Stop Comparing Yourself to "Industry Average"

If I see one more benchmarking report that says "the average handle time across industries is 6 minutes," I'm going to lose it. That number is useless. It averages a 2-minute password reset with a 45-minute healthcare claim dispute and calls it a benchmark.

AHT only means something in context. Your industry, your channel, your ticket complexity, and your customer expectations all determine what "good" looks like.

Here are the real numbers.

AHT Benchmarks by Industry (Phone Channel)

SaaS / Software

Median AHT: 4-6 minutes. Top quartile: under 3.5 minutes. Bottom quartile: over 9 minutes.

SaaS support is typically technical but well-documented. Password resets, configuration questions, billing changes. The top-performing teams achieve low AHT through strong self-service (so only the harder issues reach agents) and good internal knowledge bases.

If your SaaS phone AHT is over 8 minutes consistently, your agents are either handling issues that should be self-service, or they're spending too long searching for answers.

E-Commerce / Retail

Median AHT: 5-7 minutes. Top quartile: under 4 minutes. Bottom quartile: over 10 minutes.

E-commerce support is high-volume and repetitive. "Where's my order?" "I want a refund." "This arrived damaged." These are fast interactions when agents have order lookup tools integrated. The wide spread between top and bottom quartile usually reflects tooling, not agent skill. Teams with order management systems integrated into their helpdesk resolve faster.

Financial Services / Banking

Median AHT: 8-12 minutes. Top quartile: under 7 minutes. Bottom quartile: over 15 minutes.

Financial support takes longer because of verification requirements, compliance scripts, and the complexity of financial products. A dispute resolution or fraud claim inherently takes more time than a shipping status check. Trying to push AHT below 7 minutes in financial services usually means cutting corners on compliance.

Healthcare

Median AHT: 10-15 minutes. Top quartile: under 9 minutes. Bottom quartile: over 20 minutes.

Healthcare has the highest AHT of any industry. Calls involve sensitive information, complex insurance questions, appointment coordination across providers, and patients who need things explained thoroughly. Rushing healthcare calls is a patient safety risk.

Telecommunications

Median AHT: 7-10 minutes. Top quartile: under 6 minutes. Bottom quartile: over 13 minutes.

Telecom AHT is inflated by troubleshooting calls (rebooting routers, checking signal strength) and retention calls (customer wants to cancel, agent tries to save). The troubleshooting component could be reduced significantly with better self-service tools and proactive monitoring.

AHT Benchmarks by Channel

Phone

Overall median: 6-8 minutes. This is the baseline channel for AHT measurement.

Live Chat

Overall median: 8-11 minutes per conversation. Longer than phone because chat has natural pauses, typing delays, and customers multitasking. But because agents handle 2-4 chats simultaneously, the effective AHT per agent hour is lower than phone.

Don't compare chat AHT directly to phone AHT. They're measuring different things. A 10-minute chat costs less than a 6-minute phone call because the agent was handling two other chats during those 10 minutes.

Email

Overall median: 12-18 minutes of agent work time per ticket. This includes reading, research, composing, and review. Email AHT is harder to measure because it's not a continuous interaction. Some teams measure "touch time" (total time an agent spends working on a ticket) rather than handle time.

Email's advantage: no dead air. Every minute the agent spends is productive. On phone, agents spend 15-20% of handle time on hold, transfers, and system loading.

AI Classification

AHT: effectively zero for automated tickets. Classification happens in 100-200ms. If the classification triggers an automated action, the entire interaction takes under a second of compute time.

For tickets that are classified and routed to a human, AI reduces AHT by 30-45 seconds on average because the agent receives the ticket pre-categorized with relevant context attached.

Why Optimizing AHT Alone Is Dangerous

AHT is seductive because it's measurable, comparable, and directly tied to cost. Every 30 seconds you shave off AHT across 10,000 monthly calls saves roughly $2,500/month in agent time. That's real money.

But here's the trap.

The resolution quality trade-off

Teams that pressure agents to reduce AHT often see first contact resolution (FCR) drop. Agents rush through calls, skip root cause analysis, and apply quick fixes that don't stick. The customer calls back. Now you've paid for two interactions instead of one.

A 2025 MetricNet study found that teams in the bottom quartile of AHT (meaning the fastest handlers) had FCR rates 12 percentage points lower than median performers. The fastest agents were generating the most repeat contacts.

Customer satisfaction erosion

CSAT and AHT have a U-shaped relationship. Very short handle times correlate with lower satisfaction (customer felt rushed). Very long handle times correlate with lower satisfaction (customer felt things took too long). The sweet spot is in the middle, and it varies by issue type.

For simple issues, customers want speed. Get in, get the answer, get out. For complex or emotional issues, customers want thoroughness. Feeling heard matters more than minutes saved.

The right metric is cost per resolution, not cost per minute

If a 10-minute call resolves the issue permanently and a 4-minute call generates a callback, the 10-minute call was cheaper. Measure resolution, not speed.

How to Actually Improve AHT Without Breaking Things

Reduce agent research time

Most of the "waste" in AHT isn't the conversation. It's the agent searching for information, switching between systems, and reading past ticket history. Better tooling (unified customer view, pre-populated context from AI classification, integrated knowledge base) reduces AHT without rushing the customer.

Deflect simple tickets to self-service and AI

If your AHT is high because agents are spending 3 minutes on password resets and order status checks, the fix isn't faster agents. It's routing those tickets to automation. This naturally increases your remaining AHT (because you've removed the easy tickets) but decreases your cost per ticket overall.

Set AHT targets by ticket category, not globally

A 4-minute target makes sense for billing inquiries. It's absurd for technical escalations. Category-specific targets let agents take the right amount of time for each issue type.

Monitor AHT alongside FCR and CSAT

Never look at AHT in isolation. Report all three together. If AHT drops and FCR drops, you've made things worse. If AHT drops and FCR holds steady, you've genuinely improved.

The Bottom Line

AHT is a useful metric for operational planning and cost modeling. It's a terrible metric for performance management when used alone. The best support organizations track it, benchmark it against their own historical data (not generic industry averages), and use it as one input among several.

Know your numbers. But don't let them pressure you into faster-but-worse support.