Keyword density measures how often a specific keyword or phrase appears in your content relative to the total word count. The formula is straightforward: count how many times your target keyword appears, divide by your total word count, then multiply by 100 for the percentage. If "Ottawa web design" appears 8 times in an 800-word article, that's 1% density. Most SEO professionals don't calculate this manually anymore. Tools like Yoast SEO, Surfer SEO, Clearscope, or even free options like SEOReviewTools.com show keyword density automatically as you write. In Yoast, it's called "keyphrase density" and flags you if you go above 2.5%. Screaming Frog can also extract keyword frequency across entire sites during audits. The real question isn't how to find it, but what to do with the number. Google hasn't cared about keyword density as a ranking signal since the mid-2000s. The old "2-3% is ideal" rule is outdated SEO folklore. We've ranked pages at 0.8% and others at 3.2% for competitive terms in Toronto and Ottawa markets. What matters more is topical relevance, search intent match, and natural language use. At Ottawa SEO, we track keyword density during content audits mostly to catch over-optimization. If a page sits at 4-5% for a single keyword, it often reads awkwardly and risks looking manipulative to Google. We aim for 1-2% on primary keywords, then focus on semantic variations and related terms instead of repeating the exact phrase. Watch for these pitfalls: counting only exact matches misses how Google understands synonyms and context. A page about "SEO services Ottawa" should naturally include "search optimization," "local SEO," and "Ottawa digital marketing" without fixating on one phrase. Also, density changes with content length. A 200-word product description at 2% hits the keyword 4 times, which might feel forced, while a 2000-word guide at 1.5% uses it 30 times across logical sections. Use density as a diagnostic check, not a target. If your content sounds robotic or you're forcing keyword placement, you're likely over-optimized regardless of the percentage.