Which KPIs Do I Need for Contact Centre WFM?


We compile a set of key performance indicators (KPIs) for contact centre workforce management (WFM), highlighting how they are best used and why they are so important.

Shrinkage

It is vital at all stages of the WFM planning cycle that you are looking at your shrinkage levels.

Often, we talk about getting shrinkage down, but you need to remember that it is more than sickness and holidays.

When building a forecast, it is important to know what levels you are using in your model and if they match reality.

It is good practice to show both options in your model, e.g. if we hit targets, this is the result we can expect vs. on current numbers we expect another result.

During scheduling it is important to add as much shrinkage as possible in advance so we can optimize and reduce the possibility of it being pulled.

Real-time analysts need to be aware of shrinkages as they are happening. No one should be taken away from customer contact without the real-time team being consulted.

Real-time analysts need to be aware of shrinkages as they are happening. No one should be taken away from customer contact without the real-time team being consulted.

In the review stage, you need to monitor shrinkages to ensure they are happening. While you may want to drive down sickness levels, it is important that holidays are being used up so you don’t have an issue at the end the year.

I thumbnail photo of Ian Robertson

Ian Robertson

It is also vital that training, team meetings and coaching are happening at the correct levels. When these don’t happen, you will see sickness and attrition rise.

In any customer operation it is not unusual to have 50% of an advisor’s time taken up by shrinkage, so you need to focus on this as much as any other KPI.

Thanks to Ian Robertson at The Forum

Service Level

Service level is the target that you set yourself, as a contact centre, in terms of the percentage of contacts that you plan to answer in a specified time.

However, it’s also a KPI that you measure in real time, to ensure that your target is being met. This gives you an insight into the effectiveness of your resource planning, but it also helps in terms of intraday management.

Service level performance is also an important consideration when putting other KPIs into perspective, such as customer satisfaction and shrinkage.

Why shrinkage? Firstly, many contact centres do take shrinkage into account, overlooking the importance of tracking and planning for all the “subcategories” of shrinkage.

A thumbnail photo of Lauren Comer

Lauren Comer

It is not good enough to know that your contact centre experiences 20% shrinkage. Rather, you need to know the breakdown of that shrinkage – that 3% is due to advisor tardiness, 8% is due to training, 4% to meetings, etc.

Understanding the subcategories can help you plan and conduct what-if trade-offs and better come to terms with how that specific shrinkage affects service level performance.

Thanks to Lauren Comer at NICE inContact

Attrition

Attrition refers to the loss of employees by process of resignation, retirement, dismissal, health issues or other similar reasons.

This KPI is used for capacity planning purposes and should feed into your recruitment strategy.

When calculating attrition, it is vital to gather the full 360-degree picture by including internal attrition within this calculation.

When calculating attrition, it is vital to gather the full 360-degree picture by including internal attrition within this calculation.

When employees move out of your contact centre, whether that be externally or internally to other areas of the business, they need to be backfilled. Without taking internal attrition into account, you are not gaining a true attrition figure for your business.

Ensure you put in place a process which captures the full picture and prevent challenges later down the line!

Advisor Proficiency

The degree of expertise and competence at an individual level can vary greatly across the agent spectrum.

You have the developing proficiency level of a new starter at one end of the spectrum, and the other end of the spectrum a long-tenured, highly competent, expert-level advisor.

Reflecting levels of proficiency when scheduling resources and being careful not to assume all agents are equally proficient is key.

A thumbnail photo of Neil Draycott

Neil Draycott

Why? Because if you need ten agents to meet demand at 4pm today, but four of the people are scheduled to work at half pace (e.g. because they are new) then these four individuals only really make up two of the ten you need.

But remember, don’t just think “new starters”. Many overlook the need to reflect a lower proficiency when an experienced agent acquires a new skill. They will go through a transition period before reaching full proficiency and this should be reflected.

Thanks to Neil Draycott at Business Systems

Occupancy

Occupancy represents the percentage of time an advisor is in an available status, ready to handle an interaction, or actually handling an interaction. This includes wrap-up time after an interaction has ended.

The KPI doesn’t include breaks, lunches and coaching time, but, nonetheless, it will never reach 100%, as an advisor will not take contact after contact – they need to have small breaks in between. If they don’t, advisor burnout will quickly become an issue.

The KPI doesn’t include breaks, lunches and coaching time, but, nonetheless, it will never reach 100% as an advisor will not take contact after contact – they need to have small breaks in between.

So, occupancy is an important measurement to provide clear visibility into how well your expected volume aligns with your advisor schedules.

You therefore need to target an occupancy standard and calculate staff requirements against that, with 80-85% being an industry standard.

