I Used AI to Write SEO Content for 6 Months. Here’s When It Helped and When It Sabotaged My Rankings

My organic traffic dropped 40% over the next three months. I launched 40 articles in June using AI writing tools, expecting the traffic boost everyone talked about. Instead, Google pushed my AI-assisted articles down in rankings while keeping my older manually-written pieces stable. I spent $800 on content production and got fewer visitors than before. The traffic didn’t recover until I took a completely different approach to how I used AI.

AI became useful after I changed my workflow completely. Instead of generating full articles, I used it to draft outlines based on real competitor analysis, ask questions I hadn’t considered, and rewrite my own sentences for clarity. My traffic started recovering in month four when I used AI for the heavy lifting on data-heavy comparison tables and FAQ sections, while writing my own introduction and conclusion. The key difference: AI became a tool I controlled, not a replacement for my judgment on what readers actually needed.

Through Search Console analysis and testing, I believe my site hit a helpful content update impact. My articles had high word counts but thin on actual experience-based insights — classic AI content pattern. Google explicitly states it targets content that seems designed for search engines rather than humans. My traffic started recovering in month five when I rewrote 12 articles to include specific numbers from my own testing, personal anecdotes about what failed, and genuine conclusions based on my actual results — things AI can’t fabricate because they didn’t happen.