How to optimize agents experience using design thinking
Customer service teams are facing increasing pressure to do more with less, respond to more customer inquiries with less people while maintaining or improving customer satisfaction. Under-deliver in any of these points and you end up with rising costs or poor customer experience. The constant pressure will also negatively impact your agents’ satisfaction leading to higher employee turnover. In addition, having unhappy employees is highly correlated with poor customer experience, as it has been shown by many research studies in the past.
Improving the workflow the agents use to resolve customer issues will increase productivity, ease off the pressure, boost agents satisfaction and ultimately customers experience. In this article, we will show how to apply a design thinking methodology to improve the agent workflow of your customer service department.
Design thinking is an approach used for creative problem-solving originated from UX design but applied widely in many other fields, including architecture, engineering and business. Its focus on understanding the user (agent) and the problems the user is experiencing makes it an excellent candidate for agent workflow improvement. Also, because it involves the user at any step, your agents become part of the solution and they’ll be more open to change as a result.
The Design Thinking process can be broken down into a number of steps or phases: Empathize, Define, Ideate, Prototype and Test. Next we’ll explore how each of these can be applied in a customer service setting.
Empathize with your agents
The first step and most important one is to understand the problems that agents are experiencing with their workflow. In design thinking, understanding requires empathy which is obtained from engaging with your agents and observing their environment. This can be done in the office or remote, using video calls and screen sharing.
Select a number of agents to cover a wide range of experience and skills, from entry level to most senior. Schedule time and sit down with them. Make it as informal as possible. Keep an open mind and be clear that you are not evaluating their performance, but trying to understand the challenges they have in resolving customers issues. You need the agent to be himself and open to an honest discussion.
Start with one new customer ticket or chat and ask the agent to walk you through the resolution step by step. Encourage him or her to think out loud. Listen carefully to what he or she is saying and observe the steps taken. Take as many notes as possible – you will be referring back to them later on.
These are a couple of things to look for.
Can you spot any friction in what the agent is doing? Are there many steps he would go through to achieve something? Look at the screen. Does he have more than he needs there to solve a ticket? Any additional piece can be a distraction.
How many tabs does he have open? How often does he switch tabs? Does it have to go to other tools to lookup information on behalf of the customer? Does he need to manually move (copy/paste) data from one place to another? What kind of information he has to fill in while working on the ticket? Is it all necessary?
Watch the time. How long does it take to do each step, like composing an answer or escalating a ticket to engineering?
Is he using tools like notepad or word to copy paste snippets of responses? If you have a library of templates setup in your helpdesk, is he using them? If not, try to understand why. Is it because they’re too many and hard to choose from? Or he doesn’t trust the content?
Understanding why is crucial in narrowing down a problem to the root cause and finding the right solution. Dig deep and ask as many whys as you can. Be curious about it, don’t judge or try to correct what he’s doing. Don’t attempt to coach or suggest solutions.
Look for signs of frustration and encourage the agent to express dissatisfaction. Usually this is where the best improvement opportunities are.
Define the problem(s)
After interviewing a number of agents, bring all the notes together and read them again. Write down in a spreadsheet each problem or difficulty you’ve discovered and what impact you think it has on agent’s productivity. Frame the problem in a way that is agent-centered. For example if you find the agent needs to put together lots of data from different tools, you can state the problem as: “our agents need a consolidated view for all customer-related data”, instead of “we need to consolidate all customer-related data into one view”.
Once you have the list, get a few agents together and run the problems by them. Ask them to rate each problem by how much impact they think it has on their time and productivity. Be open and ask for feedback. Remember that the more involved they feel they are, the more open they’ll be in adopting any future changes to their workflow.
Ideate many solutions for the same problem
Now that you have the problems well understood and defined, it’s time to start working out solutions. Start with the problem that has the biggest impact on agent’s productivity.
In design thinking, this phase is not about using the first idea that comes to mind to solve a problem. Instead try to generate as many ideas as possible as solutions for the problem. Don’t evaluate or judge them yet. Get your agents involved again. This helps you get buy-in later on but also ensures that your ideas will be varied enough, coming from people who think differently than yourself.
After you’ve generated all these ideas for solving a problem, now it’s time to pick which one is the most appropriate and impactful. Again, your agents should have a voice in this decision. Try to look for quick wins as well, which solution has the biggest impact and it’s relatively easy to implement.
Prototype a solution
Now you’re ready to turn an idea into a tangible solution. Rather than trying to implement it fully, end to end, start small by choosing only an essential part of the solution. This is especially the case if the solution is complex and/or costly to implement. It will help you not only learn quickly before you commit a large budget but also generate small wins early. These wins will have an important role in convincing any stakeholders about the importance of your project.
As an example, assuming you’ve found that agents need a consolidated view for all customer-related information coming from multiple internal and external sources. Rather than doing one big project, try to break it down by source and add the information to the agent view one by one. Build up the business case for each step using wins from previous steps.
Test the solution
In this stage any prototype will be deployed and tested with the agents. Sit down again (in the office or remote) with them and observe if your solution has really improved their workflow. Ask as many questions as you need. If for any reason the solution is not working as expected, find out why. Use the methodology explained in phase one for this.
Most solutions are not done with the first prototype. Once you have the results of the test, you might need to go back to previous stages and refine the prototype or re-think the solution completely. This is easy to do because you’ve implemented it in small steps and you can afford change if needed.
As you build the solution and deploy it, make sure you get continuous feedback from the agents. Avoid taking any actions or making changes based solely on your own assumptions.
Problems and friction in agent workflow can cause productivity issues and long resolution times with impact on customer experience. While you can use a customer conversations analytics tool to understand how your agents are doing overall, mapping that with a qualitative approach can bring many benefits.
The design thinking methodology offers an agent-centric approach for improving customer service workflows. It starts by deeply understanding the problems the agents are facing before committing to any costly implementation. Also it involves the agents at every step helping them feel they’re part of the solution. By doing so you are minimizing the risk of working on something that doesn’t solve real problems or face rejection later on.