Problem
What made this hard
When a metric looked wrong, people struggled to find its source, transformations, and downstream effect. Large dependency graphs quickly became unreadable.
All case studies06 / 11
Data intelligence · Enterprise SaaS · 2022
Data Lineage
Trace a metric from source to business impact.
RoleSenior UI and UX DesignerStatusShippedContextEnquero · PowerMeEvidenceSource to output tracing · Progressive graph exploration · Impact analysis
Outcome
What changed
A progressive exploration model that pairs an overview graph with focused paths, node details, and impact analysis.
The thesis
A lineage experience that reveals how source data moves through transformations and reaches downstream business outputs.
Case-study proof
The decision record
Enough context to judge the choice, inspect the work, and separate evidence from ambition.
Why this, not that
I used progressive graph exploration instead of displaying the entire dependency network at once. A complete graph preserved every relationship but made investigation unreadable; a focused neighbourhood kept the selected metric and its immediate path understandable.
Why this case is specific
The interaction was designed for PowerMe lineage work: tracing a metric from source data, through transformations, into downstream business outputs while keeping impact beside the selected node.
What you can inspect
Shipped interaction model · illustrative reconstructionThe case explains the shipped interaction model and shows an illustrative reconstruction of the focus-path behaviour. It does not present the reconstruction as a production screenshot.
Decision to outcome
Progressive disclosure made source tracing, node inspection, and downstream impact part of one investigation path. The supported outcome is the shipped workflow; no reduction in investigation time is claimed.
Interface evidence
A graph is not yet an experience
The first job was controlling complexity. The default view shows meaningful relationships. Filters and focus states reveal depth only when a person asks for it.
Trace without losing context
Selecting a node opens details and highlights the immediate path before and after it.
01 / Model
A graph is not yet an experience
The first job was controlling complexity. The default view shows meaningful relationships. Filters and focus states reveal depth only when a person asks for it.
02 / Explore
Trace without losing context
Selecting a node opens details and highlights the immediate path before and after it.
- Clear source to output direction
- Focused neighbourhood view
- Search and object filters
- Impact shown beside detail
03 / Decision
Use detail where detail helps
Labels, edge weight, and node categories support scanning. The graph helps orientation. The detail panel supports action.
04 / Impact
Make data confidence visible
The shared model helped analysts and administrators investigate origin, explain issues, and understand the effect of a change before making it.
Supporting evidence
Use detail where detail helps
Labels, edge weight, and node categories support scanning. The graph helps orientation. The detail panel supports action.
Make data confidence visible
The shared model helped analysts and administrators investigate origin, explain issues, and understand the effect of a change before making it.
Bring me the difficult part.
Complex workflows, AI trust, and enterprise systems.
Discuss a product challenge