Empowering contact center excellence for 30 years!

ASA and AWT Definitions and Modifications

Metrics/Performance Measurement

May 16, 2005

We currently offer ASA (Average Speed of Answer) and AWT (Average Wait Time) KPI's. The timer starts on ASA after the caller completes the IVR treatment, greeting, and system status message and stops when the call is answered. AWT starts when the the call hits our call center application and then ends when the call is answered. ASA will always be less than AWT since ASA is only the hold time and AWT includes hold time, IVR and other IVR treatments. Are these accurate definitions of these two KPI's? We are also considering changing the ASA to AHT (Average Hold Time) since the caller is actually on hold during that measurement. Any thoughts? -- Scott Kirkpatrick

Scott Kirkpatrick


  • KG Posted at 12:00AM on Jun 15, 2005

    I am confused as to how you can change the definition of your KPIs. Most KPI's are defined by the switch and the reporting tools you are using. We use Avaya's CMS for switch reporting and IEX for workforce management. In our call center, we refer to AWT as average work time (after call work - or wrap up). If you use AWT as average wait time (IVR + CSR) you are correct in saying that ASA will always be less than AWT. We also refer to AHT as average handle time, which is pretty much standard in the call center industry. If you ask "what is your AHT" a person will assume you mean talk time + work time, not average speed of answer. -- KG, Cleco

  • Cheryl Helm Posted at 12:00AM on Sep 28, 2005

    Scott, Yes, they are accurate definitions since you have clearly defined them. Now I would just clarify them even further with the actual calculations and the reason why this KPI is important. What does it demonstrate to you and your organization or customer? What are you expecting to do with this information? How do they link back to your overall goals for your department and company? I would keep ASA (not AHT) as the abbreviation; it is a common term used by many contact center organizations, and it is best to measure it from the time the caller is actually queuing for an available agent until the call is actually answered. I would stay away from using AHT, since in the industy that abbreviation is often accepted as Average Handle Time. Also, in some sites we use Average Hold Time to measure the average time an representative puts a caller they are talking to on hold, in order to do some record searching, etc. For the AWT definition you gave, I believe you are trying to show the average time a caller is in your system from the start of the call into your system, choose the appropriate options - if they have any choices - and then queue and get answered. I would tend to keep the two metrics separate. One metric, ASA is "controllable" based on your forecasting accuracy, staffing, adherence to schedule and such. This is the time the customer experiences while waiting for an available agent. The other metric or part of the customer experience and is based on the design of your IVR system, its ease of use, clarity of directions and instructions and often the legal requirements of messaging, such as "your call may be monitored." It may also demonstrate how quickly your clients just zero out to reach someone, if you have that option available. Again, other variables affect this time for the caller. In both cases, we could even take it one step further and add those clients who traverse through the IVR and / or go on to queue and then abandon... to provide you with a more complete average experience in the system and average time while queuing; regardless of being answered or abandoned. What I would like you to consider, no matter which abbreviation you choose or how you decide to calculate the information, decide what actions can I take based on the information it provides, what does it mean if this metric goes up or down? What other metrics have a relationship to this and how would they be affected? How does this metric demonstrate we are doing a good job for the department, company or client? That is truly when the data we receive from all our contact center systems becomes valuable to our organizations. This is when the business analytics really demonstrate the value of what we in the contact center can provide to our organizations. -- Cheryl Helm, Helm Communications, Inc

  • ANN-MARIE CASEY-CHRISTENSEN Posted at 12:00AM on Apr 28, 2006

    As a global operations executive for the past 16 years responsible for operations in 21 countries, it concerns me to see such an approach as "WE are considering changing...." We all struggle today with the existing differences in the vernacular and variations and differences in both terminology and the fact that even with standard terminology, like ASA, that the calculation of these items varies based on technology and reporting sytems. As we all make the datamart companies wealthy over the next few years attempting to aggregate and normalize our siloed systems data into single databases so that we can create a holistic view of our centers and performance, across technology, we do not need to be re-defining the few well-known KPI's, rather focus on standardizing new KPI's to address new processes and process changes. -- ANN-MARIE CASEY-CHRISTENSEN, BUSINESS RELATIVITY

  • Azam Malik Posted at 12:00AM on Feb 9, 2007

    ASA and Average Hold Time are two different things. ASA starts when the call is in the agent's queue, and ends when the call is picked up by an agent. But, again, we can't link ASA with Average Hold Time because Average Hold Time is the average time representatives have put the caller on hold during all calls for any reason. -- Azam Malik

  • Shantanu Mohanty Posted at 12:00AM on Jul 27, 2007

    I would keep all these three terminologies separate from each other to avoid confusions. AWT would be the time the caller hits our system and if it is a charged call, the meter for the call charge starts from here. ASA would be the time the caller listens to the hold music or any other recorded messages after the initial greetings and IVR treatment is done before the call is answered by the representative. AHT would be the average handle time which includes the average call time, average hold time and average wrap time or call work. -- Shantanu Mohanty, IBM Daksh

  • Sai Narender Posted at 12:00AM on Oct 19, 2007

    Hi, I echo with what Ann-Marie had to say about standardization of the KPIs. -- Sai Narender, Deloitte

  • Vishesh Jain Posted at 12:00AM on Dec 14, 2007

    There seems to be some confusion with regards to the confusion over AWT. In this case, like mentioned, it refers to Average Wait Time at the switch level and should not be mixed with hold time or wrap time. Probably an ICMI expert signoff on the term would help. -- Vishesh Jain, IGS

Please sign in to contribute an answer. If you don't have an account you can register for free here.

Question Search

Need help with something specific? Search our entire QueueTips section to find it.

Can't find what you're looking for in a current QueueTips post? Submit a new post to us!