Two Methods in Information Architecture and How to Communicate with non-IA Stakeholders
Information architecture (IA) is widely used to describe the information on a website. It also structures and organizes the information together. Additionally, IA can plan site maps, diagrams, and spreadsheets. IA is often used to redesign a project. Some IA improvements make the information discoverable for users, because a good amount has already been introduced.
How can methods be used to uncover flaws within information architecture?
There are many methods that we can use to uncover flaws within IA, like card sorting, tree testing, content auditing, wireframing, diagramming, and data and analytics. Among these, card sorting and tree testing are the most common IA methods designers use.
- Card sorting
According to the Nielsen Norman Group (NN/g), “Card sorting is a UX research technique in which users organize topics into groups. [You can] use it to create an IA that suits your users’ expectations” (2018). Card sorting allows designers to find quick solutions at a low cost. It is a user-centric process and a powerful way to create a better user experience. When using this method, website users can participate in the website’s IA.
For example, a fashion eCommerce website (like Gap) usually starts with big categories like Men, Women, Kids, Accessories, Sales, and New Arrivals. The reason they display these big categories is to provide the highest level of information to meet customers’ needs. Customers want to immediately know where to go, start, and explore by just viewing these categories. When designing, designers make category cards and then break them into subcategories, like types of fashion items that they might be interested in: T-Shirts, Sweaters, Button-down Shirts, Jackets and Coats, Pants and Shorts, Sports Suits, and so on.
There can be many reasons that a website is not working well. However, card sorting improves the website’s IA. The ultimate goal of card sorting is to properly group, categorize, and organize the content in order to build better IA.
2. Tree testing
In order to test the effectiveness of card sorting category and subcategory discoverability, designers would need to run a tree testing. Tree testing maps out these categories into trees like accordions, so they are presented in levels. Then they would invite some participants to join the experiment and see if they can find certain information in the deep tree hierarchy.
Tree testing is a qualitative research method in which the researcher can potentially observe information that participants miss. “For example, in a recent tree test, we noticed in the pilot testing that many users avoided a certain category for the first half of their session because the label was so broad that they feared the contents would be overwhelming” (Whitenton, 2017). For example, Accessories might be too broad, because it contains Hats, Belts, Shoes, and many other possibilities.
Moreover, according to NN/g, tree testing can also combine with A/B testing to identify which IA version is preferable to users. However, that would require a good number of participants, over 100, in order to be 90% confident in the results (Laubheimer, 2019).
How do you communicate information architecture to non-IA stakeholders?
Data analytics can be a persuasive method because people tend to trust data. In the article “Using data to persuade,” the author indicates that “analysis involves compiling and examining the data for patterns or tendencies. Persuasion will probably be required to reach agreement on both the tendencies revealed by the data and what these tendencies mean. Persuasion will be required to reach an agreement on a tentative plan for using the data and knowledge gained from the study” (Covey, 2005). Data is given to the stakeholders to explain tendencies and patterns, and persuade them to agree with the analysis.
For example, a fashion eCommerce website might want to promote summer fashion items before summer comes. From a regular IA perspective, summer clothes are probably buried under each category like Men and Women, then subcategories, like Shorts or T-shirts. However, if data analysis shows that people do not visit these categories as often as they should, then it is time for the designers to persuade stakeholders to change the order of the categories. This change in IA gains more traffic to meet the commercial goal.
Designers can also use the following NN/g statement to communicate to the non-IA stakeholders: “The volume of traffic to a category is the most obvious indicator of how useful or interesting the category is to your audience. Traffic can be measured as the number of views of the main category page, or by aggregating views of all the pages within a category” (Whitenton & Sherwin, 2016). Here we can see that IA affects discovery, discovery affects visitation, visitation affects conversation, and conversation affects promotion goals.
Conclusion
IA creates well-structured content. This means that visitors can more easily find what they want on the site. Without it, visitors will leave the site unable to find what they wanted. They may also have had difficulty making a purchase. Using IA knowledge means creating a site with better usability. Card sorting and tree testing are a good combination in order to discover the effectiveness of IA. These methods help researchers improve user experience research. Researchers can then obtain qualitative data that reveals the why and how. Data analysis covers the what by gaining quantitative data. All these methods and tools are relevant for improving and refining IA.
References
Whitenton, K., & Sherwin, K. (2016, September 25). 5 Information Architecture Warning Signs in Your Analytics Reports. NN/g
Review: https://www.nngroup.com/articles/ia-warning-signs-analytics/
Whitenton, K. (2017, May 7). Tree Testing: Fast, Iterative Evaluation of Menu Labels and Categories. NN/g
Review: https://www.nngroup.com/articles/tree-testing/
Laubheimer, P. (2019, November 1). Tree Testing to Evaluate Information Architecture Categories. NN/g
Review: https://www.nngroup.com/videos/tree-testing/
Covey, D, T. (2005). Using Data to Persuade. Social Science Module. Page 82–89.
