The brief was to plan, conduct and report an evaluation of the Cancer Research UK website. The charity has two primary concerns:
Five people were recruited to participate in a 45 minute user testing session. The sessions were run in-person, moderated, in a naturalistic setting. Prior to each usability test, a semi-structured interview was conducted to establish the participant’s past experience and knowledge.
To address the client’s first concern with the homepage, Steve Krug’s exploratory homepage tour technique was used. The purpose of this method is to see whether participants can comprehend the page. Participants peruse the homepage the way they normally would when visiting a site for the first time and verbalise what they understand from touring the page. This method exposes participant’s prior knowledge of the sector and their mental models. Furthermore, it reveals what is working well and what could be improved in a more natural and effective way than asking users directly for their opinions.
An assessment test using verb-based tasks was used to examine the usability of the fundraising events and shop section of the website. Tasks were chosen based on their frequency and importance. As both areas of the site are transactional, the customer acquisition journey was the main focus.
The specific task scenario for the fundraising section was:
“Your friend from outside London wants to visit you on a weekend and participate in a fundraising event while they are visiting. Find a charity walk or run near you that you would like to join. When you are satisfied you found the right event and have enough information about it, register for it.”
To uncover usability issues in the flower shop, the following task was assigned:
“You want to buy some flowers for your mum and decide to do so using Cancer Research UK’s flower shop. Find a bouquet you think they will like and when you are satisfied that you have enough information to go ahead with the purchase, add it to the basket and organise the delivery.”
A post-task questionnaires was administered using the Computer System Usability Questionnaire to capture attitudinal data. While the metrics are not of great significance for the current study, they will act as a benchmark for future studies to compare attitudinal data of an iterative design over time.
All feedback, observational data and other comments verbalised during the think aloud protocol were paraphrased and recorded. By paraphrasing the verbal protocol, it became possible to identify comments which were expressed by more than one participant.
Problem extraction criteria was defined to apply thoroughness when combing through data.
The criteria was defined as when the user:
A source data analysis was done on each usability problem to establish the root cause for user frustrations or difficulties. Accompanying the cause, is the outcome and redesign recommendation. The example below shows a frustration verbalised by participant one. In this case, the participant was unable to discern the purpose of the social media icons.
By establishing the cause, it became possible to consolidate usability problems where one root cause fixed multiple problems that users experienced.
Social media icons have no text to explain their purpose.
The participant did not know if this was for following or sharing, so they did not interact with it.
Include label text to indicate the purpose of the icons.
Lastly, the Dumas and Redish’s severity scale was used to prioritise the issues to fix.
All participants were able to complete their tasks successfully.
Participants had difficulty describing the content and functionality they would expect to see in: ‘find out about our campaign’, ‘get involved’, ‘talk to a nurse’, and ‘cancer chat’. The severity of these issues is not high, however inaccurate labeling and lack of description could cause frustration for users and a lack of engagement.
All participants expressed a desire to be able to sort event search results by date. Four participants did not understand how search results were being organised.
A number of dark patterns frustrated users. A good example is when users selected the smallest size bouquet on a product index page, and then upgraded to the largest, most expensive size once they moved through to the product description page.
This project received a high distinction.