First response time calculator

Short answer. Drop in or paste a CSV export of your chats or tickets. The calculator finds the columns itself and shows your median first response time, in calendar hours and in your working hours, how many customers waited over an hour, and how long the ones who wrote while you were away waited. Nothing you paste leaves your browser.

Drop your CSV export here

or paste it below. Nothing leaves your browser.

Days
Columns and date format

What the numbers mean

Median first response is what a typical customer waited for your first reply, in calendar time, nights and weekends included. It is the number your customers feel. Median in working hours counts only the minutes inside the hours you set, which is the fairer number for judging how quickly your team picks things up. A big gap between the two means most of the waiting happens while nobody is working.

Replied within an hour is the share of all conversations, including the ones that never got a reply, that heard back within 60 minutes. Never got a reply counts rows with a start time and no reply. Leaving those out makes any average look better than it was.

Under the tiles, the calculator says how many conversations came in outside your working hours and how long those customers waited compared with everyone else. For most small teams that is where the slow replies are.

How to get your export

Most help desks and chat tools have a CSV export of conversations or tickets in their reports or settings. Look for a created or started time and a first reply or first response time. If your export already has a first response time column, choose "One column with the wait" under Columns and date format. If both times are in UTC, the calculator asks for your time zone so working hours line up with your day.

Two things to check before you trust the result. Some tools count an automatic "we got your message" as the first reply, which makes the number look far better than what customers felt. And if your export mixes channels, email will pull the median up compared with live chat, so compare like with like.

Bringing it down

If most of the waiting happens outside your hours, staffing more hours is the expensive fix. The cheaper one is an assistant that answers what your docs already cover at any hour, hands everything else to you with a note saying why, and tells the customer a person will reply where they are. We measured what that did to our own first response time, and where it still fell short, in first response time, with real numbers.

Questions

How do I calculate first response time?

For each conversation, subtract the time of the customer's first message from the time of your first reply. Then take the median across conversations, and count the ones that never got a reply separately. This calculator does it for a whole export at once.

How do I calculate first response time in business hours?

Count only the minutes that fall inside your working hours between the customer's message and your reply. A message at 18:00 on Friday answered at 09:30 on Monday is 30 working minutes but over 63 calendar hours. Turn on working hours above to see both.

Where do I get the CSV?

Most help desks and chat tools export conversations or tickets as CSV from their reports or settings. You need a start time and a first reply time for each conversation, or a column that already holds the first response time.

Which date formats does it read?

ISO dates like 2026-11-08 14:03:22 or 2026-11-08T14:03:22Z, dates with slashes in day-first or month-first order (it works out which), and Unix timestamps in seconds or milliseconds.

Is my data sent anywhere?

No. The file is read and calculated by this page in your browser. Nothing is uploaded, stored or sent to us.

Read next

Support metrics First response time: how to calculate it, with real numbers What first response time measures, how to calculate it from a chat export, why the median beats the average, and what happened to ours when the support chat started replying on its own. Handover When should an AI chatbot hand off to a human? The five moments an AI support chatbot should hand a conversation to a person, what the handover message should say, and how often it happened in a live support chat's first week on its own. Buyer's guide AI customer support for SaaS: a checklist for small teams What to look for in an AI support chat when your team is small and answers its own customers: sourced answers, honest handovers, learning from every chat, knowing what to build, and working with your coding agent.