Centre de contact

Average Talk Time (ATT): how to reduce it

Average Talk Time (ATT): definition, causes of long conversations, and ways to reduce it without compromising customer relationship quality.

Thomas Guenoux
Thomas Guenoux
27/03/2026
·
9
min read
Average Talk Time (ATT): how to reduce it

The average talk time (ATT) is a key call center KPI that helps answer a simple question: are agents spending too much... or not enough time with customers?

When it is too high, ATT reveals operational inefficiencies and degrades contact center performance: lower answer rates, longer queues, and increased average cost per call... At scale, a few extra seconds per interaction can represent significant costs for your center.

Conversely,an ATT that is too short can indicate superficial handling of requests. Incomplete or misunderstood answers lead to callbacks, frustration, and a decline in customer experience.

Because ultimately, the goal is not the time spent, but the quality of the response provided. The numbers don't lie: 89% of consumers are more likely to make another purchase from a brand that provides excellent customer service.

So, how can you effectively measure your ATT and find the right balance between operational performance and customer satisfaction? Our experts explain.

What is ATT?

ATT represents the average time an agent spends handling incoming calls. It includes the entire time elapsed between answering the call and its resolution, but excludes initial wait time as well as post-call processing.

The formula for calculating ATT is simple:
ATT = Total duration of conversations for the period / Total number of calls for the period

To better understand this formula, let's consider this example: if for a call center period, agents spend 300 minutes in conversation for 100 calls, the ATT is then 3 minutes.

Why is ATT a strategic indicator for customer relations?

ATT is an indicator used to track call center productivity. More specifically, this KPI provides insights into your agents' work and the quality of telephone interactions.

More specifically, by tracking ATT, you will be able to take action on the following points:

  • Answer rate and wait time : an optimized ATT frees up time for other calls, which helps improve the answer rate and reduce wait times.

  • Strengthening quality : Identifying abnormal ATT values helps pinpoint efficiency issues in request handling and allows for appropriate action.

  • Effective training : identifying gaps guides training efforts.
  • Staff planning : a stable ATT helps anticipate staffing needs.

  • Improving financial efficiency : All of the previous points affect the profitability of the call center and customer service operations.

On average, An ATT typically ranges between 3 and 8 minutes, depending on the nature of the call. Simple requests, such as order tracking, are closer to 3 minutes, while more complex cases requiring authentication, routing, payment, troubleshooting, or multiple interactions tend toward the higher end. In sectors like hospitality, we see ATT values closer to 3 minutes, whereas in banking or insurance, it can be closer to 6–8 minutes.

Therefore, to define acceptable conversation duration thresholds, it is recommended to rely on your scripts and processing workflows to set realistic ATT goals tailored to your specific use cases.

What causes a high ATT?

An ATT above the defined acceptable thresholds can be explained by:

  1. Overly complex processes: Long customer journeys, cumbersome internal procedures, or lengthy validation processes that naturally extend call duration.
  2. Lack of tools, integrations, or automation: agents who have to navigate between multiple tools, search for information, or re-enter data lose time with every interaction.
  3. Lack of training or support: some junior or insufficiently trained agents may take longer to understand the request and respond to it effectively.
  4. Inefficient routing: call transfers significantly increase ATT, especially when the customer has to repeat their problem.

Is a short ATT always a good thing?

A short ATT is obviously a good sign. However, when it is significantly lower than the target, it may reveal issues with response quality : requests being handled too quickly, with partial or unclear information, at the expense of customer relations.

To determine whether a low ATT should be celebrated or re-evaluated, it is necessary to cross-reference it with other customer service quality indicators, a point we will address later.

How can you analyze ATT effectively?

How can you analyze ATT effectively?

To get the most out of ATT, it should be analyzed with precision and context. A raw reading is not enough. In other words, cross-referencing this metric with other data and segmenting the results allows you to understand the underlying dynamics and take targeted action.

Cross-referencing ATT with other KPIs

To evaluate the quality of interactions, compare ATT with customer satisfaction (CSAT)and the Net Promoter Score (NPS). For example:

  • Short ATT and high CSAT indicate very effective call management.

  • Long ATT with low CSAT can reveal difficulties in handling arguments or unclear processes.

  • High ATT combined with low FCR indicates poorly resolved calls, often prolonged by multiple transfers.

