Skip to content

AHT

What is AHT?

Average Handle Time (AHT) is the average total time it takes to complete a customer interaction, including talk time, hold time, and after-call work (ACW). The metric is formalized in the international contact center quality standard ISO 18295 and in the COPC Customer Experience Standard, and is a core measurement reported in BenchmarkPortal industry surveys.

AHT includes:

  • Talk time (the conversation)
  • Hold time (time the customer is waiting while the agent works)
  • After-call work / wrap-up (notes, dispositions, case updates)

AHT matters because it affects cost-to-serve, staffing models, service levels, and — when managed poorly — customer experience. A common misconception is that lower AHT always equals better service; quality and context matter. SQM Group's First Contact Resolution research and the Harvard Business Review's writing on customer effort both show that aggressive AHT reduction can destroy resolution quality and increase repeat contacts.

Quick definition:

AHT (Average Handle Time) is a contact center efficiency metric that measures the average total time an agent (or voice AI agent) spends handling a customer interaction — typically including talk time + hold time + after-call work (ACW) — from start to finish.

Why AHT matters for customer operations

AHT is a foundational contact center KPI used to understand how much effort it takes — on average — to handle an interaction from the moment an agent begins working it to the moment required follow-up work is completed. Most teams think of AHT as a "call metric," but the same concept applies across channels, including chat and messaging, where "handle time" still includes the active interaction plus post-interaction work, as documented in the Genesys PureCloud (Genesys Cloud CX) glossary and the Salesforce State of Service report.

You'll often see AHT discussed alongside metrics like ASA (Average Speed of Answer), FCR (First Contact Resolution), CSAT, and CES — the Customer Effort Score developed in CEB/Gartner research and popularized in HBR. AHT is primarily an efficiency metric, but it can't be treated as an isolated target without producing unintended consequences (repeat contacts, escalations, lower satisfaction, compliance risk) — a pattern Goodhart's Law and the operational research literature describe as inevitable when a metric becomes a target.

What AHT includes (and what it doesn't)

AHT typically includes talk time, hold time, and after-call work (wrap-up). The Amazon Connect metric definitions and the ICMI handle-time guidance document the standard composition.

What AHT typically includes:

  • Talk time
  • Hold time
  • After-call work (ACW)

What AHT often does not include (or varies by platform/reporting rules):

  • Queue time before the agent answers (usually measured by ASA / wait time)
  • IVR time (sometimes tracked separately)
  • Time that happens after transfer to a different team (depends on whether you treat the transfer as the same interaction or a new one)

A crisp way to keep measurement clean: AHT is generally the time you're talking with a customer, have them on hold, or are in wrap; it does not include time waiting in queue.

AHT formula (with a worked example)

Standard formula

A widely used AHT formula is:

AHT = (Total Talk Time + Total Hold Time + Total After-Call Work Time) ÷ Total Number of Interactions

This formula (talk + hold + after-call work) is commonly used in contact center KPI definitions and matches the definition documented by Genesys, NICE, and BenchmarkPortal.

Example calculation

Assume your team handled 10 voice interactions in a reporting period:

  • Total talk time: 110 minutes
  • Total hold time: 25 minutes
  • Total after-call work: 15 minutes

Total handling time = 110 + 25 + 15 = 150 minutes AHT = 150 ÷ 10 = 15 minutes

So your AHT is 15 minutes per interaction for that period.

Reporting rules that prevent bad decisions

If AHT becomes a focus of leadership, measurement must be consistent, or it will turn into a reporting game. Lock down these rules:

  1. What is an "interaction" in your reporting? One conversation regardless of transfers, or each segment counts separately?
  2. Are you including outbound dial time? Some platforms include dialing/contacting time for outbound interactions in "handle time" calculations.
  3. Are you including "custom statuses" or pauses? Different platforms define "handled time" differently. Amazon Connect, for example, defines handled/handle time in ways that can include ACW and other durations depending on which metric/data definition you're using.

If your org uses multiple tools (CCaaS + CRM + WFM + QA), align definitions across them — or at least document the differences and prevent apples-to-oranges comparisons. The discipline is the same one ISO 18295 and the COPC standard require for any externally reported customer-contact metric.

What is a good AHT?

There's no universal "good AHT." AHT varies by:

  • Interaction type (simple FAQ vs complex dispute)
  • Channel (voice vs chat vs email)
  • Audience (consumer vs enterprise; new vs existing customers)
  • Regulatory/compliance steps (identity checks, disclosures under Regulation F, Regulation Z, or HIPAA requirements)
  • Knowledge maturity (how quickly answers and next steps can be retrieved)

A practical approach is to define AHT targets by workflow category and pair them with guardrails such as FCR and QA/compliance adherence — the approach BenchmarkPortal and the International Customer Management Institute (ICMI) recommend for benchmark normalization.

