My DEV Analytics: Reach vs. Resonance

A developer who posted 31 articles on DEV found that a post with 300 views and zero reactions attracted search traffic as a technical manual, while a post with only 47 views and five comments sparked discussion about LLM agent friction. The author concludes that discovery content and discussion content serve different purposes and advises alternating between the two formats.

My DEV Analytics: Reach vs. Resonance 300 views with zero reactions versus 47 views with five comments—that's the gap between being "discovered" and actually being "heard." I've put up 31 posts on DEV, totaling 1,222 views, and the data reveals a brutal truth about how technical content performs. My most popular post was a deep dive into Cloudflare architecture. It hit 300 views, but the engagement was a ghost town. Looking at the analytics, most of that traffic came from Google and DuckDuckGo. People weren't reading it to engage; they were using it as a manual to solve a specific technical bug. It provided a practical tutorial for a narrow problem, which is great for discovery, but it doesn't spark a conversation. On the flip side, a post with only 47 views became my most discussed piece. The topic? The friction of using LLM agents: the fact that AI writes code faster than a human can responsibly review it. I realized I was trying to make one piece of content do two different jobs. If you're building a presence in the dev community, you have to distinguish between these two formats: Stop expecting every post to go viral and spark a debate. Some posts are meant to be tools, and others are meant to be conversation starters. Now, I deliberately alternate between the two. The "Search Engine" Effect My most popular post was a deep dive into Cloudflare architecture. It hit 300 views, but the engagement was a ghost town. Looking at the analytics, most of that traffic came from Google and DuckDuckGo. People weren't reading it to engage; they were using it as a manual to solve a specific technical bug. It provided a practical tutorial for a narrow problem, which is great for discovery, but it doesn't spark a conversation. The "Conversation" Effect On the flip side, a post with only 47 views became my most discussed piece. The topic? The friction of using LLM agents: the fact that AI writes code faster than a human can responsibly review it. There were no massive code blocks or step-by-step guides. Instead, it focused on a shared pain point in the modern AI workflow. Because it presented a tension rather than just a solution, people actually responded with their own experiences. The Takeaway for AI Content I realized I was trying to make one piece of content do two different jobs. If you're building a presence in the dev community, you have to distinguish between these two formats: Discovery Content: Focuses on a precise technical problem, provides a deep dive, and optimizes for search. This is your "how-to" or "from scratch" guide. Discussion Content: Focuses on a sharp claim or a real-world failure. This is where you talk about prompt engineering frustrations or the reality of deploying LLM agents. Stop expecting every post to go viral and spark a debate. Some posts are meant to be tools, and others are meant to be conversation starters. Now, I deliberately alternate between the two. Next XGBoost: A Deep Dive into Gradient Boosting Logic → /en/threads/3199/ All Replies （3） R I've found that asking a specific question at the end usually bumps those resonance numbers. 0 C Do you think the topic category affects this, or is it mostly just the hook? 0 S My 1k views usually just mean people liked the thumbnail and then bolted. Classic. 0