Headline Analyzer

Headlines are the only thing a searcher sees before deciding to click. Per Backlinko's 11M-SERP study, headlines with specific numbers, emotional words, and power verbs consistently outperform generic ones on click-through rate, sometimes by 2-3x. The CoSchedule headline score (the original public one, 2014-2017) is the most cited model, and the team reverse-engineered it well enough that several free tools now approximate it. The score blends word balance (common vs. uncommon vs. emotional), headline type detection (how-to, list, question, numbered, generic), power-word presence, sentiment polarity, and character count into a 0-100 number with color bands: green 70+, amber 50-70, red under 50.Most low-scoring headlines fail for one of three reasons. They are too long (over 70 characters and the SERP truncates the value prop). They lean too heavily on common words like 'guide', 'tips', 'best' that every other post on the page also uses. Or they lack any specific differentiator: no number, no credential, no concrete benefit. Per Buffer's analysis of 100M headlines, the average top-performing post headline contains 1.81 power words, 1 specific number, and an odd or surprising framing. The opposite profile (generic noun phrase, no power verb, no number) consistently scores in the red band.Our analyzer gives you the same signals CoSchedule charges for: a 0-100 score, word balance across common, uncommon, and emotional categories, sentiment polarity from the public AFINN-165 dictionary, power-word count from a curated lexicon of ~200 words, and reading time. Everything computes in your browser, so the headline you are testing never leaves the device. We document exactly how we approximate the CoSchedule algorithm (it is trade-secret) and tell you where the model breaks.