Utilization

Utilization is the measurement of the percentage of advisor time that is spent logged-in, handling and expecting customer interactions, while in the contact centre.

This KPI is often confused with occupancy, but it instead helps us to determine the percentage of time that advisors are available to handle contacts within the time that we pay them.

A thumbnail photo of Patrick Russell

Patrick Russell

This is an important consideration, as your utilization figure is used as an input when calculating both shrinkage and overhead costs.

High utilization will deliver a lower cost per interaction, but high utilization can damage advisor morale, if there is nothing in place to break up long spells on the contact centre floor.

Thanks to Patrick Russell at 8×8  

Average Handling Time (AHT)

AHT is the average time it takes to handle a call or transaction from start to finish. This is from call initiation, through hold time and talk time, and all the way to any related tasks an advisor must perform post-phone call to resolve that call.

To calculate AHT, add your total talk time + total hold time + total after-call tasks, and then divide by the total number of calls. That is your average handle time.

Thumbnail image of Frank Sherlock

Frank Sherlock

From a WFM perspective, it’s vital to know your full AHT for staffing purposes, but also to know what is happening at each of these phases. This helps you to identify the potential for improvement, whether that be shorter or longer calls, and plan your resources accordingly.

By analysing the gap in handling time between different groups of advisors, you can identify how workflow can be improved at each of the steps that combine to create AHT.

Thanks to Frank Sherlock at CallMiner

Forecast Accuracy

Forecast accuracy is a percentage measure of how accurate your projected contact volumes were in comparison to the actual volumes, over certain time periods throughout the day.

The measure will help you to assess the success of your forecasting models, helping you to understand if you correctly predicted when contacts would be made and whether or not you were prepared for them.

A thumbnail photo of Peter Dempsey

Peter Dempsey

If you keep a record of our forecast accuracy, you can assess how well you incorporate annual spikes in contact volumes and continually make improvements to your forecasting model.

When you know your organization inside-out and have a record of what happened at the same time last year or in similar circumstances, it will help you to know what lies ahead and how to deal with it.

Thanks to Pete Dempsey at Puzzel

Abandon Rate

Abandon rate measures the percentage of callers that hang up before they speak to an advisor.

Clearly, you don’t want many customers abandoning with their issues going unresolved, as that will likely have a significant impact on customer satisfaction and revenues.

You need to be testing your service levels against your abandon rate, to ensure that it’s suitable.

So, you need to be testing your service levels against your abandon rate, to ensure that it’s suitable, while it is also good to dig deeper into your abandon rate figures and plot abandon curves.

By plotting abandon curves, you can use your analysis to calculate another KPI, “Average Patience”, which is the point in time in which 50% of your callers will hang up.

While this may not seem significant at first, it is in fact a key input into staffing calculations on the voice channel, if you use an Erlang Calculator for doing so.

Schedule Adherence

Schedule adherence measures the percentage of time an advisor works in comparison to the time they are scheduled to do so.

The KPI doesn’t allow for any time compensation. So, if the advisor arrives five minutes late and leaves five minutes late, their schedule adherence will not be 100%. The five minutes after the scheduled shift has finished is not included in any calculations.

Due to the nature of call arrivals, it is important that advisors are working the exact shifts that they are assigned to better meet demand. Schedule adherence therefore becomes an important measure to track, as “making up the time” logic does not apply to the contact centre.

If schedule adherence takes a hit, that will also impact on service level and occupancy, potentially damaging both the customer and advisor experience.

If schedule adherence takes a hit, that will also impact on service level and occupancy, potentially damaging both the customer and advisor experience.

By measuring schedule adherence, you can look for patterns in what may be causing issues like lateness, while you can also take the KPI down to an individual level, to highlight which advisors may individually be impacting the success of WFM plans.

Absenteeism

Absenteeism is another metric that highlights the impact that certain individuals have on your planning through calculating the percentage of unauthorized time advisors are away from the contact centre floor.

Yet this isn’t a KPI that’s only used for monitoring staff. You can split the measure across different teams to highlight where your problem areas are and use this analysis to address these areas accordingly.

If absenteeism is a big issue within your contact centre, you should also be factoring that into your staffing calculations.

A thumbnail photo of Charlie Mitchell

Charlie Mitchell

So, if you need 100 members of staff to cover a reporting period and have a 5% absence rate, logic says that you should be scheduling 105 advisors to best meet demand.

However, just remember that absence will naturally peak and dip at different times throughout year, so it is good practice to forecast absence and apply the changing rate into our staffing calculations.

Thanks to Charlie Mitchell at Call Centre Helper

For more from our panel of experts on the topic of contact centre KPIs, read our articles:


Get the latest exciting call centre reports, specialist whitepapers, interesting case-studies and industry events straight to your inbox.