You took a test, got a number, and now you want to know whether that number is any good. This page is the reference table for that question: what typing speeds look like at each skill level, what employers actually ask for in different jobs, and how speed and accuracy combine into the score you see.
First, a word about where these numbers come from
The benchmarks below are industry conventions — the ranges that turn up in job adverts, typing curricula, and test software defaults. They are not the output of one authoritative study, and no single body certifies them. Anyone quoting a suspiciously precise figure ("the average office worker types 41.4 wpm") is almost certainly repeating a number that lost its source long ago. Treat what follows as calibration, not measurement.
The one figure worth holding onto is the general benchmark that an adult on a full keyboard lands somewhere around 40 wpm — a rough centre of gravity, not a population mean.
How WPM is actually calculated
A "word" in typing measurement is not a word. It is five characters, including spaces and punctuation. This convention exists so that "hippopotamus" and "cat" do not distort the score.
Gross WPM = (total characters typed ÷ 5) ÷ minutes elapsed
Net WPM = Gross WPM − (uncorrected errors ÷ minutes elapsed)
Worked example
You type for exactly 2 minutes and produce 600 characters. Five of your mistakes are still sitting there uncorrected at the end.
- Gross WPM = (600 ÷ 5) ÷ 2 = 60 wpm
- Net WPM = 60 − (5 ÷ 2) = 60 − 2.5 = 57.5 wpm
Note the penalty scales with time. Five errors over a 1-minute test costs you 5 wpm; the same five errors over a 5-minute test costs you 1 wpm. This is why short tests produce more volatile scores.
Some tools use a different penalty — net WPM = gross WPM × accuracy percentage — which produces a slightly different number from the same typing. Neither is wrong, but it means scores from two different sites are not directly comparable. Compare yourself against yourself, on one tool.
Typing speed benchmarks by skill level
| Level | Net WPM | What it looks like |
|---|---|---|
| Beginner | Under 25 | Hunt-and-peck, eyes on the keyboard, frequent backspacing |
| Average | 25–40 | Multiple fingers but not true touch typing; comfortable for email |
| Proficient | 40–60 | Touch typing without looking; keeps pace with normal thought |
| Professional | 60–90 | Typing is no longer the bottleneck in the work |
| Elite | 90–130+ | Competitive typists, court reporters on standard keyboards |
The fastest competitive typists post scores well above 150 wpm on short passages, but those are sprint results on familiar text, not sustainable working speeds.
Typical requirements by profession
These are the ranges that commonly appear in job listings. Employers vary enormously, and many list no requirement at all.
| Role | Commonly quoted | Notes |
|---|---|---|
| Data entry | 50–80 wpm | Often specified in KPH instead (see below); accuracy thresholds are usually stricter than speed |
| Transcription | 60–90 wpm | Audio typing means starting and stopping; raw speed matters less than sustained rhythm |
| Admin / secretarial | 45–65 wpm | Rarely tested in the interview, frequently listed in the advert |
| Customer support / live chat | 50–70 wpm | Often the most speed-sensitive role, because chat queues are measured in response time |
| Journalism / copywriting | 50–70 wpm | Speed helps for interview notes; drafting is limited by thinking, not typing |
| Programming | 40–60 wpm on prose | Effective code-typing speed is lower — see below |
Decoding keystrokes per hour (KPH)
Data entry adverts often quote KPH rather than WPM. The conversion is straightforward, using the same five-character-word rule:
WPM = KPH ÷ 60 ÷ 5
So a listing asking for 10,000 KPH is asking for roughly 33 wpm sustained across an hour. An 8,000 KPH requirement is about 27 wpm. Those numbers look modest next to a 60 wpm sprint score, and that is the point — sustaining a speed for an hour is a different task from hitting it for sixty seconds. If a listing specifies "alphanumeric" KPH it usually implies a numeric keypad component too, which most people are far slower at than letters.
Why accuracy beats raw speed
Errors are more expensive than they look. Consider two typists over a minute:
- Typist A: 80 gross wpm, 12 uncorrected errors → 68 net wpm, and someone still has to proofread
- Typist B: 60 gross wpm, 1 uncorrected error → 59 net wpm, and the output is usable as-is
On the scoreboard A wins. In real work, B's output goes straight into the document while A's needs a correction pass that costs more time than the 9 wpm gap saved. This is why data entry roles set accuracy floors — often 98% or higher — and treat speed as secondary.
There is also a mechanical argument for accuracy-first practice: a mistake costs a backspace plus a retype, so each error burns two to three characters' worth of time. Typing 10% faster while doubling your error rate is usually a net loss.
Why programmers score lower than they expect
Developers often test at 40–60 wpm on prose and then feel slow when writing code. That gap is real and mostly mechanical:
- Symbol density.
{,},(,),<,>,|,&,_,$all require a modifier key. A shifted character is two keypresses and a hand movement, not one. - No word-level muscle memory. You have typed "because" thousands of times as a single motion. You have not typed
useCallback(() => {as a single motion. - Non-dictionary tokens. camelCase, snake_case, and variable names your fingers have never seen defeat the pattern-matching that makes prose typing fast.
- Numbers and the top row. Digits are slower for almost everyone than home-row letters.
None of this means you should not measure it. Testing on a code passage rather than prose gives you a number that reflects the typing you actually do. The typing speed test includes a code mode for exactly this reason.
How much can you realistically improve?
Honest expectations, based on how skill acquisition generally behaves rather than any specific study:
- If you hunt-and-peck, learning touch typing is the single largest available gain. Expect to get slower for one to two weeks before you get faster, because you are replacing a working habit with a worse-but-scalable one.
- If you already touch type at 40–60 wpm, gains of a few wpm over several weeks of short daily practice are typical. Progress is not linear; it arrives in steps with plateaus in between.
- Above about 80 wpm, improvement gets slow and specific. Further gains usually come from eliminating your personal problem digraphs and reducing backspaces, not from general practice.
- Ceilings are real. Keyboard layout, hardware, and hand size all matter. Not everyone reaches 100 wpm, and very few jobs require it.
Fifteen focused minutes a day beats a two-hour session once a week, because what you are building is motor memory, and motor memory consolidates with repetition and rest rather than volume.
Frequently asked questions
What is a good typing speed? For general office and study work, 40–60 wpm with high accuracy is comfortably good. Above 60 wpm you are faster than most people who type for a living need to be.
Is 30 wpm slow? It is on the low side of average but perfectly workable for most jobs. It falls below what many data entry and transcription roles ask for.
How long should a test be? One minute is fine for a quick check. Three to five minutes gives a more stable number, because a single mistimed keystroke has less influence on the result.
Want the number to compare against these tables? Take the typing speed test and check both your net WPM and your accuracy — the second figure matters more than most people assume. If you need a passage of neutral text to practise on, the lorem ipsum generator will produce one, and the word counter will tell you exactly how many characters it contains.