What the Gunning Fog Index measures
Gunning Fog measures two things: sentence length, like the other formulas, and the share of “complex” words in the text.
Gunning Fog Index = 0.4 × [(total words ÷ total sentences) + 100 × (complex words ÷ total words)]
Break it down:
- Words ÷ sentences is the average sentence length, the same as in the other two formulas.
- Complex words ÷ words is the share of the text made up of complex words, scaled by 100.
- 0.4 brings the combined total down to a grade-level-like number.
The output reads like a grade level, similar to Flesch-Kincaid. A Fog Index of 12 roughly means a high school senior should follow the text without much strain.
Where the Gunning Fog Index comes from
Robert Gunning developed the formula in 1952, publishing it in a book called The Technique of Clear Writing. Gunning worked as a writing consultant for newspapers and business publications, and the formula grew out of that work.
His larger argument wasn’t really about the math. Gunning believed most professional writing was needlessly dense, weighed down by jargon and long sentences that added nothing but length. The Fog Index was a way to put a number on that density. It let an editor point at a passage and say “this is why it’s hard to read,” instead of relying on a vague feeling.
That’s still the formula’s main use today. It’s less a precision instrument and more a flag. A high score means a passage is worth a second look, even if the number itself shouldn’t be treated as exact.
What counts as a “complex” word
This is where Gunning Fog earns its name. A complex word is any word with three or more syllables, with a few carve-outs:
- Proper nouns don’t count, even long ones. “California” doesn’t make a sentence harder to read.
- Familiar jargon in its field doesn’t count, though “familiar” is left for the writer to judge.
- Compound words built from simple parts don’t count.
- Words that only reach three syllables because of a common suffix, like “-es,” “-ed,” or “-ing,” don’t count either. “Creating” isn’t complex just because the suffix adds a syllable.
The idea is sound. Syllable count alone is a poor proxy for difficulty, so the formula tries to filter out words that are long but not actually hard to read.
Where the complex-word definition breaks down
The carve-outs need judgment calls that a plain word count can’t make on its own.
“Familiar jargon” depends entirely on the reader. “Torque” is familiar to a mechanic and opaque to almost everyone else. A formula applied blindly treats both readers the same way, which defeats the point of the exception.
Automated tools that implement Gunning Fog have to guess at proper nouns, usually by checking for capital letters mid-sentence. That catches most cases. It also means a capitalized common word, or a lowercase brand name, can slip through and get miscounted.
The suffix rule has a similar problem. Automated syllable counting is itself an estimate. Deciding whether a word “only” reaches three syllables because of a suffix means comparing two estimates against each other, and small errors compound.
None of this makes the formula useless. It means the number is a reasonable estimate, built on judgment calls that don’t always transfer cleanly to software.
Three passages scored by all three formulas
Here’s how Flesch Reading Ease, Flesch-Kincaid Grade Level, and Gunning Fog score the same three short passages.
| Passage | Flesch Reading Ease | Flesch-Kincaid Grade | Gunning Fog |
|---|---|---|---|
| “The dog ran across the yard. It found an old ball. The children laughed and cheered.” | 106.2 | −0.2 | 2.1 |
| “Researchers observed a modest increase in output after the new schedule began. Several employees reported feeling less rushed during the morning shift. Managers plan to review the results again next quarter.” | 49.0 | 9.0 | 8.0 |
| “The implementation of the aforementioned methodology necessitated a comprehensive reevaluation of preexisting organizational protocols, particularly regarding interdepartmental communication strategies that had previously demonstrated suboptimal effectiveness.” | −106.2 | 34.3 | 35.6 |
The three formulas mostly agree on direction. Easy stays easy and hard stays hard, across all three. They don’t agree on scale, though. The third passage, one dense sentence stuffed with jargon, pushes Flesch Reading Ease below zero and Gunning Fog above 35. Neither number means much at that point. Both formulas are really just telling you the same thing: this sentence sits far outside the range they were built for.
How to read a Gunning Fog score
Treat Gunning Fog the way you’d treat Flesch-Kincaid: as a grade-level estimate for typical prose, not a precise measurement for every possible sentence. It’s most useful in the middle of the scale, where most everyday writing actually lives.
When the number gets extreme in either direction, that’s a signal on its own. A Fog Index near 20 doesn’t mean “very smart writing.” It usually means long sentences stuffed with unnecessary complex words, and that’s worth fixing regardless of what any formula says about it.
Our piece on how readability formulas fail goes further into where all three scores run out of usefulness. It also explains why pairing a score with sentence-level detail catches problems a single number can’t. If you want the grade-level framing explained on its own, read our guide to Flesch-Kincaid Grade Level next.
Run your own text through the readability explainer to see all three scores side by side, scored the same way they’re compared here. It also flags which specific sentences are dragging your score down, so you can fix the actual problem instead of guessing at it.