Single interval
Two fingers alternating on one button, each pressing while the other lifts. The fastest technique a human can hold cleanly, and the one that runs into the hardware first.
Clicks per second of the interval
The bars fill as you click. A flat profile means you held your rate; a falling one means the interval outlasted your finger.
The interval
Butterfly click test
Butterfly clicking splits the work: index and middle finger alternate on the same button so neither has to wait for its own recovery. Done well it reaches 600–950 CPM, and it stays cleaner under load than jitter because it spends coordination rather than tension.
The cost is rhythm. Two fingers at slightly different rates produce a run full of doubled and dropped presses, and the per-second chart shows it as alternating tall and short bars.
Every mouse switch ignores state changes for a few milliseconds after a press, to reject the electrical chatter of the contact settling. That guard window sets a hard floor on how close two accepted presses can be, and butterfly is the technique that reaches it.
Above roughly 900 CPM you are no longer measuring your hand: the longest-gap and per-click figures start describing the switch's debounce interval and reset time. That is the point where the calibration scale stops naming human tiers and reads instrument-limited instead.
Calibration
Where your reading sits on the CPM scale
Move the marker to read the scale, or run an interval and it snaps to your own result. Bands are set on a ten-second interval, the length most people can hold honestly.
| Band | Clicks per minute | Per second | What it takes |
|---|---|---|---|
| Casual | 0–300 | 0.0–5.0 | An unhurried finger — one press at a time, no method behind it. |
| Steady | 300–450 | 5.0–7.5 | A relaxed wrist and a light switch. Most people land here on a first honest try. |
| Quick | 450–600 | 7.5–10.0 | Deliberate practice: the finger stays loaded and never fully leaves the switch. |
| Rapid | 600–780 | 10.0–13.0 | Jitter or a trained two-finger butterfly, plus a switch with a short reset. |
| Elite | 780–1000 | 13.0–16.7 | Butterfly at pace, held clean. Rare beyond a five-second interval. |
| Instrument-limited | 1000+ | 16.7+ | Past here you are usually reading the mouse, not the hand — drag clicking or switch chatter. |
Other click tests
The other intervals
The same clock and the same arithmetic, over a different span. A rate that holds across two intervals is a rate you actually have.
The main chronometer, every interval on one pad, and the long-form method.
Full benchMeasure your peak click rate across a single second. One second is too short to pace, so it reads raw burst frequency and projects it to clicks per minute.
OpenA five-second click speed test: long enough to need a technique, short enough that fatigue never arrives. The window jitter clicking is normally measured in.
OpenThe ten-second click speed test, the interval most tests settled on and the one Minecraft PvP adopted as Kohi. Reports clicks, CPS and clicks per minute.
OpenA thirty-second click test: the interval where a burst technique starts to cost more than it returns. Reports clicks, CPS, clicks per minute and the per-second profile.
OpenThe sixty-second click test: the only interval where clicks per minute is a straight count rather than a projection. What the chronometer reads at the close is your CPM.
OpenThe hundred-second click test, the longest interval on the bench. Measures the rate you can hold well past the point where any burst technique has given out.
OpenA fixed-distance click test: the interval closes on your hundredth press and the elapsed time is the result. Reports seconds, CPS and clicks per minute.
OpenThe Kohi click test: the ten-second CPS test named after the Minecraft server that popularised it, run on a millisecond clock and reported in both CPS and CPM.
OpenA five-second jitter click test. Measures the rate arm-tension clicking actually produces, with the per-click gap chart that shows whether the tremor stayed even.
OpenQuestions
Asked of the chronometer
Because clicks per minute is a projection. Eight clicks in one second projects to 480 CPM, but almost nobody can hold eight per second for sixty. Short intervals measure your peak; long ones measure your endurance, and both are reported in the same unit.
Yes, at the margins. Switch reset time decides how soon the button can register again, and the debounce window adds a floor to the interval between two accepted presses. Polling rate contributes about a millisecond. None of it matters at 300 CPM; all of it matters at 900.
Dragging a finger across the button makes the switch chatter, producing bursts of presses from one motion. The chronometer counts what the browser reports, so those presses do count here — but the reading is a property of the hardware, which is why the top band is labelled instrument-limited rather than by any human quality.
Most competitive servers ban it, and it wears switches out quickly.
Usually, and for different reasons. Trackpads add gesture recognition before a click is confirmed, and touchscreens sample at the display's refresh rate rather than a mouse's polling rate. The chronometer accepts both — it listens for pointer events, not mouse events specifically.
Your best score per interval is written to this browser's local storage and read back the next time you visit. It never leaves the device, no account exists, and the clear-records button removes it immediately.
Clicks per minute is the number of times you press a mouse button in sixty seconds. Most tests measure a shorter interval and scale the count up, which is why the interval always has to be quoted alongside the figure.
The clicks per minute meaning does not change between tools. What changes is the interval behind it: 480 CPM from a one-second burst and 480 CPM held for a full minute are very different results.
Yes. On this site and on every click test, CPM abbreviates clicks per minute.
It is worth knowing that CPM means cost per mille — cost per thousand impressions — in advertising, and counts per minute in radiation measurement. Same three letters, unrelated fields.
On a ten-second interval, 300 to 450 CPM is the normal range, and 384 CPM is the most quoted average. Below 300 is unhurried single clicking; above 600 is trained technique.
Normal falls as the interval lengthens. A rate that is unremarkable across five seconds is well above average held for a hundred.
Records
Click speed test world record
| Clicks per second | Clicks per minute | Clicks in 10 seconds | What produces it |
|---|---|---|---|
| 5 CPS | 300 CPM | 50 | Single clicking, no method |
| 7.5 CPS | 450 CPM | 75 | Top of the average click speed range |
| 10 CPS | 600 CPM | 100 | Practised single or early jitter |
| 12 CPS | 720 CPM | 120 | Jitter clicking at pace |
| 14 CPS | 840 CPM | 140 | Butterfly clicking, held clean |
| 16 CPS | 960 CPM | 160 | The practical human ceiling |
| 20 CPS | 1,200 CPM | 200 | Drag clicking or switch chatter |
| 50 CPS | 3,000 CPM | 500 | An auto-clicker, not a hand |
A clicks per minute test measures the average click rate you hold, the peak click frequency inside the interval, and the per-minute click consistency between them. Run the 10-second interval three times, read the mouse click speed score against the calibration bands, and take the middle result as your click rate.