17 Feb 2013 Estimated Read Time: 3 minutes, 2 seconds
Over the last month or so I have started regularly posting again on Yeah it’s Pretty Cool after almost a year of neglect. The reason being that this is probably the oldest website that I still have on the Internet and my most profitable one (meaning it has actually generated some revenue).
I am actually enjoying it because it is now easier than ever to find gadgets and home hacks as it is now in vogue at the moment, almost reaching the mainstream with everyone wanting a customize their gadgets. But rather than just blindly posting like I have been over the last few weeks, I took about 40 days of Google Analytics data to see what effect the new posting has done to the traffic and performance of the website. Here are some of the key stats:
- 272 visits over the last 42 days – 6.4 per day
- I have posted 11 new articles during that same period – one article every 3.8 days
- I have had 252 page views but only 17 have come from those new articles. The articles only got 1.5 page views each, very poor performance.
- 65% of visitors only look at one post in their visit.
What This Data Tells Me
The stats never used to be this bad. But the major difference between this version and a older version of the website was that content discovery was easier. I remember having an ugly tag cloud quite large on in the sidebar that I always hated, but it did generate a lot of page views and got the posts read number up to more than triple than what it is now.
Also, none of the category pages or the tag pages are getting looked at, or being landed on from various traffic sources. I guess that explains why people are looking at only one post and leaving.
Other points to consider:
- I need to make the category pages more obvious and highlight similar articles with a related posts plugin (something I might try writing myself using my own custom algorithm).
- My older posts that have good positions in search engines are driving the bulk of traffic. In order to get continue traffic to the website I need to improve the rankings of the new posts because they are not getting any of the traffic.
- Social is really not performing well – I don’t know if the fan counts is the problem, or the fact that it is pretty much all my friends that have liked it and not really gadget inclined and don’t feel like sharing them.
The Action Plan
So the plan based on this data is to improve content discovery on the website. It might not be pretty to start with, but if the numbers improve over the next 30 days I will be really happy with that. In a perfect world I want to double the figures I am getting at the moment.
How I am going to improve the numbers?
- Feature related posts at end of every article
- Modify the navigation bar to put around 3-4 key categories at the top
- I am not going to put the full post on the homepage, just an excerpt and the feature image to encourage the click through. I am not sure if that is a better user experience or not, but many of the major blogs do this.
By the way, there were no fancy scripts that were used to do the maths or analysis. Just a pen and paper, a few of the standard Google Analytics reports and examining the website and make some educated guesses as to what people are doing on the website. Sometimes scripts are just over kill.
06 Nov 2012 Estimated Read Time: 1 minute, 2 seconds
I have started building questionnaires for various team-mates and external parties to answer, rather than have a trail of emails or prettied up Excel files being shared all over my local machine or some share drive.
I was considering the other option of custom building something and sharing it on a small server that is laying around the office, but it would be complete over kill considering there are tools out there like Wufoo and others who have a nice pretty interface and system already built available for free or small monthly amounts.
However, the best solution I have came up with is Google Docs (which I know could be a bit late to the party on this one). But it allows you to create something pretty to share, available anywhere, anytime in the cloud and stores it all in one massive excel/csv file.
And now with Google Drive functionality, allowing me to have a direct path to pull it from in my scripts, this is by far the most flexible way for developers I think to create, store and manipulate web form data.
I have yet to give it a go by embedding it into a website, stay tuned on how that is going to go!
30 Oct 2012 Estimated Read Time: 0 minutes, 51 seconds
So as part of my ongoing learning of slowly porting everything I have ever done from PHP to Python, I have decided to setup a domain (Let’s Django) which will be powered by a new open source project called Djog.
This is something I have wanted to do for a very long time and I am finally in a position to do so. I want to document every single thought process and idea of building a blog engine from scratch and then sharing all my code (good and bad) to all the people, who will hopefully then contribute back to me via pull requests or (even better) submit a pull request as well as a blog post explaining their ideas and concepts behind their changes.
I also want to give back to the community by featuring amazing Django developers and the companies that are actively using the framework for free on the blog as well.
There is no formal roadmap as yet, but stay tuned to see how it will turn out.