Request Response Time
Definition
Request Response Time
Request response time is the interval between a customer submitting a request and someone giving them a substantive first reply. It is the wait before anything happens, and it shapes customer satisfaction far more strongly than the resolution speed itself ever does.
People tolerate a slow fix far better than silence — knowing the request landed with a real person changes the experience completely.
That is why the definition of a reply matters so much. An automated acknowledgement proves the system worked, not that anybody looked.
Counting auto-replies as responses is the most common way this measure gets quietly broken. The number improves — and the customer experience does not change at all.
Key takeaways
- Request response time runs from submission to the first substantive human reply.
- Automated acknowledgements are not responses and should never stop the clock.
- Report a percentile alongside the average, because the slow tail drives complaints.
- Elapsed time is the honest basis, even when a team works business hours only.
How it works
Timestamp the request when it arrives and again when a substantive reply is sent, then take the difference. Average those intervals across the period and publish a percentile beside the mean so the slow tail stays visible.
Two definitions decide the result. What starts the clock and what stops it can move the reported figure by hours.
| Decision | Options | Effect on the figure |
|---|---|---|
| Clock start | Submission or triage completion | Triage start hides queue time |
| Clock stop | Auto-reply, human reply, or full answer | Auto-reply can cut the figure to seconds |
| Calendar basis | Elapsed hours or business hours | Business hours can halve the number |
| Reopened requests | New clock or continuation | Continuation lengthens the tail |
Averages hide the cases that matter — a mean of two hours can conceal a group waiting three days, and those are the customers who escalate, complain, or leave.
Regulators have taken a view on response speed in at least one channel. Under the Federal Trade Commission’s Telemarketing Sales Rule, a live representative must come on the line within two seconds of a consumer answering a telemarketing call.
Service standards make the same argument for digital channels. Digital.gov frames public digital services around whether people can actually complete what they set out to do, and an unanswered request fails that test immediately.
Examples
Response expectations vary enormously by channel, severity, and what the customer is waiting to do next. Four cases show how targets get set in practice.
A B2B software support desk. Severity one requests carry a 15-minute response target and severity three carries one business day. Blending them into a single average made both meaningless.
A Manila email support team. The reported figure was 4 minutes because auto-acknowledgements stopped the clock. Measuring human replies instead gave a true figure of 6.5 hours.
A property management helpdesk. Response time is measured in elapsed hours including weekends, because a burst pipe does not wait for Monday. Weekend cover was added once the data showed the gap.
A recruitment agency. Candidate replies within two hours converted to interviews at twice the rate of same-day replies. Response speed was treated as a revenue measure, not a service one.
Related terms
Request response time sits among the timing measures and commitments that describe how quickly a request is picked up. The terms below cover the variants and the contracts around them.
- First Response Time (FRT): the contact centre name for the same interval.
- Average Response Time: the aggregated form usually published in reports.
- Internal Response Time: the equivalent measure for internal requesters.
- Support Ticket: the unit the clock is attached to.
- Service Level Agreement (SLA): the contract carrying the response target.
- Auto Response: the automated acknowledgement that must not stop the clock.
- Key Performance Indicator (KPI): the reporting family it belongs to.
FAQ
What is a good request response time?
Live channels are judged in seconds, email support in hours, and low-severity requests in business days. The right target follows what the customer is blocked from doing.
Do automated acknowledgements count as a response?
No. They confirm receipt rather than engagement, and counting them produces a figure that bears no relation to the customer experience.
Should the clock run outside business hours?
Report elapsed time as the primary figure because that is what customers experience. A business-hours version is useful internally.
Why report a percentile as well as an average?
Because averages hide the slow tail. The ninety-fifth percentile shows what the worst-served requests actually waited.
How is it different from resolution time?
Response time measures the wait for a first reply, while resolution time measures the wait for the problem to be fixed.
What improves response time fastest?
Triage automation and clear ownership rules. Most delay comes from requests sitting unassigned rather than from slow replies.
Find support partners with published response commitments in the Outsource Accelerator directory.
Independent
