top of page
Steph Lundberg

How to forecast your customer support team's capacity (and manage it well)


measuring capac

Are you prepared for your next busy season? Can you defend your staffing budget when end-of-the-year crunch time comes around? Do you know how much work your support agents can take on before they start to burn out?


Customer support capacity forecasting, planning, and management is your answer to all of those questions. 


Measuring your support team’s capacity may not always feel like the simplest exercise, but it’s mission critical. 


Good capacity planning enables you to deeply understand the work your support team is doing, what tools and processes they need to be successful, and how your team needs to grow as your business grows.


Why you need to measure support team capacity


Capacity measurement is a vital task for support leaders, both from an operational perspective and a people management perspective. 


“In my experience, there are two purposes in running capacity analysis,” says Matt Dale, CX Consultant at Moxie CX. “To get approval from company leadership for hiring or budget, and from a team management perspective to figure out how best to deploy the team, estimate inbound volume, and compare the forecast to actuals. Understanding the purpose of the exercise will help you figure out how hardcore you want to get in doing the analysis.”


As Dale points out, your executive team is going to want to know that you’re using your financial and people resources efficiently and effectively, especially in this current landscape, where tight budgets and doing more with less are common. 


And your support team needs to know you have their back – that you understand what it takes for them to offer great customer service to your customers, including building in time for rest, peer support, and professional growth opportunities.


Many organizations also have seasons where demand peaks, and you want your team to be able to responsively scale up and down to meet those demands. 


Dale notes that not properly understanding your staffing requirements ahead of customer demand can have a huge impact on your support team. 


“Understaffing can have negative consequences depending on the seasonality of your flow,” he says. “For example, at a K-12 software company our back-to-school busy season lasted from mid-September through December. Being understaffed by one or two agents meant burnout for the rest of the team because we couldn't pull folks out of the queue to go through a typical hiring/onboarding process for new teammates. 


“When we realized we were understaffed a week into busy season, we had to gut it out three more months before we could hire and onboard new teammates. We had one agent quit due to the stress and another that left a few months later.”


Solid forecasting and capacity planning helps you avoid scenarios like this. 


Key factors in capacity analysis and planning


There are a few factors and considerations that you should consider as you’re measuring your support team’s capacity:


The support channels you’re offering 


Email, chat, phone, and social media support all have different agent requirements and customer expectations. These should affect your capacity analysis. 


For instance, email support is asynchronous, which allows agents to handle multiple tickets at one time and over longer periods, giving you some cushion if you’re slightly understaffed. 


Chat is a more immediate support channel, but depending on your product complexity, your agents may be able to handle multiple chats at the same time, which also gives you some flexibility in capacity.


On the other hand, phone support is a real-time support channel. With agents only able to handle one phone call at a time, you’ll need to be very careful to measure and manage your phone agents’ capacity so that calls aren’t piling up waiting to be answered.


The coverage you’re offering


Your capacity is highly influenced by your coverage needs: do you need to offer 24/7 support? Weekend support? Multi-lingual support? Tiered support to allow for complicated escalations? 


The answers to each of these questions bring unique staffing requirements, which need to be carefully worked into your capacity planning. 


The tooling available to your team and customers


Do your agents have access to a central admin tool where they can look up information about user accounts, payments, and process refunds? Or are they looking up info from an internal database? Using a different tool for each task? Are tools like chatbots available?


Effective, well-integrated tooling has the potential to save your agents and customers time and effort. On the flip side, disjointed or dated systems can add significant time and effort into a customer interaction.


If you have any tooling projects in the works — like implementing ChatGPT into Zendesk — then you’ll want to estimate their impact on capacity and work them into your planning.


Self-service options and internal knowledge available to your customers and team


Making product and institutional knowledge readily available to your agents increases both the quality of their solutions and their confidence when dealing with customers, reducing ticket touches, resolution times, and follow-up tickets.


Likewise, even including the most basic set of FAQs can go a long way in helping users help themselves (and through that, deflecting tickets).


While it takes a little work up front, using analytics to understand your help center and making improvements to your help articles can have a big impact on future capacity planning.


