SHAKE THE TREE
An Information Architecture Project
THE PROBLEM
Shake the Tree is an existing store in Boston's North End and has a variety of items and a website but no online commerce available.
With a sprint of 2 weeks I conducted interviews with onsite and remote open and closed card sorting, competitive and heuristic analyses and usability testing to create and improve a wireframe, ecommerce-based website prototype for Shake the Tree.
HIGH LEVEL TIMELINE
Two Weeks
MAKE OF THE TEAM
Just me!
KEY GOAL
Successful information architecture and forms
UNDERSTANDING THE PROBLEM
I interviewed the store owner and six different ecommerce shoppers who fit the owner's description of her clients as women of all ages, from 20 to 70. Her store has been a local favorite for over 20 years and is a fun project she does with her daughter. She would sell online, but is scared of the extra work.
​
The shoppers I interviewed gave me important goals, behaviors and pain points to consider, displayed below.
HEURISTIC ANALYSIS
A simple heuristic analysis Shake the Tree's current website to be performing at a score of 60%, compared to competitors' average score of 96%. Two big box and one similar locally owned store websites with ecommerce were used.
PROTOTYPING, ITERATING AND TRACKING SUCCESS
After some sketches, site mapping and user flows with pencil and paper, I made a prototype in Axure.
​
Once a prototype was made, I tracked success through iterations by monitoring issues in usability testing, and by administering standard usability scale surveys to post-testing.
One error caught in testing was a missing line for an email entry on the order form, in addition to some flow issues with leaving a star rating.
All users were successful in finding the item tested (herbal tea) within the prototype's functionality, supporting a successful information architecture.
​
Early SUS scoring was low at 67% before corrections were made, and increased to 78% once corrections were tested.
​
The final site map is shown here:
Click the image below to try the prototype.
FINAL LEARNINGS
Allowing user goals found in interviews, and user behaviors observed in open card sorts, to inform information architecture choices sets a site up for success. Confirming these choices with closed card sorts and later prototype testing further supports design choices.
​
Usability testing is paramount for catching errors. Some users will inherently be less invested in usability testing. Having specific tasking and more interactive prototypes allows for better testing.