Anonymized Case Study

Engineering platform

A clearer path through complex document work

As UX/UI Designer, I researched professional document work, designed a staged prototype, and tested it with small samples. The work produced a high-fidelity prototype and implementation-facing UX guidance.

Situation
Dense metadata, mixed information levels, and established expert conventions made document work difficult to modernize.
My role
UX/UI Designer responsible for research, interaction design, prototyping, validation, and development alignment.
Contribution
  • UX design
  • User research
  • Prototyping
Focus Area
Complex operational UX
Period
March–June 2026
Verified change
Small-sample prototype validation supported a staged creation workflow and gave product and development stakeholders direction for implementation decisions. It does not establish a production outcome.

Early evidence

A staged prototype clarified complex document work

This recreated workflow presents the proposed structural change without reproducing client UI. In a two-participant prototype check, both people rated the staged direction highly for clarity and confidence and said they could complete it independently.

Starting experience

One task, many competing fields

Recreated design direction

Four clear stages

  1. 01ProjectConfirm inherited information
  2. 02DocumentDefine document information
  3. 03FileMake naming consequences visible
  4. 04ReviewCheck before completion
Recreated Validated prototype Invented labels and styling; no original client interface is shown.

Decision atlas

Three decisions clarified the work without removing expert control

The direction staged creation, made reuse easier to find, and separated shared defaults from exceptions and access states.

  1. 01 Stage creation instead of flattening metadata Overload → four stages → directional validation
  2. 02 Make reuse easier to find Expert knowledge → search → clear starting paths
  3. 03 Separate defaults, exceptions, and access states Mixed information → explicit inheritance → clearer decisions

Consequential decision

Stage creation instead of showing everything at once

The problem

Professional users could handle detailed information, but the existing flow presented overlapping names, numbers, and metadata together without a clear order or explanation of their consequences.

Evidence and reasoning

Three initial interviews surfaced overload, duplicate entry, and uncertainty about which values affected the final document. The need was clearer sequencing, not fewer expert decisions.

Design response

I designed a four-step process for data entry: project, document, file, and review. In the proposed flow, file naming was visible, duplicated entry was reduced, and people could see how each selection affected the process before completion.

Proof

In a two-participant prototype check, clarity was rated 4/5 and 5/5; both participants rated confidence 5/5 and recommendation 10/10. Initial recommendation ratings were 7/10, 6/10, and 8/10 across three interviews. These small samples indicate direction only; they are not statistically meaningful.

See the recreated workflow and ratings ↑ Validated

Consequential decision

Make reuse easier to find

The problem

Finding a starting file depended heavily on remembering technical names and a large folder tree, even though work could begin in several ways.

Evidence and reasoning

Three cross-discipline validation sessions indicated that people could retrieve files using familiar names and numbers. They also reinforced that reuse was a routine part of the workflow and that a complete information-architecture replacement was not the immediate priority.

Design response

I kept the familiar structure and added search, filters, useful context, and clear paths for templates, existing files, and previously created documents.

Proof

The recreated retrieval model shows the proposed change: familiar names and numbers become search inputs rather than the only route through the structure.

Known clueName, keyword, document or drawing number
Narrow resultsDiscipline · application · language
Choose a starting pathClean · Content-rich · Existing · Previously created
Recreated retrieval logic with invented labels
Recreated

Consequential decision

Separate shared defaults, exceptions, and access states

The problem

Project and document information were mixed, while exceptions and restricted projects needed different handling.

Evidence and reasoning

Domain interviews and two phase-four validations supported clearer project/document separation and familiar preview-and-open behaviour. Access rules still required product and security decisions.

Design response

I proposed shared project defaults, controlled document exceptions, separate technical and output data, and a limited-access preview.

Proof

The recreated information model distinguishes the proposed data path from the access decision that remained with product, security, and development stakeholders.

Recreated information and access-state model with invented labels
Recreated

Results

What changed

The work produced and checked a proposed direction for document creation, template retrieval, project navigation, metadata, permissions, and reusable interaction patterns. Research findings and prototype checks gave product and development stakeholders a clearer basis for implementation decisions.

Evidence boundary

What this does not establish

The samples were small and may include repeat participants. They do not establish production status, adoption, business impact, time savings, or reduced errors. All visuals are recreated, not original client material.

Collaboration and constraints

UX direction shaped with the team, close to implementation

My responsibility

I planned and conducted interviews, synthesized findings and requirements, designed and maintained interactive prototypes, ran validation, documented decisions, and prepared implementation-facing UX guidance across four phases.

Wider team

Product stakeholders set priorities, developers assessed feasibility and owned implementation, domain experts explained professional conventions, users tested proposals, and security and access stakeholders owned policy decisions.

Material constraints

The design had to serve occasional and frequent users, preserve useful legacy conventions, scale to large project structures, respect confidential access states, and improve incrementally while implementation planning was already moving.

Start a conversation

What has become fragmented, difficult to use, or hard to decide?

Tell me briefly about the situation. I’ll reply with an initial perspective and a sensible next step.

Email Sadi