Several factors play a role in ensuring that the email resolution rate is higher in a customer support scenario. Reading the email, understanding the issue, accurately categorizing the email, customer support agent capacity and then allocating it to the appropriate function.

We know how agonizingly painstaking it is for the customer support agents to manually read, categorize and then respond to emails. The push from the organization to improve email resolution rate, agent productivity and reduce costs is adding to the pressure on support agents. Organizations are trying various models to improve productivity.

We have done a detailed survey to bring more clarity into the issue of agent productivity especially when they deal with the most important aspect of reading and categorizing customer emails.

Email Resolution Rate Survey Methodology

At this point, I should thank my client, a leading consumer finance organization, who allowed us to conduct this survey in a real-world scenario.

We invited 10 customer support agents whose primary job is to:

• Read the customer support email
• Understand the issue
• Categorize the email in set sections
• Allocate it either to an appropriate customer support agent or upload in the CRM

The company segregates emails into 60+ categories. Each of these agents was given 100 emails each. All of them knew and were trained in the categories. Remember, this is a live project. Now, we have divided them under the following 3 scenarios:

Scenario 1 Task (S1)

Categorize 100 mails with a mix of all 60+ categories.

Scenario 2 Task (S2)

Categorize 100 mails with a mix of just 4 categories. However, these agents were asked to categorize 100 emails immediately after categorizing the previous 100 mails. So, there was no break between the first 100 to the next.

Scenario 3 Task (S3)

Categorize 100 mails with a mix of just 4 categories. These agents were given a short break after categorizing the first 100 emails.

Results

Yes, we know that the results will be varied. Let us see how varied they are. Firstly, let us look at the efficiency rate of the agents in reading the emails.

Insights

What does this tell us? Is one scenario better than the other? I don’t think so. Here is why.

Insight # 1:

Agent productivity drops as the email categories increase

We all want more from our agents. However, if we keep adding categories to their list the efficiency significantly drops.

Insight # 2:

Fatigue causes fall in the agent’s efficiency even with fewer categories

Remember, we thought that scenario 2 would perform way better than scenario 1? It will not, because the fatigue factor not only reduces efficiency but also hurts the accuracy. Even if we drastically reduce the number of categories, making an agent continuously work will be counterproductive.

Insight # 3:

Scenario 3 is good, and the only best option

Scenario 3 did really well and the only best option under the circumstances. This, however, means having an army of customer support agents to read and categorize emails.

Insight # 4:

Manual email categorization is error-prone resulting in drop in customer satisfaction

Truth be told – Manual categorization means more time and less accuracy. You are not only making the customer wait but perhaps going back to them with a wrong solution.

Final Thoughts

Traditionally, we have been reading, categorizing and allocating emails manually. This doesn’t mean that we need to continue on the same path. Customer support agent time can be very effectively used elsewhere than in a low production environment. Do not compromise on low email resolution rate, customer satisfaction and customer experience.

There are smarter solutions that will help you automate the entire process. Not just that, it can help you get insights that you never thought would be possible.

Thank you for reading these survey results. A very special thanks to everyone who took the survey and helped in the process. We tried to objectively view the data to present you with our insights, but I’d love to hear your thoughts and experiences.