CRMsoftware.pro

Average resolution time calculator

Work out how long your team takes to resolve a ticket: total resolution time ÷ resolved tickets. Paste the times from your help desk to also see the median, the outliers that pull the average up, the share resolved within your SLA, business hours vs calendar hours for a ticket, a weighted average across email, chat and phone and a comparison with 2026 benchmarks. Free, no sign-up, nothing leaves your browser.

Resolution times

Paste a column from your help desk export: one value per line, or separated by spaces, commas or semicolons.
h
Share of tickets resolved within this time

Or totals, if you have no per-ticket times

h
Of all tickets resolved in the period
In the same period

By channel (optional)

Email
h
Chat
h
Phone
h

Leave a channel empty if you don't use it.

One ticket: business vs calendar hours (optional)

Working days

Public holidays are not excluded.

Freshworks 2026 medians, sources below

Resolve the simple questions instantly

Most tickets that wait are simple questions sitting in a queue. An instant answer resolves them in seconds and leaves your team the hard ones.

CRMsoftware.pro puts an AI agent on your website and in your messengers. It answers questions about prices, orders and opening hours in seconds, day and night, and hands the rest to your team with the full conversation saved in the CRM, so nobody has to ask the customer twice.

  • AI agent answers on your website, WhatsApp and Telegram
  • Every chat saved to the contact and the deal
  • Tasks and reminders so no ticket waits without an owner

Free plan, no credit card. Paid plans from $19/month, billed yearly.

The formula

How to calculate average resolution time

Resolution time is the time from when a customer opens a ticket to when it is resolved. Average resolution time is the mean of that time across all tickets resolved in a period.

Resolution time = resolved at − opened atAverage resolution time = total resolution time ÷ resolved ticketsShare within target = tickets resolved within the target ÷ resolved tickets × 100

Count each ticket in the period in which it was resolved, not the one in which it was opened. Freshservice describes its own report this way: the total time taken to resolve tickets in the period, divided by the number of tickets resolved in that period.

Decide how to treat reopened tickets and stay consistent. Zendesk’s full resolution time runs from creation to the last time a ticket was solved, so a reopened ticket counts its whole life.

Resolution time and the metrics next to it

MetricWhat it measuresCalculator
First response timeFrom the request to the first replyFirst response time calculator
Average resolution timeFrom the request to the solution, including every waitThis page
Average handle timeTime an agent actively works on one contactAHT calculator
First contact resolutionShare of issues solved in the first contactFCR calculator
Average

15 h

‎150 ÷ 10 = 15 h

10 tickets, 150 hours in total.

Median

5.5 h

‎(5 + 6) ÷ 2 = 5.5 h

Half of the tickets took less.

Within 24 h

90%

‎9 ÷ 10 × 100 = 90%

Resolved within the SLA.

Worked example

One support team, step by step

An illustrative team, not real data. Press Show example in the calculator to load these numbers.

Last week a small support team resolved 10 tickets. Their resolution times in hours were 1, 2, 3, 4, 5, 6, 8, 10, 15 and 96. The team promises a resolution within 24 hours and compares itself with the Freshworks median for small businesses.

‎1. Average: (1 + 2 + 3 + 4 + 5 + 6 + 8 + 10 + 15 + 96) ÷ 10 = 150 ÷ 10 = 15 h

‎2. Median: the middle of the sorted list, (5 + 6) ÷ 2 = 5.5 h. The average is 2.7 times the median

‎3. Outliers: Q1 = 3.25 h, Q3 = 9.5 h, so the fence is 9.5 + 1.5 × 6.25 = 18.88 h. One ticket (96 h) is above it and holds 64% of all resolution time. Without it the average is 54 ÷ 9 = 6 h

‎4. Within the 24-hour target: 9 of 10 tickets = 90%; the 90th percentile is 23.1 h

‎5. Benchmark: the Freshworks median for small businesses is 7 h 3 min. The team’s average is 113% higher (slower), its median 22% lower

‎6. By channel, for the whole month: email 120 tickets × 20 h, chat 300 × 2 h, phone 80 × 6 h = 3,480 h ÷ 500 tickets = 6.96 h. The simple average of the three channel averages, 9.33 h, would be wrong

‎7. Business hours: a ticket opened on Friday, 13 November at 16:00 and resolved on Monday, 16 November at 11:00 took 67 calendar hours but only 3 business hours with a 9:00 to 17:00, Monday to Friday schedule

One stuck ticket more than doubled the average. The fix is not to work faster on every ticket, but to find out why that one waited four days, and to report the median next to the average.

