Key Takeaways
- Zendesk's January 2026 guidance places a good first reply at 12 hours or less for email, 5 hours or less for social media, and 1 minute or less for live chat.
- The Zendesk figures are channel guidance assembled from several sources, not observed averages from one representative sample.
- HubSpot reports that 90% of customers consider an immediate service response important or very important, while 60% define immediate as 10 minutes or less.
- SQM Group reports that customer satisfaction falls when people must make repeat calls, but this measures first call resolution rather than first response speed.
- Salesforce found that 69% of service agents had difficulty balancing speed and quality in its survey of more than 5,500 professionals across 30 countries.
First response time statistics need labels. Email, social media, live chat, and phone queues run at different speeds. A vendor target is not the same as an observed industry average, and a fast acknowledgement is not necessarily a useful human answer.
This 2026 review separates those measures. The strongest public sources provide channel guidance, customer expectations, and evidence about resolution quality. They do not establish one universal response time that predicts customer satisfaction across every company.
First response time statistics for 2026 at a glance
| Measure | Result | Scope and limitation |
|---|---|---|
| Good email first reply | 12 hours or less | Zendesk guidance updated January 15, 2026, assembled from external sources |
| Better email first reply | 4 hours or less | Published target, not a measured cross-industry average |
| Best email first reply | 1 hour or less | Published target, not a universal SLA |
| Good social media first reply | 5 hours or less | Channel guidance sourced to Sprout Social |
| Good live chat first reply | 1 minute or less | Guidance for a synchronous channel |
| Customers who consider an immediate service response important or very important | 90% | HubSpot Research Consumer Customer Support Survey, 2018 |
| Customers who define immediate as 10 minutes or less | 60% | Same survey; the report does not disclose its sample size |
| Agents who find balancing speed and quality difficult | 69% | Salesforce survey of more than 5,500 service professionals, December 8, 2023 to January 22, 2024 |
The year in this article's title is the publication year. The table preserves the collection or publication date for each source rather than implying that all fieldwork took place in 2026.
What first response time measures
First response time, also called first reply time or FRT, measures the interval between a customer submitting a request and an agent sending the first response. Zendesk's metric documentation defines ticket first reply time as the time from ticket creation to the first public agent comment.
The definition needs three decisions before teams compare results:
- Decide whether the clock uses calendar hours or staffed business hours.
- State whether automated acknowledgements are excluded.
- Name the event that stops the clock, such as any human reply or a substantive answer.
Zendesk excludes bot and automated actions from its first reply metric. Its documentation also distinguishes calendar-hour and business-hour reporting. Those settings can produce different results from the same tickets, so every dashboard and external benchmark should disclose the clock used.
An average alone can hide the longest waits. Zendesk recommends using the median because unusually fast or slow cases can pull an arithmetic mean away from the typical experience. A practical dashboard can report the median, 90th percentile, and percentage answered within the service level agreement.
Channel benchmarks are not interchangeable
Zendesk's January 2026 guide publishes separate good, better, and best ranges by channel:
| Channel | Good | Better | Best |
|---|---|---|---|
| 12 hours or less | 4 hours or less | 1 hour or less | |
| Social media | 5 hours or less | 2 hours or less | 1 hour or less |
| Live chat | 1 minute or less | 40 seconds or less | Instant |
These ranges are useful reference points, but they are not results from one Zendesk customer survey. The page attributes the figures to Toister Performance Solutions, Sprout Social, and Call Centre Helper. The source mix means the rows should not be treated as a single representative market benchmark.
Live chat is synchronous, so its clock is measured in seconds or minutes. Email is asynchronous and may arrive outside staffed hours. Social support can include public complaints and private messages. Blending all three into one company average would obscure the queue that needs attention.
Phone support needs a separate metric. Average speed of answer measures how long callers wait for an agent, while first call resolution measures whether the issue is solved without another call. Neither is identical to ticket FRT.
What customers expect from a fast response
HubSpot's 2022 State of Service report repeats two findings from its 2018 Consumer Customer Support Survey: 90% of customers rated an immediate response as important or very important, and 60% defined immediate as 10 minutes or less.
The finding describes expectations for a customer service question. It does not show that every channel should answer within 10 minutes, and it does not establish a causal improvement in CSAT at that threshold. The PDF does not disclose the survey sample size or geography on the page containing the statistic, so those details cannot be supplied from the source.
The gap between expectation and operating reality is visible in Salesforce's sixth State of Service study. Salesforce surveyed more than 5,500 service professionals in 30 countries and five continents from December 8, 2023 through January 22, 2024. It found that 69% of agents said balancing service speed and quality was difficult, down from 76% in 2022.