Your company’s time-off policies and team schedules


Realistically, your support agents won’t be working at 100% capacity during their shifts. 


Things like vacation, sick days, parental and bereavement leave, your team’s meeting and training schedules, break and professional development policies are realities. These things are often referred to as your team’s shrinkage, and they all reduce their overall working capacity. 


You’ll need to consider all of these forms of shrinkage for your particular team and figure out the number of hours the average agent works per day or per week and factor this into your capacity calculations.


Measuring your customer support team’s capacity


The experts we talked to approach measuring and managing their support team’s capacity in a few different ways. 


Gather the data


Rebuy’s Director of Customer Support, Christian Sokolowski, recommends starting by researching what the standards are for your company’s industry. 


“Establishing benchmarks tailored to your specific business and industry is crucial. While B2B and B2C may have differing expectations, the ultimate decision on what works best for your business lies with you,” he notes. “While industry benchmarks offer valuable insights, it's important not to solely rely on them, as your in-depth knowledge of your business should guide your decisions rather than just settling for a generic industry standard.”


Sokolowski then tracks metrics by channel – in his team’s case, phone and chat.


“For phone capacity, consider tracking metrics like average call duration, call volume per agent, and response time. I measure chat and email capacity in the same way, as I have them funneled into the same ticketing system,” he says. “I suggest that they be assessed by monitoring concurrent chat sessions, response time, first contact efficiency, average handle time, time to close, backlog, and cost per chat.”


Once he has solid data in place, he develops realistic but ambitious targets for his team. “Overall, build out your service level objectives and adhere to them. People who are obsessed with metrics will identify ways to improve and meet their goals,” Sokolowski says.


Sokolowski also stresses the importance of maintaining regular communication with your support team. “Check in with them regularly to understand their sentiments and ensure you're investing in their well-being. Burnout not only erodes motivation and team culture but also incurs significant costs in terms of training new hires. Prioritize what's best for your team, even if it means making necessary investments to support them.”


Choose your tools


Emre Tekoglu, Vice President of Customer Support at Zywave, uses a popular online calculator for capacity planning. 


“We use Erlang Calculator to plan for our capacity for different service level targets like answering 90% of cases within 3 minutes for phone. We also reviewed our chat, phone, [and] email case volume. Based on the numbers, we then staff our phone, chat, and email channels for 30 minute intervals. We do that via Microsoft Excel.”


Matt Dale takes a similar approach, although he cautions against trying to build complicated forecasts in spreadsheets. “If you're going in-depth, tickets are relatively easy to calculate, [but] phones and chat require an Erlang-C calculation to spread out the inbound calls over a period of time to simulate volume.”


“In these scenarios building out the model in Excel/Google Sheets is a pain in the rumpus and should be avoided. Working with a third-party workforce management forecasting tool like Aimiable or Assembled might be worth it from a cost/benefit perspective, especially if you have limited time.”


Dale notes that Assembled offers a free Google Sheets template for forecasting and planning weekly capacity and volume.


Add in the human element


Your team members aren’t just numbers in a spreadsheet row. 


Some support leaders use a more agent-focused strategy for capacity planning. “My approach to determining support capacity is a bit unusual. I take all of the analytics and then I personalize them, bringing humanity into the calculation,” says Amiee Twigg, former Director of Customer Success at Willa.


Twigg starts by gathering data by channel, looking for the team average handle time (AHT) as well as agents with the highest and lowest AHTs for both the previous month and the previous quarter. Then she cross-references that data with the overall number of resolved tickets and the number of resolved tickets for the top and bottom performers.


She says at this point it’s important to gather information about those top and bottom performers. “Things I’m looking for: What is their tenure and how does that compare with the average? Did anything occur during the reporting period that would impact performance? For example, vacation, training on new content, personal/health issues. What does their CSAT look like?”


Like Sokolowski, Twigg says it's crucial to have regular one-on-ones with her customer support team and encourage open and honest conversations so she can add their feedback to the hard data she’s gathering. 


“For the top performer, I want to know what makes them successful and how they feel about their workload. Is it manageable and is it sustainable? With the lower performers, I want to know what they feel takes the most time and what roadblocks they are hitting to keep them from being closer to the average,” Twigg says.


