Skip to main content


How to Measure and Improve Handle Time

This is the first of a series discussing the following common call center metrics specifically as they apply to cases originating via email or web form channels.

  1. Handle Time
  2. Escalations
  3. Volume
  4. First Contact Resolution Rate
  5. Customer Satisfaction


It is common to hear statements such as “Our average handle time is X.” That statement is meaningless unless it is prefaced with information about how handle time is measured. This blog lists some of the diverse methods for measuring handle time. Metrics that should be cross-reported with handle time are described and suggestions for improving handle time are offered.

The Different Kinds of Handle Time

Consumer Handle Time
Customer Handle Time is measured from the time the customer initiates a case until the time that customer agrees that the case has been addressed in terms of an answer. It is time submitted to closed.

Agent Time Assigned
Agent Time Assigned is the amount of time a particular agent takes to handle a case, even if he does not close it. It is amount of time between when that case was assigned to the agent and when the agent either closed the case or reassigned it to another agent. It does not include the time that the case is unassigned or the time while the case was assigned to a different agent.

Agent Time Worked
Agent Time Worked is the amount of time it takes while the case is being read, researched, having a response edited or, if included, having wrap-up information added. It is the amount of time the agent takes to ACTIVELY work the case.

Case Time Assigned
Case Time Assigned is the total amount of time the case was assigned to agents to be worked. If multiple agents worked the case, it is the sum of their Agent Time Assigned for that case.

Oracle - Guide to Oracle Cloud: 5 Steps to Ensure a Successful Move to the Cloud
Guide to Oracle Cloud: 5 Steps to Ensure a Successful Move to the Cloud

Explore key considerations, integrating the cloud with legacy applications and challenges of current cloud implementations.

Get the Guide

Case Time Worked
Case Time Worked is the total amount of time the case was being actively worked by Agents. If multiple agents worked the case, it is the sum of their Agent Times Worked for that Case

Total Handle Time
Total handle time is the total amount of time from when the case was submitted until it was completely wrapped up and closed. This includes the time the case was waiting to be assigned, the total amount of time each agent worked on it, and usually any wrap-up time.

Some Other Measurement Considerations for Average Handle Time

Business Hours vs. Real Time Hours
All of the different ways to measure Handle Time include separating business hours from real time hours. For example, let us say that a call center is open 9-5 Monday-Friday but closed the rest of the time. If a customer submits a case at 4 PM on Friday and the case is closed the following Monday at 10 AM, then the Total Handle Time in Business hours for that case is 2 hours, The Total Handle Time in real hours is 66 hours.

Agent Responsibility vs. Customer Responsibility
Another parameter in the measurement of Handle Time is whose court the ball was in during the lifetime of the case. Let a customer submit a case during business hours at 10 AM. Let an agent immediately spend 20 minutes researching the case, realizing that some essential information needs to be collected from the customer and another 10 minutes responding to the customer asking for the additional information. That is, at 10:30 the agent sent the email to the customer and then the case was waiting for the customer to respond. The customer sees the email at 12:15 that same day. By 12:30 the customer has responded back to the agent. The agent starts immediately opens the customer response and spends another 30 minutes responding to the customer, closing and wrapping the case. By 1 PM the case is closed. The Customer Handle Time is 10:00AM-1:00 PM or 3 hours.
The Agent Time Worked is 10:00AM-10:30AM and 12:30PM-1:00PM or one hour. (In this case the Agent Time Worked is that same as the Agent Responsibility Time. These are the same because we had the agent work the incident immediately for the sake of simplicity.) The Customer Responsibility Time is 10:30-12:30 or 2 hours.

How Are Re-opens Handled?
Let us say the case comes in at 1 and is closed at 2. Then the customer re-opens the case at 3 and then it is finally closed at 4 PM. Should that count as two one-hour Agent Time Assigned for 1-2 and 3-4 or as one three-hour Agent Time Assigned for 1-4? The question becomes whether Average Handle Times will be computed using the number of times inquiries are closed, or using the number of inquiries closed.

If using the number of inquiries closed, that raises a second question about procedures when customers tag a second question onto the first case. Usually best practice is to count a re-open related to the initial case as the same case with additional handle time. In addition, if the re-open is NOT related to initial case to copy the new case to a new case number to be tracked as a separate case, leaving the initial case closed.

Are Cases Routed to Same Initial Agent After a Customer Responds?
Usually the best practice IS to assign the same agent back to the case after update by the customer. However, this varies across organizations. Some organizations re-queue the response for the next available agent. Other organizations create a brand new case for each update. The procedure in place affects Handle Time metrics and should be clearly understood.

Is Wrap-Up Time Included?
Usually the best practice IS to include wrap-up time in the handle time computations AND to have a separate measurement just for wrap-up time alone in order to gauge how long it is taking to work the case vs how long it is taking to document the case.

Is QA Time Included?
Many organizations have cases opened and reviewed by a QA team. Usually the best practice is to NOT include QA with Handle Time. Have a separate measurement for how long the case was opened for review by QA.

Metrics Commonly Correlated to Handle Time

Prioritizing closing tickets quickly over resolving customer issues might reduce individual agent Time Assigned but it may decrease customer satisfaction. Handle Time by itself is not a success metric. It should be cross-reported with customer satisfaction, escalation rates, volume, and first contact resolution rates. Handle Time is also often correlated to particular products or categories in order to determine where to focus additional training and documentation resources. The ultimate goal, of course, is to decrease handle times while increasing customer satisfaction. This can sometimes be accomplished by well-defined escalation policies as well as improved documentation and training. It is also important to correlate Case Time Worked with use by agents of the agent knowledge base.

How to Improve Handle Time

  1. Cross-report Agent Time Worked with Cases on specific products and categories to help define where additional training is needed.
  2. Cross-report Agent Time Worked with Agent re-assignment metrics to help define where specific routing rules would be helpful.
  3. Cross-report Agent Time Worked with Agent Escalations to help define where specific escalation policies will be helpful.
  4. Cross-report Agent Time worked with Knowledge Base Metrics to define where keywords, new answers or encouraging agents to use their knowledge base more effectively will improve handle time.

Thoughts on “How to Measure and Improve Handle Time”

  1. Hi Sheri – great summary. I feel sometimes projects are focused on urgent things more than important. Your article is a valuable resource and input for gather requirement phase, where project team can use it to ensure that what will be delivered will be measured – shared w/ co-workers!!! Thanks for that.

  2. This is one of the most comprehensive analyses I have seen. I agree that Business Handle Time vs. Total Handle Time is an important distinction, and I believe it should be calculated both ways, since we cannot measure agents against time that the business is closed, but we also can’t lose sight of the customer experience.

  3. Great article. I’ve seen so many different ways also of measuring AHT too and it is nice to see it explained in one place. This will help a lot of people.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sheri Hastings

Sheri is an Oracle Certified Implementation Consultant for both the 2012 and 2016 versions of Oracle Service Cloud. She has more than 15 years of experience implementing Oracle Service Cloud- dating back to 2002 when she was hired as a senior consultant for RightNow Technologies. Prior to that she worked for eGain and KPMG which gives her more than 25 years of experience in the field of Customer Service Software. From 2002 until 2011 Sheri implemented more than 250 new RightNow customers and presented at multiple RightNow User's conferences. The areas of the product about which she is most passionate are analytics, knowledge base optimization and business rule best practices. Sheri has a BS in mathematics from and a MFA in Poetry. She is on the board of the DAM Cancer Foundation and volunteers as a math tutor in Los Angeles. In her spare time she enjoys horseback riding, gardening, and being a "foodie".

More from this Author

Follow Us