Create a system of refinements that make it easier for our customers to narrow down and modify search results to a manageable group, tailored to their specific needs and interests.
This project was a bit of a hybrid process. We worked closely with the search engineering team even though we weren't an actual scrum agile team. Our team was responsible for discovery and framing the problems, conducting in-lab usability, leading design workshops, testing prototypes with users, and delivering validated designs and features directly to the engineers for implementation.
The discovery was a high‐intensity effort that allowed us to audit the existing experience, review the competitor landscape, understand our client's vision, and begin research into user needs, behaviors and pain‐points.
For our competitive and comparative analysis, we looked at the top e-commerce sites based on traffic, sales and search functionality. Using Baymard’s research and criteria for a best in class search experience we evaluated search on Dell.com. We then conducted a 4 day in lab usability session with 12 consumer and small business customers. And lastly using Tealeaf® the team pulled customer satisfaction verbatims from the previous 6 months search experiences.
Our research revealed that that our customers were overwhelmingly frustrated and confused by Dell.com search. Many users noticed issues such as the filters blocking the content on mobile, not being able to scroll through all categories as well as filter options moving around on the page after selections were made. In our analysis we saw that 43% of respondents stated that they needed better sort/filter options with 57% of respondents not being able to find the right result at all.
Synthesizing goals from our research served as a lens to guide our brainstorming. The four goals are as follows. A good filtering and sorting experience should:
Reward every interaction with refined results
Be quick and responsive
Be consistent and predictable
Only offer controls that are essential to the query and results
We used personas constantly throughout the project to guide design decisions,priorities, and create empathy amongst the client and our team. Knowing exactly who we were designing for allowed us to better visualize how the feature would be used by our customers.
In a four hour working session we used scenarios based off real world problems to sketch concepts in a storyboard format. At the end of the session we had 100s of ideas with 26 complete solutions. These solutions formed the backbone of our requirements.
Using a risk reward chart we were able to identify smaller features to fill the backlog and bigger features which needed more extensive prototyping and testing.
The filtering solution we delivered resulted in 10x the amount of engagement from .25% to 2.26%, and a lift in effective conversion from.02% to .23%. The search experience is continually being evaluated and iterated upon.