In a center where unnecessary transfers are frequent, we often see an increase in ATT without any real benefit to the customer.

Segmenting for better insight

If cross-referencing is not enough to understand certain results, you can change your analytical angle. In other words, segment ATT by:

  • Call type : technical requests are longer than simple questions.

  • Agent : identify gaps to target training.

  • Schedule : call spikes can increase talk time.

  • Channel : voice calls differ from chats or videos.

Such segmentation allows for make concrete decisions and contextualized, rather than basing actions on a general average that lacks representative value.

Using analytics tools

The conversational analysis solutions (speech analytics) also provide significant support. They analyze the content of conversations to reveal the causes of high ATT (recurring issues, pain points).

Combined with dashboards, they offer a real-time view to allow for quick strategy adjustments. By integrating these practices, it becomes possible to transform ATT into a strategic tool for optimizing performance and customer satisfaction.

Factors influencing average talk time

Factors influencing average talk time

To reduce ATT without compromising quality, you must first gain a deep understanding of what influences it. These factors can be related to the customer, the agent, or the technical environment.

Complex requests and agent training

The more technical or emotional a call is, the longer it will last. Obviously, a complaint regarding a financial dispute will naturally take longer than a simple request for an account balance.

If an agent is poorly trained or lacks autonomy, it increases processing time. The solution involves targeted training on frequent cases, role-playing exercises, and adapted standard responses.

Scripts and business tools

Poorly designed, overly long, or inflexible scripts slow down the agent. Likewise, a omnichannel customer service poorly integrated system, a hard-to-access database, or inefficient routing can add several dozen seconds to every call.

Multiple transfers or poorly managed redirects also contribute to unnecessarily long conversations. Improving tool ergonomics and procedural clarity is therefore essential to limit ATT without impairing the customer experience.

How to optimize ATT without harming quality?

To make ATT relevant, the goal is not always to reduce it, but to align it with an optimal customer experience.

Training agents with speech analytics

The speech analytics allows you to listen to and analyze conversations to identify weak points. For example, Batvoice AI detects moments where agents hesitate or repeat unnecessary information. Based on these analyses, it becomes easy to offer targeted training. Naturally, an agent who masters common objections will easily reduce ATT without sacrificing quality.

Optimizing conversation scripts

Clear and concise scripts guide agents without locking them into rigid responses. In this regard, it is advisable to update scripts based on conversation analysis to eliminate unnecessary steps. For example, it is possible to save more than 15 seconds per call by using a simplified script for refund requests.

Integrating conversational AI

Artificial intelligence-based solutions like callbots can handle simple requests such as order tracking or routing calls to the right department. According to Gartner, integrating conversational AI can reduce request processing time by 30%. This filtering significantly lightens the load on human agents, who can then focus on high-stakes requests.

➡️ Estimate the return on investment for your project in just a few clicks with our custom simulator.

Estimate the gains of a voice agent based on your own call volumes with our ROI simulator.

Ensuring tool fluidity

A high-performance CRM and a well-organized knowledge base reduce the time spent searching for information. Conducting regular tests helps identify slowdowns or duplicates. A simple gain of 10 seconds per call, multiplied by hundreds of daily interactions, represents a significant boost in efficiency and productivity.

Measuring the impact of actions

For every action implemented, it is recommended to track results using dashboards like those offered by Batvoice. This provides the ability to observe changes in ATT, FCR, and CSAT before and after initiatives. This approach allows for continuous decision-making adjustments based on reliable, concrete data.

ATT is not enough to measure your customer satisfaction

ATT is not enough to measure your customer satisfaction

While valuable, ATT does not provide a complete picture of the quality of service provided. For a more nuanced view, it is recommended to combine it with other key performance indicators:

  • CSAT to measure post-call satisfaction in real-time.

  • NPS to estimate customer loyalty.

  • FCR to evaluate whether requests are resolved during the first contact.

Furthermore, you should incorporate the Customer Effort Score (CES). This indicator measures the effort perceived by the customer to get an answer or resolve an issue. Cross-referencing ATT with CES helps identify friction points in the customer journey and adjust processes accordingly.

How many calls can your voice agent absorb?

Estimate in one minute the number of automated calls and the savings on your own volumes, with our ROI calculator.

Read also

Our latest articles

All articles