“In both cases I want to gauge their stress levels. I also want to take a couple of mid-performers and gauge how they feel about their workloads, their stress levels, and talk about their roadblocks. Armed with all of that information, I calculate an average individual handle time that minimizes agent stress and ensures good customer experiences.”


Next, she looks at the team’s CSAT ratings, graphing out AHT compared to CSAT rating over time to get an idea of how much time the team should be taking on tickets to ensure the greatest customer satisfaction.


Then Twigg tracks ticket volume. She takes the average number of tickets created per customer within a set timeframe and multiplies that by the amount their customer base is expected to grow during that time period. 


Finally, she brings together all of the data and feedback she’s gathered to calculate the number of agents the team for the current season as well as for the coming month, quarter, or year to determine their baseline capacity needs.


Twigg shared some example data to flesh this out:

Average Handle Time Sweet Spot

10 minutes

Average working time per day (minus breaks and lunches)

450 minutes

Average working days per month

21 days

Monthly ticket volume

30,000 tickets

Number of Customers

150,000 customers

Customer growth rate per month

8%


From that information, she can then calculate their capacity:

Tickets per customer 

0.2

Average tickets resolved per agent per month

945

Number of employees needed this month

34 (adding 2 to account for time off)

Customer count next month

162,000

Anticipated ticket volume next month

32,400

Number of employees needed next month

36-37 (adding 2-3 for time off)


Forecasting and planning customer support capacity for pre-launch products


But what if you need to forecast and plan capacity for a product that hasn’t launched yet?


All of the expertise shared above applies in this situation as well, you’re just using it to inform your forecasts rather than guiding how you handle your concrete data.


I found myself in this situation when I took on building out the support function at Jyve, a company that matches skilled workers with grocery stores and other retail companies for stocking and facing products and other retail work. When I came on board, the app for Jyvers (the skilled workers from above) to sign up, claim, and complete jobs was still in its beta phase. 


I had the benefit of having some ticket volume from folks who were testing the app with us, but otherwise I had to estimate our capacity based on my own professional experience and other factors like industry benchmarks, product complexity, support channels we planned to offer, ticket complexity, user urgency, internal tooling, and self-service availability.


To begin, I looked up the support benchmarks for industries close to mine and compared them to the limited support data we had from our app’s beta program. 


I knew we didn’t need tiered support or round-the-clock coverage at that point, just the ability to cover weekends and the very early mornings that are typical for grocery and retail store product work. I decided on some SLAs and metrics that would allow us to track performance and serve as canaries in the coal mine for when the support team’s capacity was being exceeded. 


Again, these will depend highly on your product and team, but some things I kept an eye on were: median first reply time, average full resolution time, number of tickets created vs. solved (for backlog management), one-touch tickets percentage (for monitoring ticket complexity), and our SLA achievement rate. 


I also considered that we wanted our agents to have time to contribute to our fledgling help center, and that the executive team eventually wanted to offer text and phone support as well as email support.


Based on these analyses, I estimated that we could cover our immediate needs and have some breathing room left over for new channels or special projects if we hired two full-time support agents. I also concluded that we’d be in especially good shape if we could hire agents with experience in our pretty niche field – something we were actually able to do with some targeted recruiting.


That meant that we were able to shorten our onboarding process to training our agents on our app and processes. We found that since they already knew our industry extremely well, they were actually able to give us advice on how to best serve our users.


Keep your team’s humanity in your calculations


It may seem daunting to calculate and plan your support team’s capacity, but by using the reporting you already have and taking advantage of the tools and tips above, you’ll be on your way to responsibly shepherding your team’s time and energy.


Most importantly, remember what all of our experts agreed on: when calculating your support team capacity, remember that you and your team are humans who want to do their best work. 


Be curious about their experience, what might be blocking or powering them, and trust them to advise you on what both they and your customers need most.


 



Written by Steph Lundberg

Steph is a writer and fractional Customer Support leader and consultant. You can usually find her crafting, hanging with her kids, or spending entirely too much time on Tumblr.


Comments


bottom of page