TLDR:
Artworks Discovery is the first unified entry point for art content the company built. It lets users explore works from galleries, auction houses, and sale records within a single interface, across a database of more than 50 million items.
On the surface this is a horizontal integration of content. Underneath, it held three thornier challenges: data structured differently across heterogeneous sources, market semantics that conflict with one another, and a fundamental difference in how different kinds of users search.
Some users arrive with a very specific goal, looking for a particular work. Others start from an artist, an art movement, or a market type. So the scope grew from a single page into an exploration system that has to support several search intents at once while keeping the reading of information consistent.
The project spans two core surfaces: the Discovery list page, where users browse and filter across sources, and the Artwork detail page, which presents a single work in detail. This case study unpacks the design thinking behind both.
THE PROBLEM
Under the existing architecture, users had to switch between separate systems and interpret the differences across market contexts themselves. The core challenge was how to build a seamless exploration environment where information from different semantic dimensions could be understood and compared together.
Design challenge
Build a single interface for one-stop exploration of works from different sources, while keeping market semantics clear.
And avoid letting the primary market (gallery sales) and the secondary market (auction bidding) be flattened into a direct, like-for-like comparison in visuals or structure.
RESEARCH
I ran moderated usability testing with four participants closely tied to the art market: two senior gallery professionals, a gallery founder, and a high-net-worth first-time collector. Each session combined a 10-minute interview with about 20 minutes of hands-on tasks across four core scenarios, all conducted remotely.
I mapped user behavior on two axes, familiarity with technology and familiarity with the art market, and arrived at three personas: the Visual Explorer, who browses by color, subject, and how a work would feel placed in a space; the Market Strategist, who searches by specific names and listings to answer client needs; and the Gallery Promoter, focused on the exposure and visibility of the artists they represent.
The overall usability score was 4.5 out of 5 (4.75 for the Visual Explorer, 4.25 for the gallery professionals). The most visible drop in experience concentrated on a single task: searching for a specific work.
Across the four scenarios, eight notable issues surfaced. The most severe was structural, at the architecture level: users mistook the global search box in the header for a search within the Discovery list page, which sent the find-a-specific-work task completely off course.
The full research report walks through, in order, the method, the three user personas, the core insights, and the four contextual test scenarios.
1 / 13 Background: research goals and method.
THE SOLUTION
I reshaped two core surfaces in parallel: the Discovery list page brings every source together into a clear, browsable, filterable list, and the Artwork detail page presents the full context of a single work. Here they are separately:
Surface 01 — Discovery list page
An aggregated entry point that lets galleries, auctions, and sale history sit together in one smoothly browsable list.
Every source shares one design vocabulary and card structure, so browsing stays coherent. At the same time, the core business differences are preserved precisely: a gallery work clearly reads "Price on Request," a live Artnet or partner-auction lot shows "Live now" with an estimate, and a closed historical sale shows "Results Available." One layout structure holds different market states without flattening them.
The advanced filters split by source and real-time state, including Available Now, Artnet Auctions, Galleries, Auction Houses, and Ending Soon, to work through more than 50 million works efficiently. The same web structure extends cleanly from phone to desktop, so the way a single artwork reads never shifts unnecessarily with screen size.
Surface 02 — Artwork detail page
The detail view for a single work, redesigned to carry and highlight each work's market context clearly.
The redesigned detail page thoroughly restructures the information architecture and hierarchy, giving provenance, market data, and related recommended works each a clear block of their own, instead of crowding into one space and competing for visual attention.
Because galleries and auctions carry entirely different data fields and market meanings, the detail page keeps one core backbone (its layout hierarchy) while adjusting its fields specifically for each of the two sources.
Gallery work: a customized detail-page layout.
The usability testing redefined the team's development priorities. The main friction came not from interaction smoothness but from users misreading the "system boundary." So the iteration priorities were set as:
OUTCOMES
1.6×
Browsing depth
Average browsing depth per user rose from just over 1 page to more than 2. The metric clearly reflects greater willingness to explore, though it should be read qualitatively in context rather than treated as the single absolute measure.
~30%
Monthly return rate (early data)
About 30% of users returned within a month. For an early-stage product, this mainly validates whether the "one-stop exploration" mental model took hold, rather than representing final growth performance.
4.5/5
Overall usability score
An overall usability score of 4.5 out of 5 (4.75 for the Visual Explorer, 4.25 for the gallery professionals). More valuable than the number, the results were turned into a prioritized set of design fixes rather than a scattered list of bugs.
LEARNINGS
Three of the four participants were deeply rooted on the gallery side and only one was an independent collector, so the findings describe the behavior of a specific professional group precisely, rather than extending to every audience in the art market as a whole.
The essence of the research is clearly defining the qualitative scope its conclusions apply to, rather than chasing exhaustive, all-encompassing representativeness.
The most severe pain point in this testing came not from a control or a layout, but from users expecting the header's global search to automatically narrow to the current Discovery page, when the system logic actually did not.
Designing a cross-source aggregation product is, at its core, designing the boundaries between systems: users must be able to tell, at any moment, exactly what they are searching and where the results come from.