WhatsApp support staffing and response time calculator
Enter how many chats arrive at your busiest hour, how long each one takes and your first-response target. The calculator uses the Erlang C queue model to estimate how many agents must be online, and how many to schedule once breaks and other work are counted.
Runs 100% in your browser: no account, no upload. Nothing you type or open in this tool is sent to any server.
Your peak hour
At the busiest hour.
From first reply to close.
2 to 3 is common on WhatsApp.
Breaks, training, meetings, other work.
Result
- Agents online at peak
- 4
- On the schedule (with shrinkage)
- 6
- Within target
- 88.7%
- Average wait
- 40 s
Load: 5.3 chats in progress on average. Agent occupancy: 66.7%.
| Agents | Within target | Occupancy |
|---|---|---|
| 3 | 39.6% | 88.9% |
| 4 | 88.7% | 66.7% |
| 5 | 98.4% | 53.3% |
| 6 | 99.8% | 44.4% |
How to use it
- 1Describe the peak hour. New chats per hour and the minutes each chat keeps an agent busy, from the first reply to the end.
- 2Set parallel chats. How many conversations one agent handles at once. Each agent is counted as that many parallel slots.
- 3Set the target. For example 80% of chats answered within 120 seconds, plus the share of paid time agents are unavailable.
- 4Read the result. Agents online at peak, agents to schedule, the share answered within target, the average wait and a table with one agent fewer and up to two more.
- Erlang C assumes chats arrive at random and nobody gives up while waiting, so it errs on the safe side when customers abandon.
- Treating parallel chats as independent slots is an approximation: people get slower when juggling, so keep the number realistic.
- Use your own numbers from your support history; the defaults are only an example.
Frequently asked questions
What is Erlang C?
A queueing formula used in contact centers since the early 1900s. From arrival rate, handling time and number of agents it gives the probability of waiting and the expected wait.
Where do I get handling time?
From your helpdesk or CRM reports, or by timing a sample of chats from first reply to close. Our chat analyzer can show reply patterns from an exported chat.
Why is the scheduled number higher?
Agents are not available all paid hours. With 30% unavailable time, 10 agents online need about 15 on the schedule.
Is my data stored?
No. The numbers are calculated in your browser.
Related tools
WhatsApp is a trademark of Meta Platforms, Inc.; these tools and SyncZap are not affiliated with, endorsed or sponsored by WhatsApp or Meta. Results are generated in your browser from what you enter; check them before you publish or print.