Average handle time (AHT)

Average handle time (AHT) is the mean duration of a customer support interaction from start to finish. For phone support, it begins when the agent answers the call and ends when after-call work is complete. For chat and email, it covers the active time the agent spends working on the conversation, including research, writing, and documentation.

AHT matters primarily because it drives staffing and cost. If a team handles 1,000 calls per day with an AHT of 6 minutes, they need a fundamentally different number of agents than if the AHT is 10 minutes. Every minute added to AHT across thousands of interactions translates to additional headcount or longer queue times.

However, AHT is one of the most frequently misused metrics in support operations. Optimising AHT in isolation, pushing agents to end interactions faster, almost always degrades resolution quality. Rushed interactions lead to unresolved issues, repeat contacts, lower customer satisfaction score, and higher overall cost. The goal is not the shortest possible handle time but the most efficient handle time that still achieves first-contact resolution.

The most valuable use of AHT is as a diagnostic tool. When AHT increases, it signals that something has changed: a product update introduced new complexity, a knowledge gap has appeared, or system performance has degraded. When AHT varies significantly between agents handling the same issue types, it reveals training opportunities. The metric is at its best when it prompts investigation rather than pressure.

AHT is calculated by summing the three components of an interaction, talk time, hold time, and after-call work, then dividing by the number of interactions. Most contact centre and helpdesk platforms calculate AHT automatically, but understanding the components is essential for interpretation.

Talk time (or active chat time) is the core of the interaction where diagnosis and resolution happen. Hold time reflects research, consultation, or system lookup during the interaction. After-call work covers documentation, ticket updates, and any follow-up actions the agent performs immediately after the interaction ends.

AHT is the sum of its components, and each component has distinct drivers. A metric tree reveals where time is spent and which improvements will have the largest impact on overall handle time.

The tree shows that AHT is not a single number to push down but a composite of different time investments. If hold time is high, the fix is system performance and tool integration. If after-call work is high, the fix is automation and templates. If talk time is high because of complex issues, the fix may be product improvement rather than agent coaching.

The issue complexity mix branch is particularly important. A shift toward more complex tickets, perhaps after a product release or a change in self-service coverage, naturally increases AHT even if agent efficiency is unchanged. The tree helps distinguish between efficiency problems and complexity changes.

1. 1

   Provide agents with unified customer context

   Agents who must navigate between three or four systems to find account information, order history, and previous interactions waste minutes on every call. Integrate customer data into a single pane so agents have full context the moment an interaction begins.

2. 2

   Automate after-call work

   After-call work is often the easiest component to reduce. Auto-populate ticket fields from the conversation, use AI-generated interaction summaries, and provide templated notes for common issue types. Reducing after-call work by 60 seconds per interaction saves significant hours across the team each week.

3. 3

   Build guided resolution workflows

   For common issue types, create step-by-step resolution guides that agents can follow in real time. These reduce diagnosis time, ensure consistent resolution quality, and are especially effective for newer agents who would otherwise take longer to resolve familiar issues.

4. 4

   Improve self-service to deflect simple issues

   When simple issues are resolved through self-service via the knowledge base, the remaining tickets handled by agents are more complex and naturally carry a higher AHT. This is a good outcome: the overall cost per resolution drops even if agent AHT increases, because the cheapest resolutions are now automated.

5. 5

   Analyse AHT by issue type rather than by agent

   Comparing AHT across agents handling different issue mixes is misleading. Instead, compare AHT within the same issue category. This reveals genuine efficiency differences and training opportunities without penalising agents who handle the most complex tickets.

KPI Tree lets you model AHT alongside the quality metrics that must move in tandem: first contact resolution, customer satisfaction, and agent touches per ticket. This prevents the common failure mode of optimising AHT in isolation.

Decompose AHT by its three components, then further by channel, issue type, and team. When AHT shifts, the tree immediately shows which component changed and in which context. Connect AHT to cost per ticket and staffing models so that efficiency improvements are visible in financial terms. Each node can be owned by the team best positioned to influence it, ensuring that AHT optimisation is a shared responsibility rather than a pressure metric imposed on frontline agents.