DeviceBenchFree online CPS test
CPS test that starts on your first click
The clock here starts on your first press rather than on a Start button, so none of your reaction time is charged to your click rate — pick 1, 5, 10 or 60 seconds and go. The finished score is your clicks divided by the whole window rather than by the time to your last one, with the busiest single second printed beside it because a run that opens at 11 and ends at 4 is a different hand from one that holds 7 throughout. What it counts is the press events your browser received — a contact your mouse never reported cannot reach a web page, so the figure covers the whole chain from finger to tab rather than the switch alone.
- 100% free
- No signup
- 4 run lengths
- Kohi 10-second window
- Best kept in-browser
The Kohi length, and the one scores are compared in.
- Under desktop paceunder 3
- Ordinary clicking3+
- Above average5+
- Fast7+
- Very fast9+
- Past one finger12+
On a phone every tap counts, but the score sits lower than the same person on a mouse: a finger has to travel clear of the glass and back before the panel will report a second contact, while a mouse button only has to rise past the switch.
How to measure your clicks per second
Three decisions, and only the first one is yours to think about.
Pick the window before you start
Ten seconds is the length nearly every score you have seen quoted was measured over, so pick that one if you want a number you can compare with somebody else's. One second measures how fast you can start rather than how fast you click, five is a sprint you can hold all the way, and sixty tells you something completely different: whether your hand lasts.
Click the pad — the first press is the starting gun
There is no separate Start button, because pressing one and then reaching for the pad puts your reaction time inside your click rate and inflates nobody's score in a useful direction. The counter on the pad moves with every press, the border turns amber while the clock runs, and the remaining seconds count down underneath it.
Read the average, then the busiest second, then the chart
The large figure is clicks divided by the whole window, which is the number worth quoting. The busiest second underneath it is the most presses that landed in any rolling second of the run, and the gap between the two tells you how evenly you clicked: a 7.0 average with a 12 in it is a burst and a stall, while a 7.0 average with an 8 in it is a hand that held its pace.
Technical specifications
| Run lengths | 1, 5, 10 and 60 seconds; the 10-second run is the Kohi length scores are usually quoted in |
|---|---|
| What starts the clock | Your first press on the pad — there is no separate Start button to react to |
| Score basis | Clicks divided by the full window, so stopping at 8 s of a 10 s run costs you the last two seconds |
| Busiest second | The most presses landing in any rolling 1,000 ms slice of the run |
| Presses counted | pointerdown with button 0 — the primary mouse button, or one finger on a touchscreen |
| Rank bands | Six, from under 3 CPS to 12 and above, set on the 10-second window |
| Ceiling per run | 50,000 presses, which is 833 per second across a minute and well past any hand |
| What is kept | One best score per window length in this browser's local storage, cleared by a button on the page |
Frequently asked questions
What counts as a good CPS score?
Between 5 and 7 clicks per second over ten seconds is above average for someone tapping normally with one finger, and most people who have never tried land between 3 and 5. Below 3 is slower than ordinary desktop use. Past 9 the number is unusual for a single finger and past 12 it is essentially out of reach of one, which is why this page shows the band alongside the figure rather than a single pass mark — a score means nothing without knowing which window it was measured over.
Why does the timer start on my first click instead of a Start button?
Because a Start button measures your reaction time and calls it click speed. The gap between pressing Start and landing your first click on the pad is somewhere between 200 and 300 ms for most people, which is 2 to 3% of a ten-second run handed away before you begin. Starting the clock on the first press removes that entirely, and it is why scores here sometimes read slightly higher than on a site that makes you arm the test first.
Why is my one-second score so much higher than my ten-second score?
Because a one-second run is almost entirely the fastest part of your burst and a ten-second run is the average of your burst and everything after it. Most hands can sustain a rate for well under two seconds before it starts to fall, so the same person will typically read two to three clicks per second higher over one second than over ten. The two figures are not comparable, and quoting a one-second number as a CPS score is the most common way people end up with a figure nobody else can reproduce.
What is the Kohi click test?
It is the ten-second click test that ran on Kohi, a Minecraft server whose player base cared about click rate, and its length is the reason ten seconds became the standard. Nothing about the measurement was special — it counted presses over ten seconds like this page does — but because everyone was quoting the same window, the numbers could be compared, and that convention outlived the server itself.
Does my mouse's polling rate change the score?
No. Polling rate governs how often the mouse reports its position, and a button press is a separate state change that travels on the next report. At 125 Hz that report is at most 8 ms away and at 1000 Hz at most 1 ms, so a slow polling rate delays a press slightly but never drops one. If a press is genuinely missing from the count it was the switch, not the report rate.
Does clicking outside the pad count?
No, only presses that land inside the marked pad are counted. That is deliberate: the counter has to survive somebody clicking a preset button, scrolling, or accidentally hitting a link mid-run without those turning into score. Presses of the right or middle button inside the pad are also left out, so a stray one does not quietly add to the figure.
Where is my best score kept?
In this browser, in local storage, as one number per window length — nothing is sent anywhere and there is no account to attach it to. That also means it does not follow you to another browser, another device or a private window, and the Clear kept bests button on the page erases all four entries at once.
About clicks per second and how the number is built
Clicks per second is a rate, and a rate needs a window before it means anything. The same hand measured over one second, ten seconds and a minute produces three numbers that can differ by a factor of three, because a burst is not a pace and a pace is not stamina. Ten seconds became the convention almost by accident — it was the length used by the click test on a Minecraft server whose players compared scores constantly — but the convention is the useful part, and it is why this page prints the window next to every figure it shows. A score quoted without its window is not a measurement, it is a boast.
The arithmetic underneath is where click tests quietly differ. Divide the clicks by the time to the last click and someone who gives up at eight seconds of a ten-second run gets the score of someone who kept going; divide by the full window, as this page does, and the last two seconds cost what they should. The busiest-second figure exists for the opposite reason: an average hides the shape of a run, and a single rolling second is the cleanest way to show the fastest thing your hand actually did. If the whole run reads low but individual presses feel late rather than slow, the problem is latency rather than rate, and the reaction time test separates the two by timing a single response instead of a rate.
A browser sees a press, not a switch, and everything a click test can tell you sits on that side of the line. If a press never registers, this page cannot say whether the contact failed, the cable dropped it or the operating system swallowed it — it can only report what arrived. Two faults do show up here as suspicious numbers, though: a count that runs ahead of your fingers means the switch is closing twice on one press, which the mouse test confirms by button, and a rate that sits far above the top band comes from one of the three named clicking methods rather than from tapping, each of which is measured separately on the jitter, butterfly and drag click test. If you came here to count something in the room rather than to score your hand, the click counter has no timer at all.
Where your best score lives
Every number on this page is worked out by JavaScript running in the tab you are reading it in. Nothing you type, paste or open is uploaded, logged or kept, which is also why the tools carry on working after you disconnect from the network.
The one thing this page does keep is your best figure for each window length, written to this browser’s own local storage so it is still there tomorrow. It never leaves the machine, it is not tied to anything that identifies you, and the Clear kept bests button removes all four entries in one press.