Mean vs median

Why the average misleads, and what to report next to it

Resolution times are skewed: most tickets close fast, a few wait for days, and those few pull the average up.

Median = the middle value of the sorted resolution timesOutlier fence = Q3 + 1.5 × (Q3 − Q1)Average without outliers = time of the tickets below the fence ÷ their number

The median tells you what a typical customer experiences. The average tells you how much waiting your team produced in total, which matters for workload. Report both. Zendesk Explore, for example, shows full resolution time as a median.

The calculator marks a ticket as an outlier when it is above the upper fence, Q3 + 1.5 times the interquartile range (Tukey’s rule). It needs at least 4 tickets. Q1 and Q3 are calculated like QUARTILE.INC in Excel.

Do not delete outliers from your report, read them. A ticket that waited 96 hours usually points to a handoff, a missing owner or a customer who never replied, and that is the cheapest place to save time. Help Scout also suggests looking at unusually long cases separately, so that a few of them do not hide the trend.

Track the 90th percentile too: 9 in 10 tickets are resolved faster than this. It is often the best number to put in an SLA.

Working hours

Business hours or calendar hours?

The same ticket can take 3 hours or 67 hours, depending on which clock you use.

Calendar hours = resolved at − opened atBusiness hours = only the time inside the working day, on working days

A ticket opened on Friday at 16:00 and resolved on Monday at 11:00 waited 67 hours, a whole weekend. With a 9:00 to 17:00, Monday to Friday schedule, only 3 of those hours fall inside working time: 1 on Friday and 2 on Monday.

TicketCalendar hoursBusiness hours, 9:00–17:00, Mon–Fri
Fri 16:00 → Mon 11:00673
Tue 10:00 → Tue 15:305.55.5
Tue 20:00 → Wed 10:00141

Business hours show how fast your team works while it is at work, which makes them fair for comparing agents and shifts. Calendar hours show how long the customer actually waited. Help Scout makes the same point: business-hours filters help judge the team, but the customer still waits the full time.

Report both and say which one each number is. Use the calendar figure for promises to customers and the business figure for team targets. The calculator does not know your public holidays: if one falls inside the ticket, subtract its working hours yourself.

Channels

Average resolution time by channel

Chat, email and phone resolve at different speeds. Weight each channel by its tickets.

Overall average = Σ (tickets × average per channel) ÷ Σ tickets

In the example, email takes 20 h, chat 2 h and phone 6 h. The simple average of the three is 9.33 h, but chat has 300 of the 500 tickets. Weighted by tickets, the overall average is 3,480 ÷ 500 = 6.96 h.

Email has 24% of the tickets but 69% of all resolution time. That is where a faster process pays off most.

In the Freshworks 2026 data, the median resolution time is shortest on WhatsApp (2 h 6 min) and Facebook (2 h 11 min) and longest on email (6 h 21 min) and web portals (8 h 52 min). The calculator shows the matching median next to each channel.

Benchmarks

What is a good average resolution time?

Every figure below is quoted from the source linked under the page. Compare with your own past months first, then with the closest segment.

Freshworks analysed 1.2 billion tickets from 33,889 accounts in 63 industries, using anonymised Freshdesk usage data from June 2025 to May 2026. The report gives the median across organizations and the top 20th percentile, which it calls trendsetters. It does not say whether the times are business or calendar hours, so treat them as a rough guide.

Resolution time by channel: Freshworks 2026

ChannelMedianTop 20%Resolution SLA met, median
Chat4 h 14 minInstant92.51%
Phone4 h 2 min3 min92.51%
WhatsApp2 h 6 min21 min89.04%
Facebook2 h 11 min15 min88.26%
Email6 h 21 min42 min93.92%
Web portal8 h 52 min46 min86.57%

By company size and industry

SegmentMedianTop 20%Resolution SLA met, median
Small business, up to 250 employees7 h 3 min38 min93.39%
Mid-market, 250 to 5,000 employees6 h 49 min31 min92.02%
Enterprise, 5,000+ employees10 h 37 min1 h 17 min93.85%
Retail and e-commerce7 h 46 min44 min90.71%
Software and internet6 h 6 min18 min92.51%

Times are rounded to the minute. A median of about 7 hours means that half of the organizations in the data take longer. The median team meets its resolution SLA on 86.57% to 93.92% of tickets, depending on the channel, so a share far below that usually means the target or the process needs work.

