Using the Flesch Formula to Measure Readability in Technical Writing

Every technical writer faces the same quiet test: will the reader actually understand what you wrote? You can spend hours polishing a user manual or a policy document, but if the sentences are too long and the words too heavy, your audience walks away confused. This is where readability formulas step in. The Flesch Formula is one of the oldest and most trusted tools for putting a number on how easy your writing is to read. Let us break down how it works, what its scores mean, and how seriously you should take them.

The Flesch Formula explained 🔗

The Flesch Reading Ease formula was developed by Rudolf Flesch in 1948, an Austrian-born writing consultant who earned a PhD in English from Columbia University and championed the Plain English Movement. He introduced the formula in an article titled “A New Readability Yardstick” in the Journal of Applied Psychology. His goal was simple but powerful: create an objective way to measure how hard or easy a piece of text is to read.

The formula measures readability using just two features of your writing: how long your sentences are, and how long your words are. Here is the actual calculation:

Flesch Reading Ease = 206.835 − (1.015 × ASL) − (84.6 × ASW)

Two variables drive the entire score:

ASL (Average Sentence Length): This is the total number of words divided by the total number of sentences. Longer sentences pull the score down because they are harder to process.

ASW (Average Syllables per Word): This is the total number of syllables divided by the total number of words. Words with more syllables are usually more complex, so they also lower the score.

The number 206.835 is simply a baseline constant. The formula starts from this baseline and then subtracts penalties for long sentences and long words. So a text built from short sentences and simple words keeps most of that baseline value and scores high. A text packed with long sentences and multi-syllable words loses a lot of points and scores low.

A worked example 🔗

Suppose you write a passage with an average sentence length of 15 words and an average of 1.5 syllables per word. Plugging these into the formula gives you: 206.835 − (1.015 × 15) − (84.6 × 1.5), which works out to roughly 64.7. As we will see, that places the text in the “standard” range, readable by most adults.

Now imagine a dense academic paragraph with 25-word sentences and an average of 1.8 syllables per word. The penalties grow much larger, and the score can drop into the difficult range. The formula does not read your meaning. It only counts structural features, but those features turn out to be reliable signals of effort.

The Flesch-Kincaid Grade Level connection 🔗

People often confuse the Flesch Reading Ease score with the Flesch-Kincaid Grade Level, and it helps to keep them separate. The Reading Ease score gives you a number from 0 to 100. The Flesch-Kincaid Grade Level uses the same raw inputs but different weightings to translate readability into a school grade level instead. The grade-level version was later developed for the U.S. Navy to assess training manuals. Both come from Flesch’s work, but they answer slightly different questions: one tells you how easy the text is on a 0-100 scale, the other tells you roughly how many years of education a reader needs.

Flesch’s Reading Ease score 🔗

One feature makes the Flesch Reading Ease score unusual among readability metrics: higher numbers mean easier reading. This is the opposite of most other formulas, where a higher number signals greater difficulty. A score near 100 means the text is very easy to read, while a score near 0 means it is extremely complex and suited only to specialist readers.

How to interpret the score 🔗

Over time, writers and researchers settled on a widely used interpretation scale. Here is how the ranges are commonly read:

90-100 (Very Easy): Easily understood by an average 11-year-old student.

80-89 (Easy): Conversational text, easy for most readers.

70-79 (Fairly Easy): Comfortable reading for teenagers.

60-69 (Standard): Plain English, understood by the average reader around the 8th to 9th grade level.

50-59 (Fairly Difficult): Requires more effort, suited to older students and adults.

30-49 (Difficult): Best understood by university graduates.

0-29 (Very Difficult): Dense academic or professional text, often requiring a specialist background.

According to this interpretation scale, a score below 30 marks a genuinely challenging, academic-level text. For most general writing, a score in the 60-70 band is the sweet spot.

What the score means for content creation 🔗

The score is only useful when you match it to your audience. A children’s storybook should score very high. A legal contract or a research paper will naturally score low, and that is acceptable because the readers expect dense, precise language. The point is not to chase a perfect 100. The point is to write at a level your specific readers can handle comfortably.

