Skip to content
Redmoon Calculators
Text analysis

Sentence Complexity Analyzer

Free sentence complexity analyzer. Measures average clause depth, subordinate clause count, simple / compound / complex sentence classification, passive voice rate, and surfaces the longest parse-like structures.

Built and maintained by Paul Clark, Redmoon Software

When to use this

Use when your text feels hard to read but you can't pinpoint why. A high clause depth combined with high passive count is the classic recipe for "dense" prose.

How it compares

Where readability formulas give a single grade, this tool tells you what *kind* of complexity drives it — subordination, coordination, or length.

0 chars

How it works

Each sentence is tokenized and counted for subordinating connectives ("because", "although"), coordinating connectives ("and", "but"), and passive constructions.

Clause depth is roughly 1 + subordinate clause count. Sentences with no connectives are "simple"; with only coordination they're "compound"; with any subordination they're "complex".

Splits surfaced are sentences over 35 words or with three or more subordinate clauses — typical candidates for trimming.

FAQs

What does clause depth measure?

Roughly how many clauses each sentence contains. A sentence with two subordinate clauses has depth 3.

What grade level does this map to?

Not directly — use a readability formula for that. This tool is about syntactic complexity, which can be high in clear writing or low in unclear writing.

Why does it flag passive voice?

Heavy passive voice and deep subordination both increase parsing load. The split suggestions surface sentences that combine both.

Worked example

Input

Although the meeting ran long, we covered every agenda item. Decisions were made. The team left energized.

Output

Avg clause depth: 1.67 — Simple. Mix: 2 simple, 0 compound, 1 complex.

One subordinate clause ("Although the meeting ran long") gives sentence 1 a depth of 2; the other two are depth 1. The blend reads as varied but accessible.

Common pitfalls

  • Heuristic detection — does not parse syntax fully.
  • Dialogue confuses the simple/compound classifier (heavy "and" usage).
  • Long sentences without subordination are not flagged — only deep nesting is.

Structure, not length

Complexity here is about how a sentence is built rather than how long it is. The measure looks at subordination and embedding — clauses inside clauses — because that is what actually taxes a reader, and it is invisible to the word-count measures every readability formula relies on.

The difference is easy to feel. "The report, which the committee, whose chair had resigned, had commissioned, was late" is 16 words and nearly unreadable, because two relative clauses are nested inside the subject and the main verb arrives last. A 40-word sentence of coordinated short clauses is longer and far easier.

Reading it alongside a readability score

This is the measure to consult when a readability score says a document is fine and readers say it is not. Formulas will pass nested prose that stays within their sentence-length band, and this is the check that catches it.

The edit is usually not to shorten but to unnest: pull the embedded clause out into its own sentence, or move it after the main verb so the reader gets the subject and predicate before the qualification. Sentence count goes up, average length goes down, and comprehension improves without any word being removed.

Read more about this

Related tools

Send feedback

We read every message. Tell us what could be better or what you love.