The Commenting and Blog Traffic experiment started here.
Last week we found that the traffic to this blog spiraled below 20 views/day, and was heading towards 15/day.

This week, I started visiting random blogs and leaving comments. The goal was one comment on one new blog per day, and I’ve attached my commenting history at the end of this post.

The preliminary result is out:

Commenting seems to increase traffic.

Now, this is a very preliminary result, and there are some significant problems with it.

1. Did my comments on other people’s blogs attract more views, or did I write more interesting posts in the past week?
2. Or simply because there are more blog entries for people to browse through this week than last week?
3. Why are the traffic for Saturday and Sunday of week 1 so high?
4. What’s up with the dip on the Sunday of week 2?

Answering #1 is difficult, because I don’t know whether my comments on other blogs attracted traffic directly. I think people who follow the link from my comments to my blog are not registered as being “referred” to me, whereas if you click on a link in Website X to this blog, then Website X is registered as a “referrer”.

However, my popular posts such as “Shood hands with Stephen Lewis” and “How good should we feel when we take the bus?”, both of which were written during week 1, showed similar traffic generation pattern in posts in week 2 (”Tina is Legend“), where each post got around 13 visitors on the first day, and the number dropped dramatically to 1 or 2 in the subsequent days. So no, I didn’t write more interesting posts in week 2.

#2 may be of some concern, because the old posts are still generating traffic. However, in the last weeks of this experiment when I stop commenting, if we see a decrease in traffic, then we know commenting is a traffic booster.

#3 give me a headache. I think it’s the residual traffic from the initial launch hype (record making 142 visits on January 10th), which throws off this experiment a bit. But that should be remedied by more data.

#4 This is the problem with doing this experiment on a small blog like mine. I only have a few dozen visitors, and if they happened decided to take the same day off reading blogs, then this experiment is thrown off.

* * *

Besides these findings, I also found some other things and decided to change the experimental design accordingly:

1. Finding a good blog to comment on is HARD!!! It takes me an average of 20 minutes to finally find something interesting and commentable. As a result, I might drop the 4-comments/day week, depending on the result of the next week.
2. To be consistent, I try to write one blog entry a day. If I can’t do that, I try to find old posts to hopefully attract readers. If I skip a day or write many entries in one day, then of course the traffic will be different.
3. The data for Fridays, where this experiment transit from one phase to the next, will be discarded because it’s unclear which week the traffic should belong to, and listing the posts that I commented on below may also increase traffic, which is not part of this experiment.

* * *

Here are the places I left comments:

• Day 1: Ten Reasons I Don’t Buy Into The Global Warming Hype
• Day 2: IMDb message board I Am Legend
• Day 3: Scientists restore walking after spinal cord injury
• Rules of Thumb for Writing
• Day 4: Do spankings “work?” Unintended consequences
• Day 5: 暴龍與暴龍的故事。
• Day 6: The Scientific Vacuity of Intelligent Design Creationism

Your response: