Readability Scores for ESL Writers: What to Actually Target

If you write English as a second language and want a concrete number to aim for, here it is: target a Flesch Reading Ease score of 45–60 for IELTS and TOEFL essays, 30–45 for journal abstracts, and 60–70 for business email. Higher scores mean easier reading. And the direction of the problem is almost never what ESL writers expect. Most worry about sounding too simple. In practice, the drafts I see — including my own early ones — score too low: somewhere between 0 and 25, deep in “very difficult” territory, because the writer stacked long sentences and Latin-derived vocabulary to sound academic.

The fix is mechanical, not mysterious: keep average sentence length under 20 words, replace noun-chains like “the implementation of the enhancement of” with plain verbs, and give each sentence one idea. Below are three real rewrites with exact score deltas — an essay sentence going from 0 to 51, an abstract sentence from 0 to 39, an email from 24 to 76 — plus the reasoning behind each target band. You can measure any text in seconds with a free readability checker.

What the Flesch bands actually mean

The Flesch Reading Ease formula looks at exactly two things: how long your sentences run, and how many syllables your words carry. That is all. It does not know whether your grammar is correct, your logic holds, or your facts are true. A grammatically broken text can score 90. A flawless one can score 12.

Two things follow from the formula being this dumb. First, it is honest — the score responds instantly and predictably to concrete edits, so it works as a feedback loop while you revise. Second, a low score is not automatically bad writing. A dense methods section can legitimately sit at 35. The mistake is sitting at 15 when the content never required it.

Why ESL writers score too low: the imported-register trap

Many academic cultures treat sentence length and abstraction as markers of seriousness — Ukrainian and Russian certainly do, and colleagues tell me the same about German, Arabic, and Chinese academic prose. Ukrainian academic style taught me that a serious sentence is a long sentence built from nouns. I still draft tricky paragraphs in Ukrainian first, and the first English version reliably comes out ten to fifteen Flesch points harder than it needs to be, because I imported the register along with the ideas. That imported register is a big part of why your English sounds translated even when every word is technically correct.

Three habits do most of the damage:

1. Nominalization. Turning verbs into nouns: investigate becomes conduct an investigation of. Every nominalization adds an “of” chain and a four-syllable word where a one-syllable verb lived.
2. Passive stacking. “It has been demonstrated that improvements were observed” spends eight words before saying anything. If you are not sure how much of your draft leans on this, a passive voice detector will show you in one pass.
3. Thesaurus inflation. Use becomes utilize, help becomes facilitate, show becomes elucidate. The longer synonym carries zero additional meaning and one to three additional syllables.

There is a psychology underneath: rare words feel like proof of proficiency, insurance against being read as basic. But the people who grade, review, and reply to your writing read at volume. The text that never needs rereading is the one that wins.

What to target, context by context

IELTS and TOEFL essays: Flesch 45–60

The scoring rubrics reward precision, coherence, and controlled variety — not rarity. A 45–60 essay has some subordinate clauses and some less common vocabulary used correctly, and the examiner never rereads a sentence. Above 75, your sentences are probably uniform and choppy, which costs you on structural variety. Below 40, you are almost certainly forcing vocabulary you do not fully control — and forced vocabulary is where collocation errors come from, the exact errors that cap band scores.

Journal abstracts: Flesch 30–45

The abstract is the most-read part of any paper: reviewers skim it to decide their mood, and databases surface it to people far outside your subfield. Published abstracts routinely score 15–25. You do not win by matching them. Your technical terms are non-negotiable syllables — heteroscedasticity costs what it costs — so everything around the terms must be cheap: short verbs, short connectors, sentences under 22 words. An abstract at 38 next to competitors at 18 reads like relief.

Business email: Flesch 60–70 or higher

Your reader is on a phone between meetings and decides in seconds whether to act, defer, or ignore. Formality lives in tone, not syllable count: “could you send” is perfectly polite; “kindly provide the aforementioned” is a fax from 1995. Every support and partner email I send for Diglot gets the same treatment — one idea per sentence, the actual request within the first two lines.

Three rewrites with exact score deltas

I computed these with the standard Flesch formula on the exact sentences shown. Single sentences are noisy — full documents average out — but the deltas hold at any length.

Essay sentence.

Before (scores below zero — most checkers display 0): “The utilization of modern technology in educational institutions has facilitated the enhancement of teaching methodologies. Consequently, the acquisition of knowledge by students has undergone significant transformation.”

After (Flesch 51): “Modern technology has changed how universities teach, and students now learn in ways that did not exist a decade ago.”

Twenty-six words became twenty. Syllables per word dropped from 2.6 to 1.6. Nothing was lost except the fog — and 51 lands exactly in the essay band.

Abstract sentence.

Before (raw score −15; displayed as 0): “The present study was conducted with the objective of investigating the potential correlation between prolonged exposure to second-language academic environments and the subsequent development of advanced written proficiency.”

After (Flesch 39): “We studied whether long exposure to academic English improves writing skill in a second language.”

Twenty-nine words became fifteen. “Was conducted with the objective of investigating” — seven words, fourteen syllables — became “studied.” The claim is identical and now sits in the abstract band.

Email request.

Before (Flesch 24): “Pursuant to our previous correspondence, I would like to kindly request that you provide the aforementioned documents at your earliest convenience.”

After (Flesch 76): “As discussed, could you send the documents by Friday?”

A 52-point jump, and the second version is more likely to get the documents — because it names a deadline instead of hiding behind “earliest convenience.”

Notice what changed in all three: not the information, and not the politeness. Only the number of clause layers between the reader and the point.

A revision workflow that lands in the band

1. Draft without checking. Think in whichever language your thinking flows; do not optimize syllables while forming ideas.
2. Run the score once. Note the number and, more importantly, your five longest sentences.
3. Fix the longest sentences first. Sentence length is the biggest lever in the formula, and a 45-word sentence usually hides two or three separate ideas anyway.
4. Hunt nominalizations. Search your draft for -tion of, -ment of, and the … of the — each hit is usually a buried verb.
5. Recheck and stop when you are in band. Do not chase 80 in an abstract. In-band is the goal, not the maximum.

This is also where a purpose-built ESL writing tool earns its place over a generic checker: seeing readability alongside grammar and word-choice feedback means one revision pass instead of three, and you learn which of your habits — usually imported from your first language — drag the score down.

What the score cannot tell you

Flesch Reading Ease is a smoke detector, not a fire inspector. It cannot tell you that utilize should be use — both are grammatical, and the formula only counts syllables after the fact. It cannot catch a wrong collocation, a missing article, or an argument that contradicts itself in paragraph three. A text at 55 can still read as translated, stiff, or wrong. Treat the score as one instrument on the dashboard: fast, objective, and blind to everything it does not measure.

But for ESL writers specifically, it is the instrument that points the right direction. Your instinct after years of academic training in another language is to add — syllables, clauses, formality. The score gives you permission, in numbers, to subtract.

If you want to see where your own writing sits, paste a paragraph into the free readability checker at diglot.ai — it takes ten seconds and shows the band immediately. And if you would rather see the score live while you write, next to grammar and phrasing suggestions tuned for non-native writers, that is exactly what the Diglot editor does.