Spreadsheets

Average resolution time formulas for Excel and Google Sheets

One ticket per row: resolution time in hours in column B, from B2 to B1001. For the channel mix, tickets in C2:C4 and the average per channel in D2:D4. Opened and resolved timestamps in E2 and F2.

WhatFormulaExample result
Average=AVERAGE(B2:B1001)15 h
Median=MEDIAN(B2:B1001)5.5 h
90th percentile=PERCENTILE.INC(B2:B1001,0.9)23.1 h
Share within 24 h (format as %)=COUNTIF(B2:B1001,"<=24")/COUNT(B2:B1001)90%
Outlier fence=QUARTILE.INC(B2:B1001,3)+1.5*(QUARTILE.INC(B2:B1001,3)-QUARTILE.INC(B2:B1001,1))18.88 h
Weighted channel average=SUMPRODUCT(C2:C4,D2:D4)/SUM(C2:C4)6.96 h
Resolution time from two timestamps=(F2-E2)*2467 h
Business hours, 9:00 to 17:00, Mon–Fri=((NETWORKDAYS(E2,F2)-1)*("17:00"-"9:00")+IF(NETWORKDAYS(F2,F2),MEDIAN(MOD(F2,1),"17:00","9:00"),"17:00")-MEDIAN(NETWORKDAYS(E2,E2)*MOD(E2,1),"17:00","9:00"))*243 h

The formulas use English function names, which Google Sheets always understands. Excel uses the names of its interface language, for example MITTELWERT for AVERAGE in German or MOYENNE in French; switch to the CRMsoftware.pro page in your language to see them.

NETWORKDAYS counts Monday to Friday. To skip public holidays, add a range with their dates as the third argument of each NETWORKDAYS. The Excel formula button in the calculator copies the average, and the CSV download holds every number from your result.

FAQ

Average resolution time questions

How do you calculate average resolution time?

Add up the resolution time of all tickets resolved in the period and divide by the number of those tickets. 10 tickets with 150 hours in total have an average resolution time of 15 hours.

What is the difference between resolution time and first response time?

First response time ends with the first reply. Resolution time ends when the problem is solved, so it includes every wait in between: for the customer, for another team, for a fix.

Should I report the average or the median?

Both. The median shows the typical ticket and is not moved by a few stuck ones. The average shows the total waiting your team produced. In the worked example one 96-hour ticket lifts the average to 15 h, while the median is 5.5 h.

What is a good average resolution time?

In the Freshworks 2026 data the median is 7 h 3 min for small businesses, 6 h 49 min for mid-market and 10 h 37 min for enterprise companies, and ranges from 2 h 6 min on WhatsApp to 8 h 52 min on web portals. Compare with your own past months first.

Should I use business hours or calendar hours?

Business hours for team targets, calendar hours for what customers feel. A ticket from Friday 16:00 to Monday 11:00 is 3 business hours but 67 calendar hours.

How do I calculate the average across channels?

Multiply each channel’s average by its tickets, add the results and divide by all tickets. Do not average the channel averages: in the worked example that gives 9.33 h instead of 6.96 h.

Do pending and on-hold times count?

Full resolution time counts everything from opening to solution, including time spent waiting for the customer. Some help desks can also report resolution time without pending or on-hold time. Pick one definition and keep it.

How can we reduce resolution time?

Start with the slowest tickets and channels and find out where they wait. Clear ownership, fewer handoffs, a knowledge base and an AI agent that answers simple questions instantly all shorten resolution time.

Does the calculator store my data?

No. Everything runs in your browser. The share link keeps your numbers in the address after the # sign, which is not sent to our server.

References

Sources

  1. Median and top-20% resolution time by channel, company size and industry, resolution SLA %, 1.2 billion tickets from 33,889 accounts, June 2025 to May 2026: Freshworks, "The Customer Service Benchmark Report 2026" (PDF, report page), checked November 20, 2026
  2. Full resolution time from ticket creation to the last solve, reported as a median: Zendesk, "Explore recipe: Reporting on full resolution time", checked November 20, 2026
  3. Average resolution time as the total time taken to resolve tickets divided by the tickets resolved in the period: Freshservice, "How is the Average resolution time calculated in reports?", checked November 20, 2026
  4. Business-hours filters and the full wait of the customer, looking at unusually long cases separately: Mathew Patterson, Help Scout, "Time To Resolution: What It Is, Why It Matters, and How to Reduce It", published March 27, 2024, checked November 20, 2026

More free support tools

Related guides and templates

Answer visitors before they leave.

Start on the Free plan: an AI agent answers questions on your website day and night and saves every conversation to one CRM, so your team can follow up. No credit card.

Start free →