How to Actually Measure Your Typing Progress (Not Just WPM)

Open TypeBlind's practice mode in a background tab before you keep reading. Nothing to sign up for, nothing to configure - the point of this article is to show you what to actually watch instead of one noisy number, and the fastest way to see it is to run a session while you read.

Why Doesn't My WPM Go Up Even Though I'm Practicing?

Kevin Whitaker ran a real two-month self-experiment on his own typing speed, tracking daily sessions rather than guessing. His honest result: "If you fit a straight line through the data, it slopes upward, showing an improvement of 5wpm over a month and a half." Five words per minute, after weeks of deliberate 20-minutes-a-day practice. He also caught himself being fooled by a metric: isolated drills on his weakest letter pairs looked like they were working, until he checked against a no-drill control period and found matching gains either way.

This isn't a rare story. On Quora, real questions from people at wildly different absolute speeds ask the identical thing: "stuck at 70 WPM," "stuck at below 35 WPM" after two years of practice, "plateaued on typing faster" with no idea how to improve further. The number changes; the stuck feeling doesn't. Another Quora question captures the opposite kind of frustration - inconsistency, not a flat line: "Sometimes I type at 30 WPM but sometime it reduces to 20-23 WPM, it's like my fingers forget the key positions."

Both patterns - flat lines and wild swings - point at the same root problem: raw WPM is a single, noisy number, and typing progress is not one-dimensional.

What's the Difference Between Gross WPM and Net WPM?

Most typing tests quietly report gross WPM by default: (total characters typed ÷ 5) ÷ minutes, full stop. It rewards raw finger speed and doesn't care whether what you typed was right. Net WPM answers a different, more useful question: gross WPM minus an error penalty, so it reflects usable output - the text you'd actually be able to send.

Two typists can post identical gross WPM and have very different net WPM. If your accuracy climbs from 90% to 96% while your raw speed stays flat, your net WPM goes up even though the number most typing tests show you first hasn't moved at all. That's real typing progress a single glance at "WPM" hides completely.

Is a Single Speed Test Even a Reliable Number?

Short sprints flatter the typist: you can hold your absolute peak pace for a few seconds in a way you can't sustain for a real working session, which is why many certification and workplace typing tests run multiple minutes rather than a quick burst. A lot of the "my WPM isn't moving" frustration is partly an artifact of comparing two numbers that were never measuring the same thing - a quick burst on a good day against a longer, harder test on a tired one.

1. Match the test length every time you compare - don't judge a 15-second score against a 60-second one.
2. Run more than one test before drawing a conclusion - a single number is one sample, not a trend.
3. Track the trend line over weeks, not the day-to-day wobble.

Building a fair comparison yourself means re-running the same test length and logging every result by hand. TypeBlind's typing test tracks this automatically across sessions, so the typing progress you see is a trend, not a single noisy sample.

What Does "Consistency" Actually Measure?

Monkeytype is open source, so its scoring isn't a marketing claim - it's inspectable code. Two of its formulas matter here:

Accuracy = 100 * correctChars / (correctChars + incorrectChars)
Consistency = a 0-100 score derived from the coefficient of variation
              of your typing speed measured second-by-second within one test

Consistency is a formalized version of a question you can answer just by watching yourself type: are you moving in smooth, even bursts, or jerking forward in stops and starts? A rising WPM average can hide the fact that you're still lurching through certain words - and a flat WPM average can hide real typing progress in how evenly you're typing everything else. Pravit Ampapathini, who logged 832 of his own Monkeytype sessions over 18 months, put it plainly: "Consistency offers a more stable measure of performance. It's like a barometer for mental clarity."

Why Does One Small Error Cost More Than One Keystroke?

Anna Feit, one of the researchers behind a 136-million-keystroke study of real typing behavior (Dhakal, Feit, Kristensson and Oulasvirta, CHI 2018), put the mechanism directly: "Correct motor execution is the key to fast typing. Slower typists make more errors and require a long time to identify and correct them." Slower typists aren't just slower at pressing keys - they're slower at noticing and fixing what went wrong, and that noticing-and-fixing tax is its own measurable cost, separate from raw finger speed.

