How to Build a Call Center Staffing Model
“Please hold. You are a valued customer and an agent will be with you shortly.”
Everyone hates waiting on hold. But if you’re a fast-growing company, you also can’t hire unlimited customer service agents to ensure you always have an agent available to help your customers.
What are you supposed to do?
This dilemma highlights how staffing a call center can be a huge challenge for most customer service leaders. If you offer phone support, figuring out the right call center staffing model can make or break the experience you’re providing for your customers. And with 83% of customers saying that good customer service is the number one factor for them in deciding what to purchase, learning how to staff your call center is truly a business-critical issue.
In this article, we’ll cover everything you need to know to build out your staffing model:
- The benefits of an accurate call center staffing model
- The Erlang C staffing model (the most popular way to determine your staffing needs)
- Strategies to improve your staffing as you scale
The benefits of an accurate staffing model
Why do you actually need a staffing model?
That might seem like a simple question, but it’s important to think through. No one creates a staffing model just to have a fancy spreadsheet they can show off when they’re done. Your staffing model is meant to drive hiring decisions. It’s a tool to help you build the right team for your call center and business.
If you’re like most customer service leaders, you don’t just have a blank check that allows you to hire as many people as you want. You’ll need to be able to make a strong case that shows how many people you need to hire, when you need to hire them, and—most importantly—why you need to hire them.
Your staffing model is a foundational piece of how you answer those questions.
There are three lenses you can use to understand the benefits of a sound staffing model.
Your staffing model benefits your customers by forcing you to think through the level of service you want to provide them.
Without an accurate staffing model, your customers are likely to experience inconsistent service. The first time they call they may get to an agent immediately, then the next time it may take thirty minutes. Talk about frustrating. Your staffing model should help you reduce the likelihood of situations like these happening.
If you’re staffed correctly, your customers should experience a consistent level of service and minimal wait times.
An accurate staffing model for your call center brings a lot of business benefits. Three key benefits are worth mentioning here:
- Cost-savings. Hiring too many agents for your call center is costly. If you overhire, you’ll be wasting money. But if you don’t hire enough agents, you’ll burn out your existing team and have issues with employee retention. Your staffing model should help you avoid each of these extremes.
- Scalability. Your staffing model is a forecast. It shows you how many agents you need today and how many agents you’re likely to need in the future as your business grows. This enables you to plan ahead and determine how to scale your support team efficiently.
- Brand reputation. Bad customer service means a negative reputation and lost business. Your staffing model helps you stay off lists like this by helping you deliver the level of service your customers expect.
Support team benefits
Your support team benefits directly from your staffing model in two ways, both of which have been hinted at above.
First, an accurate staffing model should ensure enough bandwidth across your team to avoid burnout. Call center agents aren’t robots. They need time to breathe between calls and to take breaks. You can give them this time by accounting for it in your staffing model formula.
The second benefit for your support team is related to customers. Because your staffing model helps you deliver a consistent customer experience and avoid long wait times, your support agents won’t have to deal with customers who are irate from sitting on hold forever. This should equate to shorter support interactions and more pleasant conversations, which makes every support agent’s job more enjoyable.
The Erlang Model
The Erlang C formula may be the most common call center forecasting formula. Although it’s over 100 years old, it’s still used today by call center managers all over the world.
Erlang C has been tweaked over the last century to include additional variables like occupancy and shrinkage (more on these below). Its usefulness stems from its ability to model the relationship between your staffing, call volume, and response time. These three factors are fundamental to every call center’s daily operations, which makes it applicable across the board.
Here’s the bottom line: Erlang C will help you understand how many agents you’re likely to need during a given period, based on the data you supply.
It can’t always account for outliers or unexpected situations — like a major outage — but assuming your inputs are accurate, it should give you a sound baseline for staffing your call center on a normal day.
There are a number of Erlang model calculators available for free online, but to make the most of them you’ll want to read on to make sure you understand the inputs they rely on:
- Call Centre Helper’s Erlang calculator
- Agenses’ Erlang C calculator
- Planetcalc’s Erlang C calculator
Erlang C formula inputs
There are three primary inputs for the standard Erlang C formula:
- Call volume. The number of calls your call center gets in a given time period.
- Average Handling Time (AHT). The amount of time it takes an agent to handle a phone call. This includes both the time on the phone and any necessary wrap-up time before they’re available for the next call.
- Service level. The percentage of calls you want to answer within a given timeframe. A common baseline has always been answering 80% of calls within 20 seconds, but this is really determined by your company’s support philosophy and any contractual agreements you’ve made with customers.
You’ll also want to make sure that you’re looking at the right time interval.
For instance, say you get 100 calls per hour. If the formula is looking at 30-minute intervals and you enter 100 calls as your normal volume, then the staffing projections it recommends will be way higher than necessary (because 100 calls in 30 minutes equals 200 calls per hour, making it look like you have double your actual call volume).
Some of the more recent additions to Erlang’s formula—available in some calculators—include occupancy, shrinkage, and average patience.
- Occupancy rate. Occupancy rate is the percentage of time your agents spend handling customer calls versus waiting for calls (idle time). While you might think 100% occupancy is the goal, that’s a sure recipe for burnout. Many experts say that an occupancy rate of 85-90% is ideal for most call centers.
