Artsy Web Navigation
Artsy is an online art marketplace that aggregates artworks from leading galleries around the world. Users can shop for art, browse shows and auctions, and learn about the art world at large using Artsy’s online platform or iOS app.
Background
Artsy’s navigation had never been considered holistically from a user perspective. It reflected an older business model that prioritized partner visibility and education, as opposed to its marketplace. Things were organized in multitude of ways and didn’t follow any sort of hierarchy.
We often heard feedback from users stating how difficult it was to find what they were looking for. We knew we needed to create a navigation experience that was both user-friendly and commercially-focused. Our plan was to approach this problem with a series of smaller, iterative steps and evaluate our strategy along the way.
Previous Design
Initial User Studies
Ideally, high-level navigation reflects a hierarchical structure that aligns with widely accepted patterns within a given industry. For example, clothing departments generally follow a similar flow and has now become commonplace among retailers. However, in the online art space, we were setting forth into new territory. In order to work toward a solution that users could easily navigate, we needed to understand how they thought about categorizing an online inventory of art.
We organized a study in which we walked users through a series of designs and asked them to rank their choices along the way. They provided context for their decisions throughout the process and helped us gather observations with key takeaways we could fold into our overall vision for the site navigation.
Key Observations & Trends
Regionality is a major catalyst in a collectors browsing behavior.
People would initially browse on Artsy to see what was available, but then want the option to view the work in-person if possible. This was especially true of many of the established collectors we spoke to (many of them living in major metropolitan cities).
Power-browsers prefer visual inspiration and an abundance of pathways.
When users were in initial discovery mode, they preferred lots of visual inspiration when landing on Artsy’s homepage. They wanted multiple pathways that were easily accessible to help them gather inspiration, and eventually, narrow their focus to something that fit their taste and budget.
Collectors have a strong desire to learn about artists' identities.
Multiple collectors mentioned interest in the identities or nationalities of artists they liked. Knowing an artist’s background grounded the work in an important context. Collectors noted they would be interested in exploring multiple identities which would lead to extended browsing time.
Emerging artist status suggests accessibility and affordability.
Collectors noted they would want to check back frequently to look at what artists were new and emerging. Many of them desired to follow their trajectory as they entered the art world and build up their rapport. Works by emerging artists were more financially accessible to collectors and potentially less intimidating to pursue.
Collectors discern quality and crave freshness.
Several users expressed interest in browsing curated lists as an indicator of quality. Collectors seem to trust Artsy but realize galleries are still uploading a lot of "junk" to the site. They wanted Artsy to consistently provide fresh, relevant recommendations.
Medium is the most common conceptual model for browsing art.
Among all specific sub-groupings shown to users in the study, "medium" was most often called out as a method by which they would want to browse. Many collectors targeted a specific medium within their collection (ie: photography) and therefore this seemed to be the easiest pattern for them to follow.
In summary
There was an even split among participants selecting Artists and Artwork as their first step in browsing. This gave us enough support to push forward with these vectors in future navigation tests. Overall, the strongest signals we got from users were around regionality, artist identity, career stage, and artwork medium.
Nav Bar Reorganization Test
Our next steps built upon the findings from the above user research and led us to a lightweight yet insightful AB test that involved reorganizing some of the existing navigation items on the site.
Our test would expose “Artists” and, in turn, would move “Galleries” and “Fairs in its place. The outcome would tell us whether the top-level items we presented were an effective combination for users to begin browsing by.
Control
Experiment
AB Test Outcome
We ran the AB Test for 10 days in which we saw “Artists” attract more click-throughs than “Artworks” and “Auctions”, both of which perform higher in terms of commercial action. This resulted in an insignificant difference in conversion to shoppable art between the two variants. However, we recognized that the landing experience for “Artists” wasn’t optimized for that intent . The initial click-throughs alone were a clear indicator that users were interested in the resurfaced “Artists” nav item and gave us confidence to move forward with this direction.
Nav Bar Drop-downs
From prior tests we knew two things: users desire to browse by “Artists” and “Artworks” and their following steps were most often influenced by regionality, artist identity, career stage, or medium. We wanted to push this idea further with quantitative data by introducing category drop-downs underneath these two top-level vectors.
We consulted with our internal curation team to create an initial set of subcategories to trial with users. We monitored initial behaviors and planned to continually reevaluate the performance of these subcategories, while also monitoring overall conversions to artwork pages via the navigation.
Our goal was to convert more user sessions into artwork page views via navigation drop-downs. Each link within the drop-downs would route users to curated artwork collections, or a filtered artwork view. By showing users a more narrow subset of works, the likelihood of match-making increased significantly. We also hoped to learn which subcategories performed better than others. We would continue to iterate on this into the future.
Future Vision
We created a holistic recommendation for a future vision for web navigation that included product, business, and brand considerations. With this proposal, our goal was to give users an immediate understanding of what Artsy has to offer and the most important services provided.
Navigation had to also speak to the two sides of Artsy’s business model—sellers and buyers alike. We had to strike a balance that told the right story and provided necessary pathways for all types of users.
Design Principles
A sense of purpose should be
immediately understood.
Our aim was to provide users with the right amount of context to intuit what services Artsy offered and what could be done on the platform.
Language should feel clear, concise and actionable.
We wanted to speak to the user and empower them to make selections with ease and confidence.
Information density should
feel gradual.
To avoid overwhelming users, we recommended providing fewer pathways initially and slowly open more as the user engages.
Trends within the marketplace
should be evident and clear.
Users should have an idea of what is trending or popular on the site and not have to decipher information on their own.