Google wanted to know when people use images to solve an everyday information need. We ran a seven-day diary study with 30 participants in the US and India, then used their entries in 16 one-hour interviews. The research identified six mental models, three journey types, and one eight-step behavior map. Visual search was not a default habit. Saving and revisiting images, especially while shopping, mattered more.
Google wanted to know when people use images to solve an everyday information need. We ran a seven-day diary study with 30 participants in the US and India, then used their entries in 16 one-hour interviews. The research identified six mental models, three journey types, and one eight-step behavior map. Visual search was not a default habit. Saving and revisiting images, especially while shopping, mattered more.
Google wanted to know when people use images to solve an everyday information need. We ran a seven-day diary study with 30 participants in the US and India, then used their entries in 16 one-hour interviews. The research identified six mental models, three journey types, and one eight-step behavior map. Visual search was not a default habit. Saving and revisiting images, especially while shopping, mattered more.
Visual search and daily information needs
Text-based searcher, image-collector, image-sharer, feedback-finder, reverse-image searcher, and true visual searcher. Each carries a distinct expectation about what a visual search tool should do.
The product team could now decide where visual search belonged based on the behavior in front of it. Image collection became part of the shopping journey instead of an adjacent feature.
Methodology
Phase 01
weight 3
Seven-day remote digital diary with 30 participants across the US and India. Captured visual information needs in the moment, including triggers, tools, and outcomes.
Phase 02
weight 2
Clustered diary entries into mental-model archetypes and journey types. Flagged the highest-signal entries for follow-up exploration in phase 03.
Phase 03
weight 3
Sixteen sixty-minute depth interviews, grounded in the participant's own diary entries. Pushed on mental models, decision moments, and the role of image collection.
Phase 04
weight 2
Produced six mental-model archetypes, three journey types, and an eight-step behavioral map. Framed the implications for visual-search product direction.
Findings / 5
01
Text-based searcher, image-collector, image-sharer, feedback-finder, reverse-image searcher, and true visual searcher. Each carries a distinct expectation about what a visual search tool should do.
02
Visual search sits outside most participants' go-to patterns. It is reached for in specific moments rather than used as a daily query mode, which reshapes how and where visual search surfaces should appear.
03
Saving, grouping, and revisiting images is how many participants move through shopping and planning. A tool that supports collection directly matches a real, repeated behavior.
04
Image-initiated, text-initiated, and exploratory journeys all fit a shared eight-step behavioral map (trigger, frame, choose tool, query, scan, refine, act, save / share). The journey type, not the step list, determines product design choices.
05
US and India participants differed in what started a visual information need more than in how the need resolved. The behavioral shape is stable. The entry conditions are market-specific.
Structure / 1
01
Method, sample, and output shape as delivered.
datasheet
Deliverables / 6
Outcome
The product team could now decide where visual search belonged based on the behavior in front of it. Image collection became part of the shopping journey instead of an adjacent feature.