{"id":1725,"date":"2026-02-17T12:00:35","date_gmt":"2026-02-17T17:00:35","guid":{"rendered":"https:\/\/sites.temple.edu\/assessment\/?p=1725"},"modified":"2026-02-24T13:01:59","modified_gmt":"2026-02-24T18:01:59","slug":"revisiting-the-bento-style-everything-page-in-library-search","status":"publish","type":"post","link":"https:\/\/sites.temple.edu\/assessment\/2026\/02\/17\/revisiting-the-bento-style-everything-page-in-library-search\/","title":{"rendered":"Revisiting the Bento-Style \u201cEverything\u201d Page in Library Search\u00a0"},"content":{"rendered":"\n

Note: <\/strong>This post was written by Michelle Mackinsky and Jackie Sipes<\/p>\n\n\n\n

Background  <\/h2>\n\n\n\n

Temple University Libraries offers access to books, articles, databases, archival collections and more through Library Search<\/a> \u2014 our primary unified discovery interface.  The \u201cEverything\u201d results page uses a bento-style layout to group results by resource type and gives users a high-level snapshot of what\u2019s available across our collections. <\/p>\n\n\n\n

In Fall 2025, the Discovery Oversight Group, a team that stewards the ongoing development of Library Search, proposed adding a new resource type to the \u201cEverything\u201d results page: Archival Collections. With our special collections finding aids now searchable in ArchivesSpace,<\/a> increasing their visibility felt timely and aligned with our broader access goals.  <\/p>\n\n\n\n

At first this seemed simple: add another bento. But our existing page was already a tightly packed three-column, two-row grid. Adding a seventh box risked turning a scannable snapshot into visual overload. What began as a request to add a new bento, ultimately prompted us to step back and revisit the overall design. If we were changing the page anyway, how might we improve it? <\/p>\n\n\n\n

\"\"<\/a>
Screenshot of former Everything page design<\/em> <\/figcaption><\/figure>\n\n\n\n

User Research <\/h2>\n\n\n\n

Joi Waller, Michelle Macinsky, and Jackie Sipes started by reviewing discovery layers from peer institutions and revisiting our own past user research on Library Search. As we compared sites from other libraries, we paid close attention to how they structured their results pages \u2014 whether they favored tighter grids or more editorial, column-based layouts; how they handled visual hierarchy; and where key navigation elements like \u201cSee all results\u201d were placed. <\/p>\n\n\n\n

We were especially interested in whether short, plain-language descriptions could clarify distinctions between resource types. In a unified search interface, categories like \u201cDatabases\u201d or \u201cDigital Collections\u201d are not always intuitive. Small cues can make orientation easier \u2014 particularly for novice researchers.  <\/p>\n\n\n\n

This idea wasn\u2019t new. In previous Library Search testing, participants had suggested adding brief descriptive text to clarify what types of content are included in each bento. Research from the University of Michigan Libraries<\/a> affirmed similar findings. <\/p>\n\n\n\n

Google Analytics Data <\/h2>\n\n\n\n

Before redesigning anything, we looked at Google Analytics click events on bento headers over a two-month period (April 9\u2013June 8, 2025): <\/p>\n\n\n\n

Bento Resource Type<\/strong> <\/td>Header Clicks (4\/9\/25-6\/8\/25)<\/strong> <\/td><\/tr>
Books & Media <\/td>707 <\/td><\/tr>
Articles <\/td>395 <\/td><\/tr>
Databases <\/td>27 <\/td><\/tr>
Journals <\/td>12 <\/td><\/tr>
Library Website <\/td>10 <\/td><\/tr>
Research Guides <\/td>2 <\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

Books & Media and Articles were already positioned at the top of the \u201cEverything\u201d page, and their high engagement re-affirmed this placement. The data also clarified which bentos had lower engagement. If we needed to reposition categories to accommodate a new resource type, Library Website and Research Guides were logical candidates. <\/p>\n\n\n\n

One caveat: this data captured header clicks only, not interactions with individual results inside each bento or overall impressions of the page. To understand user expectations more fully, we needed to talk to people directly. <\/p>\n\n\n\n

Prototyping & User Feedback <\/h2>\n\n\n\n

Based on our research, we developed four prototype designs. Each included the new Archival Collections bento and incorporated elements that could improve usability and readability. We intentionally varied the designs to gather feedback on those specific elements with users. <\/p>\n\n\n\n

Across the prototypes, we explored different column densities, (two- versus three-columns), compared evenly aligned grid-based structures with layouts that stacked sections in vertical columns for a  more newspaper-style layout, tested whether lower-priority categories should be visually grouped or set apart, experimented with boxed styling for clearer separation, and adjusted the placement of Library Website and Research Guides within the hierarchy. By isolating these variables, we were able to ask participants not just which layout they preferred, but why \u2013 and which elements meaningfully affected clarity and scannability. <\/p>\n\n\n\n

\"\"<\/a>
Four prototype options shared with users for feedback<\/em> <\/figcaption><\/figure>\n\n\n\n

We asked participants to compare multiple prototype layouts of the \u201cEverything\u201d page, talk through their immediate reactions, what felt clear or confusing, and indicate which version they preferred.  They also completed a ranking exercise to identify which resource types they would want to see first when searching. Nine participants (4 undergraduates, 2 graduate students, 2 staff, and 1 faculty member) took part; most had previously used Library Search (7 of 9), typically to find either books or articles. <\/p>\n\n\n\n

Key Findings<\/h2>\n\n\n\n
\"\"<\/a>
Prototypes set-up in the Charles Library lobby for user feedback<\/em> <\/figcaption><\/figure>\n\n\n\n

Prototype Preferences<\/em> <\/p>\n\n\n\n