Support teams obsess over resolution time. How fast did we solve the problem? But the metric that actually predicts whether a customer is satisfied is different: how fast did we say "I hear you"?
First response time (FRT) correlates with CSAT 3x more strongly than total resolution time (Zendesk CX Trends Report, 2025). Customers who wait 30 seconds for a first reply but 2 hours for full resolution rate their experience higher than customers who wait 5 minutes for a first reply but get resolved in 30 minutes.
That is counterintuitive. But it reflects how humans process waiting.
The Psychology of Waiting
Waiting without information feels longer than waiting with a reason. This is the "uncertain wait" effect documented in operations research since the 1980s (Maister, 1985). Disney uses this principle in their theme parks. The posted wait time at a ride is always longer than the actual wait. People feel better when they wait 20 minutes after seeing "30-minute wait" than when they wait 15 minutes with no information.
Customer support works the same way.
When a customer sends a chat message and gets nothing for 3 minutes, they do not know if anyone is there. They do not know if the system is broken. They do not know if their message was received. The uncertainty creates anxiety.
When they get a reply in 15 seconds saying "I see your question about billing. Let me check on that," the uncertainty vanishes. They know they have been heard. They know someone is working on it. They relax.
The Data: FRT Predicts Satisfaction
The Zendesk CX Trends Report (2025) analyzed 4.4 million support interactions across 94,000 companies. Their findings:
FRT under 1 minute: Average CSAT of 92%. FRT 1-5 minutes: Average CSAT of 83%. FRT over 5 minutes: Average CSAT of 68%.
That is a 24-point drop between "instant" and "slow." Now compare the resolution time data from the same report:
Resolution under 1 hour: Average CSAT of 88%. Resolution 1-24 hours: Average CSAT of 82%. Resolution over 24 hours: Average CSAT of 75%.
The CSAT spread for resolution time is 13 points. For first response time, it is 24 points. FRT has nearly double the impact on satisfaction.
Forrester (2024) found similar results. Customers who received a first response within 30 seconds gave an average satisfaction score of 4.6/5. Those who waited over 3 minutes gave 3.2/5, regardless of how quickly the issue was ultimately resolved.
Why Resolution Time Gets More Attention
If FRT matters more, why do most dashboards emphasize resolution time?
Resolution time is easier to manage. It has a clear start (customer contacts you) and clear end (issue closed). Managers can set SLAs around it. It fits neatly into reports.
FRT feels less actionable. Most teams think the only way to improve FRT is to hire more agents. That is expensive. So they focus on what they can optimize: the resolution process.
The metric is misunderstood. Many teams measure "average response time" across all messages, not specifically the first one. This dilutes the signal. The first response is special. Subsequent responses matter less.
How to Reduce First Response Time
Five approaches, ordered by impact.
1. AI auto-reply (biggest impact)
An AI chatbot responds in under 3 seconds. For 60-80% of questions, it provides the full answer. For the rest, it acknowledges the question and starts gathering information.
This single change can drop your average FRT from minutes to seconds. If AI handles 70% of first contacts instantly and your team handles 30% in 90 seconds, your blended FRT is about 27 seconds.
Helpable's AI answers feature does exactly this. The chatbot reads your knowledge base articles and responds to customer questions immediately. When it cannot answer, it gathers the question details and routes to your team.
2. Auto-acknowledgment messages
For conversations that need a human, send an immediate auto-acknowledgment: "Thanks for reaching out. A team member will be with you shortly." This is not a full response, but it tells the customer their message was received.
The effect on perceived FRT is significant. Even though the "real" first response comes later, the auto-acknowledgment breaks the silence. Customer anxiety drops.
3. Team routing rules
Round-robin assignment spreads chats evenly but ignores availability. An agent who is deep in a complex conversation gets assigned a new chat. Their FRT for that new chat suffers.
Better: route to the agent with the lowest current workload. If Agent A handles 2 chats and Agent B handles 4, the new chat goes to Agent A. Most chat platforms support workload-based routing.
4. Canned responses for common openings
The first reply does not need to be the full answer. It needs to acknowledge the question and set expectations. Build canned responses for your top 10 conversation openers.
Example: Customer asks "How do I add a team member?" The canned first response: "Great question. You can add team members from Settings > Team. Let me walk you through it." This takes 5 seconds to send. The detailed walkthrough follows.
5. Peak-hour staffing
Your FRT average is misleading if it blends quiet hours (FRT: 15 seconds) with peak hours (FRT: 4 minutes). Staff for the peak, not the average.
Check your volume-by-hour chart. Find the 4-hour window with the most chat volume. Add one agent during that window. This improves your worst FRT numbers, which are the ones customers complain about.
The Resolution Time Trap
Focusing only on resolution time creates perverse incentives.
Agents rush resolutions. If resolution time is the target, agents close conversations quickly whether or not the issue is actually resolved. Re-contact rates go up. Actual customer experience goes down.
Complex issues get deprioritized. An agent can resolve five simple questions in the time it takes to investigate one complex bug. If resolution time is the metric, agents avoid the hard problems.
First response gets ignored. A team optimizing for resolution time might let a customer wait 5 minutes for a first reply because the agent is finishing a previous conversation. The resolution metric looks good. The customer feels ignored.
The healthier approach: set an FRT target that is aggressive (under 1 minute) and a resolution target that is realistic (varies by issue type). Track them separately.
FRT Benchmarks by Channel
| Channel | Good FRT | Average FRT | Poor FRT |
|---|---|---|---|
| Live chat (with AI) | Under 10 seconds | Under 30 seconds | Over 1 minute |
| Live chat (humans only) | Under 1 minute | Under 2 minutes | Over 3 minutes |
| Under 1 hour | Under 4 hours | Over 12 hours | |
| Social media | Under 30 minutes | Under 2 hours | Over 6 hours |
Live chat with AI sets the bar. Customers who experience 3-second AI responses on chat expect faster email responses too. Raising the bar on one channel raises expectations across all channels.
FAQ
Is first response time the same as average response time?
No. First response time measures only the gap between the customer's first message and the first reply. Average response time includes all messages in the conversation. FRT matters more for satisfaction because it determines the customer's first impression of the support experience.
Does an AI auto-reply count as a first response?
Yes. If the AI provides a relevant answer or acknowledgment, customers perceive it as a first response. A generic "thanks for reaching out" is less effective than an AI response that addresses their specific question. The quality of the auto-reply matters.
What FRT should I target for live chat?
Under 30 seconds with AI. Under 1 minute for human-only teams. The industry average is 1 minute 35 seconds (Tidio, 2025). If you are above 2 minutes, customers are abandoning your chat before you reply.
How do I balance FRT with resolution quality?
Use AI for the first response and humans for the resolution. AI brings FRT to under 5 seconds. Your team focuses on solving the actual problem without time pressure on the first reply. This decouples speed from quality.
Should I set different FRT targets for different customer segments?
Yes. Enterprise or paying customers should have stricter FRT targets than free-tier users. Helpable's live chat routes conversations to your inbox where you can prioritize by customer type. The goal is to respond faster to higher-value conversations without ignoring anyone.