Discussions about how technology such as artificial intelligence and machine learning will affect customer experience and customer service sometimes leave out a crucial component: the agent. A positive experience with an agent can result in brand loyalty and, when the experience is really positive, the coveted customer referral. Driving toward positive customer experience outcomes can be a platform for alignment with business leaders.

Alignment starts with clearly articulated performance expectations that fit within and enable the organization’s business goals. Performance expectations may cover areas such as customer experience standards, call handling techniques, average handling time or one call resolution. It is likely that the evolution of AI and predictive analytics will continue to alter the call center landscape and the role of the agent.

Fundamentally, agent training should:

    • Focus on improving agent performance.
    • Engage learners by being relevant to their performance improvement.
    • Improve results by giving learners opportunities to practice in a way that looks and feels like real-world call situations.
    • Connect knowledge and skill content to role-specific situations, cases and scenarios.

What are some strategies for identifying and reproducing real-world scenarios? What factors do you need to take into account when selecting the most appropriate scenarios? Is there a way to benchmark performance? Each of these questions is an essential consideration. These tips should help guide the thought process:

Identifying Scenarios

Here are some tips for identifying real-world situations:

    • Use data that is already available, such as call volume (by transaction, by customer segment, etc.) and error reporting, to identify the most common and impactful situations. Then, work with agents to understand how they handle those calls.
    • Listen to calls and interview agents with a range of experience to gain perspective on what’s easy or difficult, what’s complex or not complex, and what trips them up.
    • Talk to agents, team leaders or supervisors to capture the context of the interaction. What’s the customer’s mood and behavior? What are the technical aspects of the interaction, and how does the system enable or inhibit the call? How can a call change based on what the customer says?

Another way you can improve agent training is to look at the experience through the customer’s eyes. Spend some time with the marketing team to learn about why customers are calling, what are they hoping to achieve and what they compare their experience to.


Here are some tips and ideas for the design process:

    • Use the organization’s call handling expectations to assess agent progress and provide coaching that mirrors what will happen on the floor. Consider using (or mirroring) the quality monitoring process and criteria that their supervisors will use when assessing agents’ job performance. Observe calls to develop an understanding of the pace of a call, and set expectations near or at a production pace.
    • Consider creating a base call flow that reflects a typical call from end to end. Then, determine how you can adjust the base flow to address different scenarios, the context clues agents use during calls and other considerations for improving customer service.
    • Use a simple matrix to organize scenarios by levels of technical and interpersonal difficulty and complexity. Align the scenario complexity with developmental stages of performance milestones.
Technical Complexity High
Low Medium High
Interpersonal Complexity


    • Determine what information is available and accessible through the system, workflow and job aids, and determine how difficult it is to access that information while working.
    • Keep in mind that call center work is repetitive. It is not necessary to teach and practice all of the scenarios; the task itself will provide practice.
    • Try to select an appropriate number of practice scenarios based on the complexity of the call center agent role.
    • In the call center, some situations are unique and, therefore, may not require formal training. Train agents on how to identify and pass those calls to more experienced team members.

There are a few issues to watch out for when designing practice for these scenarios, or the learners may not be successful. They include practicing:

    • In slow motion, which does not reflect the actual speed of the job (for example, if the average call handling time were 300 seconds but the practice took 600 seconds).
    • Only the optimal path or call flow, which leaves out complex or challenging situations.
    • Interacting with customers and using systems separately.

Learning Ecosystem Considerations

It’s also important to consider the organization’s learning ecosystem and what it will — and won’t — support. You can deliver engaging, relevant and effective training in a variety of ways, including eLearning, on-the-job training (OJT), instructor-led-training (ILT), virtual-instructor-led-training (VILT), video and coaching. If your ecosystem doesn’t include a training sandbox for the system environment, doesn’t have performance support capabilities or doesn’t have adequate coaching resources, some modalities will be off the table, while others should be front and center.

Finding ways to provide great training often means creatively working with the strengths and weaknesses of the ecosystem. For example, when there isn’t a place for safe practice in the production system, because there’s no sandbox, it may be necessary to use eLearning with practice to simulate scenarios or guided real work that puts a coach shoulder to shoulder with a new agent to step in when necessary and provide real-time feedback.

Research indicates that as AI and other technologies improve and become more prevalent, businesses will be more efficient, and customers will be more satisfied. The goal for training professionals will be to see the positive influence of these evolutions and to marry them with a learning strategy that creates measurable and positive change that drives the business.

When measuring a customer service call center, leaders often look at statistics like key performance indicators (KPIs), time to proficiency, turnover and employee satisfaction. In that light, you can view call center learning strategy as a critical competitive differentiator. There are few levers you move that can have such a profound effect on the customer experience. Improve agent engagement by giving them real-world practice that mirrors what they will have to do when they hit the floor. Practice makes better, if not perfect. By providing your customer service call center agents with engaging and relevant content, “better” is inevitable.