Benchmarking the right way: compare like with like

Instead of one AHT target for an entire operation, create AHT bands for:

  • Simple, repeatable inquiries (status, hours, basic how-to)
  • Standard servicing workflows (account changes, scheduling)
  • Complex or regulated workflows (identity verification under FFIEC authentication guidance, disputes, recovery arrangements, exception handling)

This prevents you from "improving" AHT by discouraging complex calls or rushing high-risk interactions — a pattern repeatedly flagged in SQM Group and Gartner CX research.

The trap: optimizing AHT at the expense of outcomes

Lower AHT can be a false win if it causes:

In other words: AHT is an efficiency metric; it should move in the right direction without breaking resolution quality. The Genesys definition explicitly calls out that lower AHT isn't automatically better without context.

What drives AHT higher?

AHT rises because of friction. Your fastest path to lowering AHT is rarely "tell agents to be faster." It's removing friction from talk time, hold time, and after-call work.

Common drivers:

Knowledge friction

  • Agents searching multiple systems for basic answers
  • Inconsistent or outdated knowledge articles
  • Unclear policy exceptions
  • Lack of a single "source of truth" — a data quality discipline catalogued in the DAMA DMBOK

Workflow friction

  • Manual identity verification steps that are slow or inconsistent
  • Repeating disclosures because the workflow doesn't guide them
  • Copying details between CCaaS, CRM, billing, and internal tools ("swivel chair" work) — the "screen-shuffling" pattern documented in Forrester productivity research

Transfers and routing problems

  • Unclear ownership boundaries between teams
  • Misrouted calls that require transfers
  • Incomplete intent capture up front

Customer comprehension and emotion

  • Customers repeating themselves because context wasn't captured
  • Confusion about next steps and requirements
  • Time spent de-escalating frustration caused by prior failures

Language and audio issues

  • Mishears and repeated questions
  • Multilingual needs requiring handoffs
  • Poor audio quality leading to repetition, measurable via Word Error Rate in NIST speech recognition evaluations

Systems and latency

  • Slow CRM lookups
  • Authentication friction
  • Timeouts and re-auth cycles — meaningful on calls because ITU-T G.114 sets 150 ms as the threshold for high-quality interactive voice

If your AHT is elevated, separate it into its component drivers. You can reduce AHT substantially without "talking faster" if you reduce holds and reduce ACW.

How to reduce AHT without harming customer experience

Treat AHT improvement as a set of levers. Each lever maps to one of the AHT components.

Reduce talk time (without sounding rushed)

Talk time drops when the interaction is structured, the right questions are asked early, and the conversation doesn't loop.

High-impact tactics:

  • Improve intent capture in the opening (one or two clarifying questions beats five minutes of backtracking).
  • Standardize the "goal → required inputs → resolution → next steps" structure.
  • Use guided scripts for required disclosures, but keep tone natural.
  • Provide agent-assist prompts that surface the next best question/action at the right time, a pattern catalogued in Gartner's research on conversational AI and agent assist.

The goal is fewer loops and fewer repeats, not fewer seconds of empathy.

Reduce hold time

Hold time is usually a symptom of searching, waiting, or needing another team's help.

High-impact tactics to reduce hold time:

  • Enhance agent knowledge and efficiency: Provide agents with a readily searchable knowledge base for immediate guidance on next steps and proper dialogue.
  • Speed up customer data retrieval: Implement systems for faster, ideally unified, access to a customer's complete profile.
  • Improve call routing: Employ more intelligent routing rules to minimize unnecessary call transfers.
  • Simplify escalation and procedures: Clearly define policies and escalation criteria to reduce the time agents spend on hold seeking supervisor guidance.

If you reduce hold time, you often reduce both AHT and customer effort simultaneously — the relationship documented in HBR's customer effort research.

Reduce after-call work (ACW)

ACW is frequently the most automatable part of AHT.

High-impact tactics:

  • Auto-summarization of the interaction into a structured note
  • Auto-tagging and disposition suggestions with human approval
  • Automated CRM updates and case creation
  • Templates for standard outcomes (resolved / follow-up / escalated / recovery plan offered)

Amazon Connect explicitly tracks After Contact Work (ACW) as part of handled/handle time in its data definitions for handle time.

Automation that reduces AHT end-to-end

Automation can reduce AHT materially when it:

Where voice automation typically helps most:

  • Repetitive service inquiries
  • Appointment setting and rescheduling
  • Status and next-step calls that don't require deep judgment
  • Payments and recovery calls that follow consistent rules, with exceptions routed to humans.

The measurement rule: if AHT drops but repeat contacts rise, you didn't reduce work — you displaced it.

AHT vs related metrics (and why SEO readers care)

AHT becomes more useful when you compare it to the adjacent metrics that determine whether faster is actually better:

AHT vs ASA: ASA measures how long customers wait before an agent answers; AHT measures the total time to handle the interaction once connected. The AWS Amazon Connect metric definitions document the boundary clearly.

