Insights of a testing session

11/11/2012 00:00

Just recently some colleagues started a new (German) Website for our developers. Just when it was made publicly visible, I chose to test it myself for about 30 minutes. After that I missed to do a proper debriefing, but what for do I have a blog?!

As I posted everything which I stumbled on directly to our company-chat I can still review most of my steps. This was the only way of notetaking I chose to use. Good about it is, that other colleagues testing the website could (and can) see my notes. The one who published the site actually fixed some bugs directly and put issues to his backlog, when he saw the necessity of a conversation with other colleagues. The bad thing about just rattling the bugs/issues/questions down in a chat is the linearity. Two issues related to the same topic are now at very different positions in that conversation. As we use yammer as company-chat I won't miss any of my entries by logging out, but due to a little fault of mine, I even reported the bugs in two different threads. If the page had already been bigger, this might have been frustrating.

Here's how I tested: At first I didn't even stay on the starting page but immediately went to the page about the team. Comparing the Pictures with the names I stumbled upon a false couple, made a post to yammer and moved on to the next items in the menu one after the other to see, where I can get. Just when I saw issues with the format or spelling mistakes, I stopped to make a note and moved on.


I made my way horizontally through the menu items and then downward through the links which I saw on all pages. When I was moved to other websites of our company I just had a quick look on them, scrolled if possible and made notes about what was unexpected for me in the first five or so seconds.

After that I turned to the site notice where I stopped to write about having the content mashed together and to the site map, which told me that I left out some pages. Quickly I found five subpages which are available only for one menu entry and took a quick tour. As I recently drew some pictures for an article of one of my colleagues I missed a picture at the TDD entry. After going to the next point I realized that I didn't miss a picture on the first of the five subpages but only after seeing the second one. But there were no pictures in all five subpages (at the time I tested first). I would have missed that detail if I had not drewn the TDD picture some days earlier.

Then I finally returned to the starting page and looked for links in pictures like the it-agile-logo, the CSDevs-logo and began to read the starting page. As there were two videos on them which seemed very small I tried to make them run, succeeded but couldn't really identify things because they were so small that one couldn't have used even the movie-controls. I made a note and returned to read the remaining texts.

After having seen all pages and subpages at least once, I returned to the page with the most links and checked if they were all running and leading to the right pages. Satisfied, I checked the last PDF (which I knew because I was involved in the writing of that particular story) briefly for its contents and stumbled over a strangely formatted link which I reported as well.

After that I just did a lot of reading of contents, menu items, lists and so on and reported if anything was missing or when I found typos. Finally I reconed half an hour was over (I missed by two minutes) and stopped testing. While I was redoing my steps for this blog I even found another typo. Altogether I found 19 things that I stumbled on in 28 minutes of testing. What could I have done differently?

What I did do was a tour of

●      all the pages,

●      all the links and

●      all the content

I was aware of. The only possibility I saw to interact with the page was to click on links or to scroll up and down. I'm asking myself what I left out. Or which other approaches there are to test the page.




comments powered by Disqus