Glossary
Average Handle Time (AHT) is a customer support metric that measures the average total duration of a customer interaction from start to resolution. It includes active conversation time, hold or wait time, and any post-interaction wrap-up work (notes, ticket updates, follow-up actions).
AHT is calculated as:
AHT = (Total Talk Time + Total Hold Time + Total After-Call Work) / Number of Interactions
For different channels: - **Phone**: Talk time + hold time + after-call notes (benchmark: 6-8 minutes) - **Live chat**: Active conversation time + typing delays + wrap-up (benchmark: 8-12 minutes) - **AI chatbot**: Time from first message to resolution (benchmark: 30 seconds - 2 minutes) - **Email**: Time spent reading, researching, and composing a response (benchmark: 5-10 minutes per response)
AHT is a double-edged metric. Lower AHT means more efficiency, but pushing AHT too low can sacrifice resolution quality, agents may rush through conversations and fail to fully resolve issues, leading to repeat contacts.
In practice, average handle time (aht) should be evaluated by what it changes in the support workflow. Ask whether it improves answer accuracy, reduces repeated agent work, clarifies handoff decisions, or makes reporting easier. If the answer is only "it sounds modern," the concept is not yet operational.
A concrete example is ai chatbot vs human aht comparison: A support team measures AHT across channels: AI chatbot resolves "What are your hours?" in 12 seconds, while the same question takes a human agent 3 minutes (greeting, lookup, response, closure). For FAQ-type questions, AI AHT is 90% lower than human AHT.
The simplest takeaway is: AHT measures the total time to handle a customer interaction: conversation + hold + wrap-up
Benchmarks vary by channel and industry. Phone: 6-8 minutes. Live chat: 8-12 minutes. AI chatbot: 30 seconds to 2 minutes. Email: 5-10 minutes per response. Focus on reducing AHT through efficiency (AI, better tools, templates) rather than rushing agents.
AI reduces AHT by: resolving simple issues instantly (eliminating human handling entirely), pre-gathering customer context before handoff (saving 2-3 minutes of agent information gathering), suggesting responses to agents (reducing typing time), and auto-generating wrap-up notes (cutting after-call work by 80%).
No. Extremely low AHT can indicate agents are rushing through conversations, not fully resolving issues, or skipping documentation. The goal is efficient resolution, not fast resolution. Track AHT alongside first-contact resolution rate and CSAT to ensure quality is maintained.
A team discovers their AHT is high because agents spend 3 minutes per ticket writing wrap-up notes. They implement AI-generated conversation summaries that automatically fill in ticket details, reducing after-call work from 3 minutes to 30 seconds without sacrificing documentation quality.
AHT is a core input for workforce management. If you handle 5,000 conversations per month at 10-minute AHT, you need approximately 833 agent-hours. Reducing AHT to 7 minutes drops this to 583 hours: a 30% reduction in staffing needs. AI chatbots handling 60% of volume further reduces human agent hour requirements.