I started the week by completing the survey questions and sending it to Tigger. This needed a lot of thought put into it because originally I was asking the same questions as I did Tigger in our meeting. These were “What do you think the website is for?”, “What do you like about the website” and “What do you dislike about the website?”. When asked in a face-to-face environment I was able to engage in conversation and bounce topics off each other. The reason I couldn’t use these on an online platform is that the questions are too open-ended, the answers I get from the selected user group are final and there is no opportunity for discussion. In the worst case scenario, the data gathered could even be completely irrelevant. To overcome this hurdle I adapted my questionnaire to say “What aspect of The Greenhouse Trust are you interested in?” and “Do you have any additional comments or suggestions you would like to add?”. The first question is more closely tailored to defining the correct user groups of the website which is my main goal. I really need to know who I am designing for and why/how they can benefit from an improved user experience. To start with it also said “What aspect of The Greenhouse Trust website is you interested in?” but I thought this was too specific to the website. Due to the nature of the business and their clients being more familiar with interacting physically (e.g. coming into the cafe or visiting the bookstore) than with their online presence, it is essential for my online platform to be used as a compliment to this service, rather than replace it. If I can find out how people interact with The Greenhouse Trust as a whole then I can start catering to their needs appropriately. For the final question, I needed more of an open-ended question to enable people to let their voices be heard. I am working on the assumption that as this is part of a creative community I may find myself presented with ideas I hadn’t considered before. Additionally, I think it will make the user feel a lot nicer about doing the questionnaire because they have an opportunity to say anything they want after having to fill out my mandatory questions.
Later on this day I developed a wire-frame of the information architecture of the main menu system.
As you can see from the key, the items in boxes with solid borders indicate page names and items in boxes with dotted borders indicate what information that page will hold. This was an extremely useful exercise because it made it clear to me in my mind what needed to go where. To be able to map out a complex 2 tier menu system with multiple options was essential for the development of my project.
After this, James helped me set up my new WordPress site. We added password protection as per the Memorandum of Understanding and I could start looking at the menu system I had created on paper in action. I chose a theme and familiarised myself with it but soon came to realise I was going down the wrong track for prototyping. Granted, this was all very valuable knowledge for me to be learning, however, I was drifting away from what I was going to be testing. This is what I realised I needed to change my test rationale if my hard work into the information architecture was going to be put to good use. In my last post, I outlined that I would be comparing my new site to Tigger’s old one. I have now decided that they are too visually different and the test wouldn’t be fair to the user. The results would be tainted by letting the user use poor visual design over a simplistic, modern one. Along with this, the front page, menu structure and images would be different. I thought of a few ways around this, one of which included duplicating the menu structure from The Greenhouse Trust website to fit the same visual design as the one I am working on now, but the test conditions would be awkward as I would have to change the menu structure manually in-between tests as WordPress doesn’t support interchangeable menu options. This would, in turn, be time-consuming and leave too much room for human error (I could input the wrong menu option). With all this in mind, I decided it was best to start from square one again. A simple Google search of “How to test Information Architecture” revealed so many useful insights into how industry standard testing is conducted. Two of the quickest and easiest ways is performing card sorting and tree testing exercises. I believe these will be the most effective ways to single out my information architecture (IA) and not distract the users with any visual prompts which were one of my original problems. Because of this, however, I predict I will have to conduct another lot of tests which uses the results of this test in the digital environment I have created.
I managed to get my survey out to Tigger and he forwarded it to 10 people for me. I heard back from 4 people on Friday so I am going to wait until I have a few more responses until I start to analyse the data more closely.