the HR system that you align to, it makes it a lot easier to use that to reconcile with your agents in the workforce management system. Be sure to reconcile headcount at least monthly, if not weekly.  

This is the foundation for your staffing forecast.


Accounting for Full Time vs. Part Time in your Staff Requirement

In many markets, everyone works full time.  If this is the case, then tracking the number of part time employees vs. full time isn’t necessary.  

Where you do have people working less than 40 hours, this needs to be considered. Your demand forecast (from our last blog) provides requirements FTE, so your supply of agents has to have the same methodology.   


There are a few ways to factor in part time workers: 

1. Have all part time workers as 0.5 FTE.  

This is the simplest method and requires only that you know they are part time.

This works best if you have a small percentage of part timers or don’t have the bandwidth to keep track of a lot of part time workers.  

It’s less accurate, but gives you a directionally correct number so you don’t overstate your workforce

2. Have part time workers in the plan based on their actual scheduled hours.  

This is the more accurate method and unless you have a lot of changes, then the work is in the initial set up when you add an employee.  

This is pretty straightforward to calculate. Just take the number of hours they work and divide it by the number of hours that make up an FTE in that market.  

If a person works 24 hours and your FTE is 40 hours, then the math is 24 divided by 40, which is 60%. So this person would be in your staffing model as 0.6 of an FTE


Consider Voluntary Attrition in Planning

Voluntary attrition occurs when an agent leaves the company voluntarily.  Often, you’ll see capacity plans with just “attrition” in it.  

In those cases, you’re blending together two very different drivers to attrition - voluntary and involuntary. Planning for these independently allows you to adjust your assumptions based on the different factors. 

When you plan voluntary attrition, you’ll look at historical data to see what the trends are telling you. Much like with forecasting volume, you can look at how the rate has changed year over year and apply any seasonality.  

You may notice that attrition goes up after people get end of year or quarterly bonuses. You may also notice a change in voluntary attrition based on the tenure of the population.  

New hires (less than a year) generally have a higher attrition rate than employees who have been with the company longer. So you’ll want to determine the attrition rate for new hires as well as tenured employees.  

A good way to build the model is to look back at the last several years (I like 36 months, based on experience, but it varies by industry and how stable your business is).  

Track the attrition rate for employees less than a year and then do the same for 1+ years. Use this to form your model for each population.

The simplest way to do this is to average each of the 3 years by month (e.g. average of Jan 2015, 2016, 2017 becomes the January model).  

When you’ve done this for each month, you now have a full year model. Then to get to the total voluntary attrition, you have to weight those based on the population.


Example: Attrition rates for January 

Step 1: Establish a model for attrition for each month


Step 2:  Determine the weighted average for the population for January



Repeat this process for each month.  It’s a simple process, because you can just build this in excel with the formulas and just copy/paste.

Involuntary attrition is completely different from voluntary attrition.  

This attrition takes place when employees are terminated by the company, usually for violating policy or from the performance management processes.

While this ultimately impacts the same number (how your current staff attrition down to your future staff), the driver is different, so you need to forecast for it separately.This is a tougher one to forecast, to be honest.  

Look at the historical data gives you a baseline. But what that data represents is a result of both your policies and the enforcement of your policies.  

Let’s say your company has had an absenteeism policy where if an agent calls out unplanned 3 times in a month, they are terminated.

Then they change it to 2 unplanned absences results in termination. Your historical data is based off of one policy, but going forward is a new policy that is more aggressive.  

In theory, that would signal you to increase your involuntary attrition going forward. But as i mentioned the other layer is how it’s actually enforced.

This is where you really need to understand why the policy is in place (or in this case, changed), and how the operations plans to enforce it.  In this example, if the objective is to send a message, but not actually terminate more people, your forward-looking assumption probably shouldn’t change.

And if they do intend to enforce it, then you’ll need to look at the number of employees who call out unplanned twice a month vs. three times a month.  If the number is the same, then essentially the policy won’t terminate more people and your assumption will need to stay the same. 

So as you can see, the work to determine involuntary attrition can be complicated, and requires a lot of engagement with your partners in operations.  It also requires interpretation. But the payoff is huge in terms of getting to the right number of staff.


Lastly, let’s talk about staff moves.  

This is an area that’s often problematic as well.  

When an employee moves from one staff group to another, do you have a systematic way to track and report the change in your workforce management system?  

Better yet, does your operations team let you know when a change has been made?

You need to make sure you have a clear process for having these communicated to your workforce management team. Make sure you include situations where they don’t move team leaders, but take a different call type. Your intraday performance reports have to account for the correct call type.

You’re looking for any adds (new hires), moves (moving from one call type/group to another) and deletes (terminations).  For new hires and for terminations, ideally you get connected into your HR system, because that’s the system of record.

This same process can be leveraged to report any change in status. For example, if someone moves from part time to full time.



The Must-have Guide to Contact Center Forecasting


>> Download now <<


In Summary

Forecasting your staff is key to a well-built staff plan and often doesn’t get the focus it deserves.  Take the time to really build this out.

You should have regular meetings on this topic, and if it doesn’t take up a significant amount of your (or your team’s) time, then you’re probably not spending enough time on it.  

It’s also important to note that the forecasting error for this can be large, because you rely so heavily on insights and observations in addition to the historical data.

Measure your historical accuracy and keep that in mind as you’re making decisions. If your forecast accuracy for the workforce is +/- 1%, you can make strong business decisions off of it.

If the accuracy is +/- 20%, then it’s just guidance, and you have to assume a high probability that you’ll have a different number of staff then planned. You’ll need flexibility options (overtime, voluntary time off, schedule changes, etc..)

>> Also read: 5 Mistakes that might cost your contact center forecast accuracy.


What's up next:

Now that you’ve forecasted your demand and the supply of agents, you’re ready to start incorporating best practices to closing the gaps.  

In our final blog in the Contact Center Forecasting Fundamentals series, we’ll cover techniques for efficiently aligning staff to demand both short and long term.

We’ll look at how to determine the best method to cover staffing needs for a few weeks or several months.

Check out part 3 now:

Contact Center Forecasting Fundamentals #3: Closing the Staffing Gap



The new automatic injixo Forecast

With self-learning algorithms for the best available forecast. Considers trends, holidays and events.

Save up to 75% of your current forecasting time.

Learn more




Free webinar on February 28th
How to efficiently manage service level with the right technology
Join now!