PROJECT SCREENSInside the design
Auto-scrolling — hover any screen to pause & lift
A SaaS data-management dashboard — dense information organized into decisions, not decoration.
Data dashboards fail when they display everything and explain nothing. This concept organizes a data-management product around the decisions users actually make.
Users drowning in tables need hierarchy: what changed, what needs action, what can wait. Charts must earn their pixels.
Dashboard IA, data-visualization choices, table systems, and the component patterns that keep dense screens readable.
The landing view answers “what needs me today?”
Summary rows expand to detail — density on demand.
Visualizations chosen for the question, not the aesthetic.
Any record reachable in two interactions.
Discovery focused on the jobs users hire the product for: what decision does this screen serve, what action follows, what can wait behind a click. Auditing comparable SaaS tools showed the same trap everywhere — dashboards that display everything and prioritise nothing.
Users open the product to do something. The landing view must answer “what needs me today?” before showing anything else.
Dense tables are fine — undifferentiated dense tables are not. Summary first, detail on demand.
In daily-use tools, users get fast by building muscle memory; consistent patterns are a performance feature.
Lives in the product, values speed and keyboard-friendly flows, and notices every inconsistency. Needs dense-but-ordered screens that reward familiarity.
Drops in weekly for status and numbers. Needs summaries, trends and export — without relearning the interface each visit.
A restrained interface where colour means something: neutrals for structure, one accent for primary actions, and semantic colour reserved for status. Tables use progressive disclosure, cards summarise before they detail, and every component comes from a documented library so new screens ship consistent by default.
Prototypes were tested against realistic data volumes — hundreds of rows, awkward edge cases, long names — because SaaS designs that demo well often collapse under real data. Iterations tightened empty states, loading states and error states, the three screens teams usually forget.
Auto-scrolling — hover any screen to pause & lift
A dashboard that turns data into next steps — full case study on Behance.
View full project on Behance