← All free tools

Free tool

A readability checker for interface copy, not blog posts.

Every other readability checker was built for bloggers and SEO writers. This one is built for error messages, form labels, onboarding text and empty states -- the copy your interface actually shows people, usually at the worst possible moment. Paste your copy below and get an instant score, no signup required.

Check your copyHow it works ↓

Paste your copy

An error message from a real product is loaded by default. Replace it with your own error message, form label, empty state or onboarding copy.

Verdict

Well above the target reading age. UK public-sector guidance aims for reading age 9 (Flesch 60-70) -- this copy scores 16.2, reading age 20.

Flesch Reading Ease

16.2

0-100, higher is easier

Flesch-Kincaid Grade

15.3

US school grade

UK reading age (approx.)

20

years -- approximate

Words

86

Sentences

5

Avg sentence length

17.2 words

Passive-voice sentences

0

Long words (3+ syllables)

25

Specific problems found

Common UX-copy failures

  • "please try again later" -- this tells someone nothing about what went wrong or what to do next.
  • "an error has occurred" -- this tells someone nothing about what went wrong or what to do next.
  • "please try again" -- this tells someone nothing about what went wrong or what to do next.
  • "contact support" -- this tells someone nothing about what went wrong or what to do next.

Sentences over 20 words (2)

  • "In order to proceed with the utilisation of this functionality, users must first authenticate their credentials in accordance with the aforementioned security protocol." -- 23 words. Consider splitting it in two.
  • "If you continue to experience difficulties, please do not hesitate to commence the troubleshooting procedure outlined in our comprehensive documentation, or alternatively, initiate contact with our customer support representatives at your earliest convenience." -- 33 words. Consider splitting it in two.

Jargon and complex words (13)

  • "at your earliest convenience" → try "as soon as you can" instead.
  • "in order to" → try "to" instead.
  • "in accordance with" → try "following" instead.
  • "please do not hesitate to" → try "feel free to" instead.
  • "proceed" → try "go ahead" instead.
  • "utilisation" → try "use" instead.
  • "functionality" → try "feature" instead.
  • "authenticate" → try "sign in" instead.
  • "aforementioned" → try "this" instead.
  • "termination" → try "ending" instead.
  • "commence" → try "start" instead.
  • "comprehensive" → try "complete" instead.
  • "initiate" → try "start" instead.
Want a second opinion

Get a human read on your copy, not just a score.

Tell us what you are writing and we will send you a short, honest note on how it reads -- no report generator, an actual person looks at it.

By submitting, you agree to receive course updates and our occasional newsletter. Unsubscribe any time. Privacy policy.

How it works

What the scores mean, and where they come from.

Flesch Reading Ease

A score from 0 to 100 based on average sentence length and average syllables per word. Higher is easier to read. It is calculated as 206.835 - 1.015 x (words / sentences) - 84.6 x (syllables / words).

Flesch-Kincaid Grade Level

The same underlying measurements, expressed as a US school grade: 0.39 x (words / sentences) + 11.8 x (syllables / words) - 15.59. A score of 8 means the text is written at roughly the level of an average 13-14 year old.

UK reading age

We convert the US grade level to an approximate UK reading age (grade + 5 years, which is a commonly used rule of thumb, not a precise conversion). We label this clearly as approximate because it is a translation of a translation. Use the Flesch score as your primary number and the reading age as a rough, more intuitive way to talk about it with colleagues.

The target

UK government content design guidance aims for a reading age of about 9, roughly Flesch 60-70. That target exists for public services used by everyone; we think it is at least as relevant to a product interface, where people are often stressed, in a hurry, or reading on a small screen while doing something else.

Be honest with yourself

These formulas were built for prose, not interfaces.

Flesch and Flesch-Kincaid were developed in the 1940s-70s for continuous text -- training manuals, textbooks, newspaper articles. They measure sentence length and syllable count, and nothing else. A three-word button label and a two-sentence error message will both often score as "very easy" even when one of them is confusing, because the formulas cannot see clarity, tone, ambiguity or whether the reader knows what to do next.

Treat every number on this page as a signal, not a verdict. A low Flesch score is a reliable sign of a problem. A high one is not proof the copy is good -- it just means it is not failing on sentence length and word complexity. The syllable counter itself is an approximation (vowel-group counting, not a dictionary), so expect it to be slightly wrong on words like "wanted" or unusual names. Read your copy aloud, or better, watch someone else read it, before you trust it.

Questions

Common questions

What is a good Flesch Reading Ease score for interface copy?

UK public-sector design guidance aims for a reading age of about 9, which is roughly a Flesch Reading Ease score of 60-70. Interface copy -- error messages, labels, confirmations -- should usually aim at the easy end of that range or easier, because people read it under stress, in a hurry, or on a small screen.

Why is this different from a normal readability checker?

Most readability tools (Hemingway, Yoast, and similar) are built and tuned for long-form writing: blog posts, articles, marketing copy. Interface copy is a different job -- it is short, read under time pressure, and often the only thing standing between someone and giving up. This tool applies the same underlying formulas but flags the specific failure patterns that show up in error messages, form validation and onboarding text, not essay structure.

How accurate is the syllable counter?

It uses vowel-group counting with rules for silent trailing e and -le endings, which covers most English words correctly. It is not a dictionary lookup, so it will get some words wrong -- most commonly words where an -ed ending adds its own syllable (like "wanted" or "needed"), and unusual proper nouns. Treat the scores as a good approximation, not an exact measurement.

Does this tool store or send my text anywhere?

No. Everything runs in your browser -- nothing you paste into the text box is sent to a server. The only place data leaves your browser is if you choose to use the form below to ask for a human review of your copy.

Writing copy across a whole product? See in-house UX training for teams.

Learning UX design yourself? Watch the free UX/UI masterclass.