- Shrinkage. Shrinkage is a measure of the things that take agents away from their ability to answer customer calls. This includes things like team meetings, sick days, vacation, and training. It’s usually expressed as the percentage of time an average agent won’t be available due to shrinkage (e.g. 25%).
- Average Patience. Also known as Average Time to Abandon (ATA), Average Patience is the average length of time a caller will stay in your call queue before hanging up.
Many of the top phone support platforms make it easy to pull reports that include these metrics.
Strategies to improve your staffing as you scale
Learning how to staff a call center isn’t rocket science, but it’s also not as simple as plugging data into a calculator and doing exactly what the calculator says. The Erlang C formula gives you a great foundation to understand how many agents you may need. But how should you use that data to inform your approach to supporting your customers?
You need to understand how to combine Erlang C data with your company’s unique values to help you scale your customer support effectively.
Here’s a proven approach that will help you achieve this successfully.
Step 1 – Figure out your approach to customer service
Every company says they care about their customers.
But when push comes to shove—or when you take a look at their budgets or customer reviews—it quickly becomes clear that there is a wide range of approaches to customer service across brands and industries.
Some companies view customer service as a cost center and a necessary evil. These companies focus on minimizing their customer service expenses, often leading to frustrating or negative customer experiences.
Other companies view customer service as a differentiator and a driver of growth. These companies invest heavily in tooling and resources for support teams. They are strategic about the customer experiences they create. They find ways to leverage a great support experience to drive additional revenue for the business.
While you might like to say we’re the latter type of company, that may not be the case. Ask yourself questions like these:
- How does our executive team view customer support?
- How much money are we investing in customer service and how has that changed over time?
- What percentage of our revenue is given to supporting customers?
- If we were starting from scratch today, what kind of experience would we want to create for our customers?
Questions like these can help you better understand your current reality and your ideal future state. For instance, maybe you recognize that right now support isn’t a priority, but you’re able to get consensus across your leadership team that in the next several years you want your support experience to be a differentiator.
Step 2 – Let your findings inform your Erlang inputs
Use the information you’ve uncovered from your question asking to inform the numbers you plug into your Erlang model.
Perhaps right now you’re only answering 70% of calls within 30 seconds, but you recognize you’d eventually like to answer 90% of calls within 20 seconds. That’s a significant change, and it’s going to have a big impact on your staffing needs. Use an Erlang Calculator to understand what the difference in staff required, and then work out a plan to staff up so you can achieve that goal.
Step 3 – Recognize the levers you can pull to influence staffing needs
None of the Erlang C inputs referenced above are set in stone.
If you view each input as a lever, you can adjust each one as needed to influence your staffing needs and the resulting customer experience you create. The only exception may be Average Patience, as that’s something that’s difficult for any call center manager to control (although if you can minimize wait times, this becomes less important).
- Call volume. As your business grows, call volume may grow. But by implementing self-service best practices and building out a knowledge base, you can reduce call volumes significantly. This means fewer agents are needed. You may also be able to use the budget you’ve freed up (from having fewer agents) to invest in other ways to improve your customer experience.
- Average Handle Time (AHT). Training your call center agents is a great way to improve customer service productivity. The faster agents can solve customer problems, the lower your AHT becomes and the more efficiently you can scale your support. Be cautious about pushing AHT too low, because if customers feel like your agents are pushing them off the phone, it may create a negative perception.
- Service level. You can increase or decrease your service level targets to influence your staffing need. Answering more calls faster creates a better experience, but it also requires more agents and costs more money. You’ll need to figure out how to balance these factors to create the experience you’re aiming for.
- Occupancy rate. The occupancy rate you aim for dictates how much ‘breathing room’ your agents get between calls. Increasing occupancy can lower costs, but if occupancy is too high then you’ll likely see higher employee attrition rates.
- Shrinkage. Less shrinkage means agents are available to help customers more frequently. You can adjust your shrinkage by being strategic about when you pull agents away from the phones. Can you shorten meetings or deliver training more efficiently through a quick video? While this can mean less shrinkage, make sure you’re giving your team enough time to rest in order to prevent burnout.
Step 4 – Leverage outside expertise when needed
Business process outsourcers (BPOs) like Peak Support exist to help companies grow efficiently without sacrificing customer service quality. Any good BPO will bring significant experience to the table regarding how to staff and grow your call center as you scale.
For instance, at Peak Support, the results of the erlang C calculator are just the start of modeling staffing needs. From there, we’ll do things like apply the results across a client’s hours of operation, divided into thirty-minute intervals. We’ll then apply a linear optimization model to this data, which gives us more precise visibility into how many support agents are needed during each interval. We’ll layer in additional data to make the most of your agents’ availability while reducing the need for overtime and extra pay.
If you’re already a guru at creating staffing forecasts, more power to you. But if you’d prefer not to reinvent the wheel or could benefit from a little help, you can leverage BPO expertise to help you achieve your goals.
If you want to explore your options to grow your customer service team with Peak Support, contact us to set up a discovery call today.