Trinket Trove: A toy e-commerce website experience

I was given the task to create a working prototype and information architecture of an e-commerce website for a toy store based in Melbourne that had to keep “the personal touch of the small shop that started it all”.

The Goal

Based on the brief provided and through interviews with its owners, the main objective established was to create a website that will allow Trinket Trove customers to find and purchase easily those toys they are looking for with a very personalised and complemented service online and off-line.

Duration: 2 weeks

My role: Researcher & Designer

Techniques Used: Card Sorting, User Flows & Sitemaps, Sketching, Prototyping & Usability Test.

Tools used: Pen, Paper, Optimal Workshop & Axure.

The research approach

With the details of personas’ description provided I came to the conclusion of the typical behavior these personas will reflect while buying toys online as the baseline for the design approach.

Research Methods and Activities:

To approach the information architecture structure I used a base of the top 30 most popular products to categorise them on a user’s perspective using: open card sort off-line with some potential general users to gather qualitative information and a hybrid card sort online based on the first exercise to confirm categories with a broader audience and targeting people that matched the personas behavioral profile.

The results led me to a total of 4 categorisation types and 7 subcategories as shown on the website sitemap shown below. Gender was excluded as a category, because even though it comes natural to users, it’s a category Trinket Trove values and positioning don’t allow.

Additionally a detailed user flow technique was used to build a first time and return user journey as well as the check out process based on users’ feedback and expectations. The key findings were:

• The first option for first time and return user to look for a toy is a search feature. This is something that is expected as a standard. However it wasn’t include in the final prototype as it was an specific requirement on the brief provided.
• In the absence of a search option, an easy to understand categorisation system is required.
• Detailed direct links to product pages are required as users can come to them directly from search engine links.
• The check out process is perceived as a standard process that shouldn’t request anything beyond what is widely know as minimum required.

Finally, competitor’s analysis allowed me to compare and define good and bad practices within the industry:

• I confirmed the search option as a key feature.
• Very similar categorisation system at the first level.
• Extensive but confusing and overwhelming subcategorisation up to third level that should be avoided.

Sketching, Prototyping and Design Iteration Process

The initial low fidelity sketches and iteration with potential users allowed me to get input not only into the information architecture structure but the logic of navigation within the site that contributed to the final sitemap, and a the same time gave me some insights for the ‘click-trough’ prototype structure.

The ‘click-trough’ prototype was tested in three iteration cycles allowing me to improve it through the users eyes, recruiting people that matched the personas behavior.

During the first prototype iteration the key findings were:

•Products and categorisation as the predominant features.

•Search function is really important.

• Most popular / recent products demonstrate expertise.

The second and third iteration processes through usability tests allowed me to refine the prototype balancing users’ needs and business requirements as shown on the home page annotations below.