Depth-First began with a single post on August 12th, 2006. One year and 193 posts later, I thought it would be interesting to reflect back on how my expectations about blogging diverged from reality. Many of my observations come from statistics compiled by Google Analytics. An investment in time to install and use analytics software more than pays for itself within a few weeks.
With these thoughts in mind, here are ten things that surprised me:
- The Undiscovered Continent of "Gray Literature." Like television networks, scientific journals serve subscribers with often wildly differing interests from yours. Some content will simply never make it into a print journal, no matter how useful or newsworthy. When you publish a blog, the only OK you need is your own. This means you can tackle subjects that are literally impossible for scientific journals to cover. In fact, your blog can contain valuable content not available anywhere else. After you wrap your brain around this simple concept, the most interesting things can start to happen.
- It takes about six months of constant writing to see signs of being noticed. This is the phase during which I suspect many bloggers throw in the towel. Unless you already enjoy world renown in your field (in which case you're probably not blogging), expect to pay your six month dues.
- Syndication matters. A lot. When I started D-F, I couldn't understand why people seemed so excited about "syndication." But I did notice a steady build-up of traffic on my server logs dedicated solely to accessing my RSS and Atom feeds. I decided to track this traffic using FeedBurner and discovered, to my amazement, that it accounted for about a third of all non-robot traffic on my site.
- Readers take the weekend off. Traffic drops off rather significantly (50-60%) on weekends. Conversely, traffic peaks quite consistently on Wednesdays. Although the former is understandable, the latter continues to puzzle me.
- For niche subjects, AdSense sucks. Chemistry is a subject dominated by niches. Any recent ACS program will show 40+ divisions; chemistry is all about the long tail. In the early days of D-F, I experimented with AdSense and got nada. This made sense (pun intended) because the ads being shown had almost nothing to do with what I was writing about. Adwords can work very well for broadly-appealing subjects. But if you're writing about your area of scientific expertise, which by definition is about as niche as you can get, AdSense is more likely to be an eyesore than a source of income.
- Your most valuable asset: the archive. On a typical day, only about 15% of D-F's pageviews come from a user landing on the homepage itself. The rest come from pages previously found and linked to from other sources, bookmarks, or from search engines. When seen from this perspective, (which is hard to grasp when your archive contains ten articles) it becomes very important to maintain access to the archives and ensure that links can always be followed. On two occasions I've nearly hosed my database, but my backups saved the day. Never assume your archive of articles and comments will be there tomorrow.
- You never know who's reading. The vast majority of readers never post a comment or write an email. I should have expected this, since the vast majority of blogs I read never get a comment or email from me. On the other hand, I've met a few very intelligent and friendly people online as a direct result of one or more D-F articles, which is the ultimate reward.
- Writing begets more writing. Forcing yourself to write regularly about a subject you know about is very therapeutic. More importantly, being forced to back up your arguments in writing on a regular basis makes you examine your own assumptions more carefully. Most important of all, writing regularly brings new ideas into view that you would have missed otherwise.
- Forget about getting Dugg, try to get StumbledUpon instead. Being a regular reader of Digg, I was well aware of its massive audience and the surge in traffic that follows a site being featured on the Digg home page. The effect usually lasts a couple of days. But the traffic surge produced from a listing in StumbleUpon, continues unabated for weeks, and results in permanently higher overall page views.
- Robots everywhere. Google Analytics doesn't record activity from robots and web spiders. But my server log reveals a staggering amount of traffic due to non-human visitors. The software interacting with my site is doing everything from indexing it for Google to trying to post annoying ringtone spam comments. Still, the benefit of automated user agents more than outweighs the inconvenience. Make sure your robots.txt file allows any and all user agents.