Did you know…
- 84,8% of users visit the ETH Library website via their desktop?
- Firefox is the most used browser?
- with 79,6%, most users are from Switzerland?
ETH Library regularly collects anonymous usage figures for its websites and platforms to optimise its user experience. In the process, a statistical analysis of these usage figures is essential to identify optimisation measures.
In this blog entry, we would like to outline five steps to perform this kind of web analysis.
1. Define goal and task
Formulate a clear goal which you would like to achieve with your web analysis. Work out a task which you can accomplish and makes sense for your activities/endeavour.
Make a selection of subpages which you would like to analyse, for instance, and set a time limit that you would like to analyse.
2. Set suitable KPIs
Set suitable key performance indicators (KPIs) to work out and achieve your goals.
These KPIs might be the following, for instance:
||New visit to a website involving several page views (A visit ends when the visitor closes the browser, deletes the cookies or has been inactive for 30 minutes.)
||Number of pages visited during a visit
||Keywords entered in a search engine to find the website.
à helpful for SEO (search engine optimisation)
||The previous page the visitor was on before clicking on the page to be analysed (original page)
||Reveals how many visitors leave the website on the first page visited, without visiting any other ones.
à the lower the percentage, the better
||Visitor’s country of origin
||Website visited via desktop or mobile device
3. Evaluating data over a longer period
Activate the desired KPIs in your analytics tool. This enables the key figures to be evaluated over a longer period, which is very important for a meaningful analysis. Fix a time limit (e.g. the last three months) and make sure that the desired period is always set in the analytics tool when you begin the individual analysis.
4. Compare usage figures
To draw useful conclusions, it is important to compare the KPIs evaluated.
Especially scrutinise the figures in the event of outliers or deviations and try to compare the figures.
The ratio of page views to visits or the average time spent on each page is ideal here.
5. Draw up measures
And something very important to finish on: collecting KPIs and presenting them in pretty diagrams is not enough by itself.
Work out any need for action and measures that have to be taken to render your website more successful.
Starting points for recommendations:
- Selection of compatible browsers
- Website design
- Optimisation of certain subpages
- User-friendliness of the website on the web
- How easy the website is to find on the web
Over to you: tackle the web analysis one step at a time. After all, a good web analysis does not just improve the efficiency of the website’s design; it also enables the success of your website to be monitored in the long term.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International Public License.
DOI Link: 10.16911/ethz-ib-3026-en