For technical writing, this balance matters a great deal. Technical documentation often sits in the 40-60 range, which reflects the reality that the subject carries some inherent complexity. A good target is to meet your readers a little below their actual ability, so the material feels easy rather than taxing. If your user manual scores 35 but your audience is the general public, that is a clear warning that your sentences are too long or your vocabulary is too dense.

For college students and a general adult audience, aiming for the standard band of 60-70 keeps your writing accessible without dumbing it down. The score gives you a quick, objective check before you publish, helping you catch passages that are quietly turning readers away.

When to use readability formulas 🔗

Here is the honest part that many tutorials skip. Readability formulas are useful, but they are not the final word on whether your writing is good. The expert consensus, built over decades of debate in technical communication, is that these formulas work best as guidelines rather than rigid rules.

The expert debate 🔗

A well-known review of the technical writing literature captured a real controversy. Some authorities recommend readability formulas as long as the writer treats them as a limited tool. Others argue more forcefully that formulas can mislead writers, either by creating a false sense of security or by tempting them to write stilted, choppy prose just to hit a target number.

Usability researcher Janice Redish has been one of the most influential voices here. A literature review of her work points out several technical weaknesses of readability formulas: they were originally designed for children’s school books rather than adult technical documents, they ignore differences between individual readers, and they overlook the effects of content, layout, and navigation aids. They emphasise what can be counted at the expense of subtler things that actually drive comprehension.

This is the core limitation. A formula counts syllables and sentence length, but it cannot tell whether your paragraphs are logically organised, whether your headings help readers find information, or whether the reader already knows the subject. A short, simple sentence can still be confusing if it sits in the wrong place.

The technical vocabulary problem 🔗

There is a specific trap for technical writers. Specialised terminology is often made of long, multi-syllable words. Terms like “configuration”, “authentication”, or “infrastructure” push your syllable count up and drag your score down, even when your target readers know exactly what those words mean. A jargon-heavy document can score poorly yet still be perfectly clear to its intended professional audience. The reverse is also true: a text can score well and still be hard to follow if the ideas are poorly arranged.

How to use the formula wisely 🔗

Treat the Flesch score as a screening device, not a grade. Here is a sensible workflow:

Run an initial check: Use the score to spot passages that are unusually dense. Tools like Microsoft Word include built-in readability statistics, so the check costs you almost nothing.

Revise with purpose: If a section scores low, look for long sentences you can split and complex words you can simplify, but only where it does not sacrifice accuracy.

Combine methods: Pair the formula with peer review and, where possible, usability testing with real readers. The guidance from readability experts is that scores sound more precise than they really are, so they should be interpreted in a general way rather than taken as exact measurements.

Respect necessary complexity: Sometimes a topic is genuinely complex, and forcing the text into short sentences would distort the meaning. Know when to stop simplifying.

The bigger picture: plain language 🔗

Readability formulas fit inside a broader movement toward plain language, which is communication the audience can understand the first time they read it. This idea is now embedded in policy in several countries; the U.S. Plain Writing Act of 2010 defines plain writing as text that is clear, concise, well-organised, and appropriate to the intended audience. Importantly, plain language does not mean oversimplifying complex ideas. It means making sophisticated information accessible through careful word choice, logical structure, and good document design.

For technical writers, the lesson is that the Flesch Formula handles one slice of this goal, the word and sentence level. The rest, organisation, formatting, headings, and reader awareness, sits in your hands. The formula tells you where to look; your judgement decides what to do.

What do you think? If a technical manual scores poorly on the Flesch Reading Ease scale purely because of unavoidable specialist terminology, should the writer try to raise the score, or accept the low number as appropriate for the audience? And in your own writing, would you trust a readability score over feedback from a real reader?

References

1. https://readabilityformulas.com/learn-about-the-flesch-reading-ease-formula/
2. https://charactercalculator.com/flesch-reading-ease/
3. https://gorby.app/readability/flesch-reading-ease-formula/
4. https://www.checkreadability.com/formulas/flesch-reading-ease-fre
5. https://journals.sagepub.com/doi/abs/10.2190/U4FF-0L5Q-FPD4-2DCJ
6. https://dl.acm.org/doi/10.1145/344599.344637
7. https://www.ahrq.gov/talkingquality/resources/writing/tip6.html
8. https://digital.gov/resources/an-introduction-to-plain-language