This is exactly what error-recovery time tracks: how long each mistake actually costs you, rather than how many you made. A typist who makes rare but expensive mistakes and one who makes frequent, cheap, quickly-fixed ones can post the same error count and have very different real throughput - and very different typing progress underneath an identical accuracy percentage.

Are You Actually Faster Than You Think?

The 136-million-keystroke study (Dhakal, Feit, Kristensson and Oulasvirta, presented at CHI 2018 and run jointly with the University of Cambridge) is the largest dataset of real typing behavior ever collected - 168,000 volunteers, over 200 countries. Lead author Antti Oulasvirta was direct about the ceiling: "The fastest users in our study typed 120 words per minute, which is amazing given that this is a controlled study with randomised phrases."

A separate 2016 study from Vanderbilt University, comparing touch typists against people who never learned formal technique, found something that cuts against a common assumption: touch typists averaged 80 WPM against 72 WPM for everyone else - a real but much smaller gap than the doubling that gets repeated as folklore. Lead researcher Gordon Logan noted the result was smaller than expected: "Our results raise the question of the value of remedial training for nonstandard typists."

What Is "Rollover" and Why Do the Fastest Typists Use It?

Rollover isn't a trick you can memorize - it's a direct consequence of resting all ten fingers on a stable home row, each one an independent lever that doesn't have to wait for its neighbor. Hunt-and-peck technique, even a fast and practiced version of it, relocates the same two to four fingers across the whole keyboard for almost every keystroke, which makes rollover mechanically difficult to reach no matter how much you practice the motion itself.

The same study also found something worth knowing on its own: modern typists mostly make substitution errors (hitting the wrong letter) rather than the added-or-omitted-character errors that were common on typewriters - a small, concrete reminder that typing technique research is still an active, evolving field, not settled folklore from decades ago.

Why Does Accuracy Have to Stabilize Before Speed Does?

As a specific sequence - a common letter pair, a short word typed the same way every time - gets more practiced, the time between those keystrokes shrinks and becomes less variable. That's consistency, the same 0-100 score Monkeytype computes from second-by-second speed variance, and it isn't a vanity stat: it's a direct, measurable signature of real typing progress happening below the level a WPM number can show. Speed gains that show up before that stabilization tends to come at accuracy's expense; the pattern typing coaches describe informally - "accuracy first, speed follows" - has a real mechanism behind it, not just tradition.

How Do You Actually Track These Metrics Without Buying Software?

1. Pick one test length and stick to it for every comparison - don't mix 15-second sprints with 5-minute sessions.
2. Log accuracy alongside speed every time, not just WPM - a rising accuracy percentage at flat WPM is still real typing progress.
3. Note whether the session felt smooth or jerky, even without a formal consistency score - your own sense of "even" versus "stop-start" tracks the same thing Monkeytype's formula measures.
4. Compare week to week, not session to session - single sessions are noisy on their own, for the same reason a short burst test scores higher than a longer one.

What Should You Do When You Hit a Real Plateau?

Whitaker's own conclusion, after his two-month experiment produced only 5 WPM of measurable gain, was refreshingly honest: "I would have probably needed at least an order of magnitude more than the 15 minutes a day and/or the two months I spent on this to make real progress." That's not a discouraging note - it's a realistic one. Typing progress on the metrics that matter (accuracy, consistency, error cost) moves on a longer and quieter timeline than a single WPM number ever will.

How long before I see real typing progress?

There's no single verified number for this - the research in this article describes the mechanism (accuracy and consistency stabilize before raw speed climbs) rather than a fixed timeline, and self-reported accounts vary from weeks to months depending on practice frequency. Watch the metrics in this article rather than a calendar.

Is a WPM plateau normal, or does it mean I'm doing something wrong?

It's normal. Multiple independent accounts in this article - a deliberate two-month self-experiment, recurring Quora questions from typists stuck at completely different speeds, and workplace-forum threads - describe the same flat-line experience regardless of starting skill level. A plateau in the number you're watching is common; check accuracy and consistency before assuming nothing is improving.

Watching the right numbers is half the work - the other half is practice that actually targets what's slowing you down instead of drilling everything equally. TypeBlind builds every session from your own weak keys, five minutes a day.