Sentence Counter

Sentence count is most useful as a denominator. On its own it tells you little, but divided into word count it gives average sentence length, and divided into paragraph count it gives sentences per paragraph — the two ratios that actually predict how a page reads. It is also the unit that matters for translation quotes, subtitle segmentation, and any workflow that processes text sentence by sentence.

Detecting sentence boundaries is harder than it looks, because a full stop does not reliably end a sentence. Abbreviations like Dr., Inc., e.g., and approx. all contain terminal punctuation mid-sentence, as do decimals, version numbers, file names, and ellipses. This tool applies the usual heuristics — a following capital letter and whitespace, with common abbreviations excluded — which is accurate for ordinary prose and less so for text dense with citations or technical notation.

That means the count is an estimate on some inputs, and it is worth knowing which way it errs. Text heavy with abbreviations will typically over-count, splitting one sentence into several. Text using unusual terminators, or lists written without full stops, will under-count. For ordinary editorial writing the error is small enough to ignore; for legal or scientific text with heavy citation, spot-check before relying on a derived average.