Keyword density between 1–2% is a reasonable baseline, meaning your primary keyword appears once or twice per hundred words. That said, the metric itself is outdated. Google's NLP models understand synonyms, related terms, and context, so stuffing exact-match keywords to hit a magic number will hurt more than help. You're better off writing naturally and ensuring your primary keyword appears in the title tag, H1, first 100 words, and a few subheadings, then letting semantic variants carry the rest. We see diminishing returns above 2% and active penalties if you cross into obvious repetition. A 1,500-word article might use the exact target phrase 15–20 times maximum, but you should also weave in close variants, LSI terms, and question-based phrases that align with user intent. If your keyword is "Ottawa web design," you'd naturally mention "web design services," "website development," "design agencies," and "Ottawa-based designers" without forcing the exact phrase into every paragraph. What matters more than density is topical depth. Cover the full scope of the query—if someone searches "how to optimize images for SEO," they expect file formats, compression tools, alt text best practices, and load-time impact, not just the phrase "optimize images" repeated twenty times. Google rewards comprehensive answers, not keyword spam. At Ottawa SEO, we audit pages by reading them aloud. If a keyword feels forced or repetitive when spoken, it's over-optimized. We also run TF-IDF analysis to see which terms top-ranking competitors use, then ensure our content matches that semantic footprint without mechanical repetition. Tools like Surfer SEO or Clearscope can guide term frequency, but human readability is the final filter. If your content flows naturally and covers the topic thoroughly, keyword density will land in a healthy range without micromanagement.