AHT vs FCR: FCR measures whether the issue is resolved in one interaction. Lower AHT with lower FCR is usually a net loss — the SQM Group's FCR research documents this tradeoff in detail.

AHT vs CSAT/CES: These tell you whether your speed improvements reduced customer effort or created rushed, incomplete outcomes. Industry-scale CSAT is tracked by the ACSI; CES was developed in Gartner research and popularized in HBR.

AHT vs Transfer Rate: Transfers can make AHT look "better" for one group while increasing total labor across the operation — a measurement issue the COPC standard flags explicitly.

How Acclaim helps reduce AHT

Acclaim is an AI CX platform deploying GOAL-driven AI agents that recover more in collections, resolve service requests, and delight customers — built for banks, credit unions, and fintechs, and live in only weeks on your infrastructure.

To improve AHT (Average Handle Time), interactions must be more direct and less repetitive, minimizing any need for manual searching and documentation. This is achieved by reducing hold time, decreasing After-Call Work (ACW), improving call routing efficiency, and eliminating repetitive information requests that just increase talk time.

How Acclaim typically contributes to lower AHT:

  • Goal-oriented interaction flows prevent drift and reduce backtracking
  • Stronger intent capture and routing to reduce transfers
  • Context-aware responses that reduce repetition and clarification loops
  • Multilingual handling to reduce handoffs
  • Guardrails that keep required steps consistent, reducing rework and audit exposure — consistent with the deterministic-control pattern documented in NIST AI 600-1
  • Operational controls that let teams refine workflows without long development cycles

Where AHT reduction is often most visible:

  • High-volume service interactions with repeatable patterns
  • Scheduling and appointment setting
  • Payments and recovery interactions that follow defined policy and disclosure steps
  • "Status and next steps" calls that don't require extended human judgment

Frequently Asked Questions about AHT (Average Handle Time)

What does AHT stand for? AHT stands for Average Handle Time, the average total time spent handling an interaction end-to-end. This typically includes talk time, hold time, and after-call work.

What is included in AHT? Average Handle Time (AHT) includes talk time, hold time, and post-call work (wrap-up). Organizations usually exclude customer queue time, tracking it separately with metrics like ASA or simple wait time.

How do you calculate AHT? AHT is calculated as (total talk time + total hold time + total after-call work) divided by total interactions for the period.

Is a lower AHT always better? No. Lower AHT is only beneficial if resolution quality stays high. If AHT drops because interactions are rushed or required steps are skipped, you often see repeat contacts increase and satisfaction decline.

A lower Average Handle Time (AHT) is only advantageous when the quality of resolution stays high. A decrease in AHT that results from rushed interactions or skipped essential steps often leads to a rise in repeat contacts and a drop in customer satisfaction — a relationship documented across SQM Group, HBR, and ACSI research.

What is the difference between AHT and ASA? ASA measures how long customers wait before an interaction is answered. AHT measures how long it takes to handle the interaction once it is connected with an agent, including wrap-up.

What is the difference between AHT and FCR? AHT measures time per interaction; FCR measures whether or not a customer's issue is resolved in a single interaction. AHT can improve while FCR gets worse if any efficiency gains are the result of rushing or deflecting.

How does after-call work affect AHT? After-call work is a direct component of AHT. If agents spend a significant amount of time documenting, tagging, or updating systems after the interaction, then AHT increases even if talk time stays constant.

How can AI reduce AHT safely in regulated industries? First-rank AI CX technology is effective at reducing Average Handle Time (AHT) by minimizing hold time thanks to faster information access and workflow completion, decreasing After-Call Work (ACW) via automated logging and call summaries, and ensuring compliance and quality with guardrails for proper escalation, disclosures, and verification — controls that align with the NIST AI RMF and the CFPB's compliance management system expectations.

What is a good AHT for customer service calls? A "good" AHT depends on interaction type and compliance steps. Set targets by workflow category (simple/standard/complex) and evaluate AHT alongside FCR, CSAT/CES, and QA/compliance measures.

How do you reduce AHT without reducing empathy? Remove friction rather than rushing: improve intent capture, reduce transfers, speed knowledge retrieval, and automate documentation. You preserve trust and empathy when customers don't have to repeat themselves, wait on hold, or call back for the same issue — the operating principle behind the customer-effort thesis in HBR's research on customer loyalty.

Key takeaways

  • AHT is the average total handling time per interaction, typically talk + hold + after-call work.
  • AHT targets should be set by workflow, not as one number for everything.
  • The biggest AHT improvements usually come from reducing hold time and after-call work, not "talking faster."
  • AHT should be managed with quality guardrails like FCR, CSAT/CES, and compliance adherence.
  • Automation can reduce AHT materially when it removes repetition and rework while enforcing escalation and policy steps.