That result is an agent-reported workload measure. It does not provide a channel response-time average or a CSAT cutoff. It does show why a speed target should be paired with a quality measure instead of optimized alone.
Customer satisfaction depends on resolution too
Public evidence connecting a precise FRT threshold directly to CSAT is limited. SQM Group publishes a clearer relationship for first call resolution. In its benchmark data, customer satisfaction was 78% when the issue was resolved on the first call and 64% when repeat calls were required. It also reports that top-box CSAT drops 15% on average each time a customer has to call back about the initial issue. SQM Group
SQM says its benchmark program covers 500 North American contact centers. The figures describe call resolution, not email, chat, or social first reply time. They should not be used to claim that shaving a fixed number of minutes from FRT will produce the same CSAT change.
The safer scorecard pairs speed with whether the customer received a useful answer:
| Metric | Question it answers |
|---|---|
| Median first human response | How long did a typical customer wait for a person? |
| 90th percentile first response | How long did customers in the slow tail wait? |
| SLA attainment | What share received a reply within the stated promise? |
| First contact resolution | Was another contact needed for the same issue? |
| Reopen or repeat-contact rate | Did the problem return after the initial answer? |
| CSAT by response-time band | How did surveyed satisfaction vary across actual wait bands? |
| Survey response count | How much customer feedback sits behind the CSAT result? |
CSAT by response-time band should come from the team's own data. Compare similar channels, issue types, severity levels, and staffed-hour rules. A simple password question and a complex billing dispute should not be pooled without segmentation.
How to set a defensible first response target
Start with the channel and request type. Use external guidance only as calibration, then compare it with arrival volume, staffed hours, backlog, and the team's current percentile distribution.
Write the measurement rule before publishing an SLA. State the start event, stop event, timezone, business-hour treatment, and exclusions. Keep automated receipts separate from the first human response.
Review reply quality alongside speed. Sample messages that met the target and check whether they answered the question, routed the case correctly, and avoided unnecessary follow-up. If faster replies raise repeat contacts or lower CSAT, the speed target is working against the customer outcome.
Teams that need staffed coverage can review customer service support options. The service model still needs documented response and escalation rules; adding people does not make an undefined SLA measurable.
Frequently asked questions
What is a good first response time in 2026?
It depends on the channel. Zendesk's January 2026 guidance calls 12 hours or less good for email, 5 hours or less good for social media, and 1 minute or less good for live chat. These are reference ranges compiled from several sources, not universal observed averages.
What first response time do customers expect?
HubSpot reports that 90% of customers consider an immediate response important or very important and 60% define immediate as 10 minutes or less. The cited research is from 2018 and does not disclose its sample size on the report page. It concerns customer service questions generally rather than a universal channel SLA.
Does a faster first response improve customer satisfaction?
Speed can matter, but the public sources reviewed here do not support one universal minute-by-minute CSAT effect. SQM's contact-center evidence shows that satisfaction falls when customers need repeat calls. That supports measuring resolution and repeat contact beside FRT.
Should an automated acknowledgement count as the first response?
Track it separately. Zendesk excludes bot and automated actions from its first reply metric. Reporting the first human response prevents an instant receipt from hiding a long staffed-queue wait.
Should a team use average or median first response time?
Use the median for the typical experience and add a high percentile for slow cases. An average remains useful if reported consistently, but a few extreme waits can distort it.
Sources and method notes
- Zendesk, "First reply time: 9 tips to deliver faster customer service," updated January 15, 2026. Used for the FRT definition and channel guidance. Zendesk attributes the ranges to three external sources rather than one shared field study.
- Zendesk, "Understanding ticket reply time," updated May 1, 2026. Used for the product metric definition, automated-action exclusion, and calendar-hour versus business-hour reporting.
- HubSpot, Annual State of Service Report, 2022. Repeats findings from the HubSpot Research Consumer Customer Support Survey, 2018. The report does not disclose sample size or geography for those findings.
- Salesforce, State of Service sixth-edition release, April 23, 2024. Double-anonymous survey of more than 5,500 service professionals in 30 countries and five continents, fielded from December 8, 2023 through January 22, 2024.
- Salesforce, State of Service, sixth edition. Used to confirm the reported difficulty balancing service speed and quality and the survey's international scope.
- SQM Group, "First Call Resolution Operating Strategies". Benchmark findings for North American contact centers. Used only for first call resolution and customer satisfaction, not as a first reply time study.
Tags
Ready to put this into practice?
Book a free 15-min match call
Tell us what role you're filling. We'll match you with a pre-vetted virtual assistant - or tell you honestly if we're not the right fit.
Book a free call →