In this post, I’m going to share web analysis methodology that I think will be useful for a web analysts. This web analytics methodology is good for optimizing website performance & increase customers satisfaction through the websites.
- Define Issue/Question
- Set Hypothesis
- Investigation:
- Optimization Actions
- Apply Testing & monitor results
I think this is the first step for successful analysis process. Define the issue, question or the problem that you want to find answer for. For ex. check the top content report at Google analytics, order pages by the highest Bounce rate, then detect 5 pages that have bounce rate greater than 80%.
Second step is to set hypothesis for the causes of the problem/issue. For ex, In the above example, the causes of high bounce rate is may be non relevant traffic, none convinced content, no clear call to action or boring design layout.
Third step is to investigate to make sure which of these hypotheses the real cause of the problem. In our example, after investigation to the traffic for these pages, I found that traffic is relevant. So the problem cause may be the in the content or design layout. So, you may compare between successful pages & these pages to figure out what difference between them in design & content.
In this stage, you have to apply updates/changes that most properly will fix the issue. These updates you can get from successful parts at your site or your competitors. In this example, you may update content to be more convinced content, update layout design to be more attractive.
After make some updates, you have to apply testing such as A/B testing or multivariate testing to make sure if the new updates working better than the old version.
Let me what do you think, do you apply that before? Let me know your feedback about this post.
