Blog layout decision making process

You may have noticed that I used to propose two different layouts for the main page: one with summaries and one with full articles. The goal wasn't to let it work for ever, but to chose which layout was the best. (click here to see the layout with full articles)


To decide which one I would keep I have 3 different tools:

• a pool
• I can let the web surfers which layout they want
• I can use a metric to measure the efficiency of both layouts

Here is how I collected data and used it to decide which layout to keep.

The pool was supposed to let me know which layout you say you prefer. I said "supposed" because I didn't have enough answer to consider it as useful and I don't mind since I don't care about what people think they need. It appears that most web surfer aren't really aware of what's best for them.

Then, I let the web surfers decide which layout they like to have. Then I counted the number of clicks and this way I know which layout web surfers wanted to use. This tells me which layout was the most convenient according to the visitors. Here again, I don't have enough data to draw any conclusion. I should have had more visitors or let the tests run longer. But my statistic tool was slowing the blog and it killed my rank in Google. Therefore I decided to rely on the third tool.

Finally, I set up a metric to measure the efficiency of both layouts. My goal is to have as many readers as possible so I want to increase the ratio readers vs incoming visitors. This ratio is the efficiency of my layout. The interest of my articles aren't to be taken into consideration since it the same articles for both layouts.
To implement this ratio, the first step was to define what a visitor was. I defined a reader as a visitor who stayed at least 1 minute on a page on which a full article was written. I enter this definition in a segment on Google analytics and waited for the results.

• For the layout with full articles, the efficiency was 21%
• For the layout with summaries, the efficiency is 38%

I'm so surprised to see how low these ratios are. Around two third of my visitors are leaving my website without even reading any thing. That's tough for me...
Then I realized that the layout with summaries was way better. So this layout has a better conversion rate.

But this isn't the only data I'm interested in. I also want to know if a reader goes from one article to another. This is quite simple to measure in the layout with summaries but it's just impossible for the other layout. Therefore I decide to take the average time on my blog into consideration as I think it is shows quite well if visitors can find the articles they are interested in easily.

• The average time on blog with the layout with full articles was 1 min 05 (5 min 11 for the readers)
• The average time on blog with the layout with summaries is 3 min 06 (8 min 04 for the readers)

The second layout has a better average time on blog for all visitors and for readers.
So the second layout is the best for both metrics. This is the reason why I kept it. I hope you like it!

Share this article

Tweet