Find what you need with quickness & ease
Product designer
User Interviews
Wireframing
Prototyping
2 engineers
1 product manager
3 months
Abound is an online two-sided marketplace for buying and selling wholesale. Consistent feedback from our retailers is that our search is "bad" & they can't find what they're looking for, but we don't know exactly why that is. So before I could make any improvements to the search experience, I needed to figure out: what exactly is our problem with search?
I started this project by looking at the ui/ux of our current search experience to identify pain points. I did this to get a deeper understanding of the problem space & so I could explore these insights further with the retailers I planned on talking to.
I interviewed 8 retailers via Zoom––the participants I chose either expressed being unhappy with our search (found by looking through old interview notes & PMF survey responses) OR they appeared to have had a poor search experience (found by watching FullStory recordings).
I started by asking them questions about their current process for finding products/brands on Abound & then did a task with them where they actually search for something.
Based on observations during the interview task, this tends to be after 3-4 pages. At this point they'll either try another search term or move on to other things. Is there a way we can surface more products they'd want in these first 4 pages and/or provide them related search terms?
General search terms (e.g. "coffee") & search terms with descriptors (e.g. "wooden knife") seem to face this issue the most
They search for a brand when they need to reorder and/or want to see if a brand is on Abound. Sometimes they don't know the exact brand name or mispell it so none of that brand's products comes up in the search results. Plus there's confusion that the search results page is actually the brand shop page...
On the product team we differentiate direct searches using the search bar from browsing via our category pages. However our retailers don't have this same distinction & consider both as "search", so some insights from browsing the category pages were:
I brainstormed opportunities & improvements based on my interview findings & the pain points I identified:
I worked with my product manager to determine which ideas we wanted to test & then prioritized them based on how much they might impact GMV and engineering effort:
We prioritized building features that require low engineering effort & have the potential to have high impact on GMV (the upper left section in the priority map).
Now that we had our features prioritized, I did some competitive research to get some inspiration and see if there were any established patterns. I looked at Uber Eats & Instagram for search ui/ux, Etsy & Instcart for brand CTA on search result pages, and Faire & Tundra for image carousel inpsiration.
previous
current
I focused my changes for this page on making it more obvious it's a search results page & making it easier to get to a brand shop page:
previous
current
Retailers will be able to click through up to 5 product images on brand cards to get a better sense of what the brand offers.
previous
current
Improve search dropdown ui/ux: After releasing the ui/ux enhancements, we saw frustrated clicks in the search dropdown decrease & clicks on the brand suggestions increase. Plus there are still more ideas I'd love to test—on the roadmap is showing trending search terms & personalizing the featured brands in the search dropdown.
Updates to search result pages: The addition of the brand CTA led to more retailers finding the brand shop page from the search results page. Right now the brand CTA only appears if the search term exactly matches a brand name. We could make this even better for retailers by having the CTA show up for slight misspellings.
Image carousel on brand cards: During testing we saw performance issues due to adding so many images to the page. As a result, we didn't end up releasing this feature because we didn't want to greatly hinder performance for these highly trafficked pages. Page performance was a big takeaway for me coming out of this project! It's now something I think about when designing & discussing designs with engineers.