Glossary
AHT (average handle time)
Average handle time (AHT) is the average time it takes to handle one contact: the conversation itself, hold time, and after-call work. AHT drives how many agents you need, because the workload on your lines is the number of contacts multiplied by AHT.
Why it mattersAHT drives how many agents you need and what each contact costs.
Also called: Average handle time, Average handling time
AHT = (talk time + hold time + after-call work) / number of answered contactsAll values for the same period and queue. Abandoned calls are not counted: they never reached an agent.Example
Formula and example
In one day, agents answered 120 calls. Talk took 480 minutes, hold 60 minutes, and after-call work (notes, CRM tickets) 120 minutes. That is 660 minutes in total, so AHT = 660 / 120 = 5.5 minutes, or 330 seconds.
Only answered calls go in the denominator. Add abandoned calls and AHT comes out lower than it really is, so a staffing plan based on it will be short of agents.
Details
What AHT is made of
- How complex the questions are and whether the call reaches the right agent the first time
- How quickly the agent finds the answer in the knowledge base and CRM
- How many applications they have to open during the call
- How much manual work follows the call: notes, forms, tickets
- Transfers and holds while the agent consults a colleague
Metrics
Typical values
AHT depends heavily on the topic: an order status check takes 2-3 minutes, technical support 8-12 minutes or more. So compare one queue's AHT across periods, not different queues against each other.
How AHT affects staffing: 100 calls in half an hour at 180 seconds each is a load of 10 erlangs, and a target of 80% of calls in 20 seconds needs 14 agents on the line. If AHT grows by 20% to 216 seconds, the load becomes 12 erlangs and you need 16 agents.
Practice
How to cut AHT without hurting quality
Do not cut AHT at any cost. If an agent rushes and leaves the issue unsolved, the customer calls back and the total workload grows. Watch AHT together with first contact resolution and customer ratings.
- Show the customer record on the agent's screen when the call arrives, so they do not search the CRM
- Keep answers in the knowledge base, not in the heads of senior agents
- Turn on AI suggestions: the right answer appears while the customer is still talking
- Find the longest calls and work out why they dragged on
- Reduce manual after-call work: templates, auto-filled fields
In CONTACT.WORK
How CONTACT.WORK uses it
During the call, the AI assistant pulls answers from the knowledge base for the agent, and it writes the call summary itself, which cuts after-call work. Reports split AHT into talk, hold, and after-call work, so you can see right away what made it grow. Speech analytics transcripts show at which stage conversations drag on. Shift planning uses the actual AHT from the platform, so staffing for each interval is based on real data.
More
Where to use it
Backlinks
Where it appears on the site 10
Calculators 2
Blog 8
- Contact center workforce management: forecast, schedule, adherence
- Custom development of your contact center platform for your processes
- How many agents does your contact center need? Erlang C in plain English
- How to build a contact center from scratch
- How to reduce average handle time without sacrificing quality
- Remote agents: a contact center in the browser
Glossary