UX case study

Designing trust into an AI that builds your software

FieldWorrk collapses requirements-gathering, architecture, and build into one AI-guided conversation. My job as lead designer wasn't making the AI smarter — it was making a risk-averse enterprise user comfortable enough to click "Approve" on what it produced.

Role Lead Product Designer Team 1 PM · 2 designers · 6 eng · 1 researcher Timeline 5 months, discovery to GA Product FieldWorrk
58 → 76 SUS usability score, round 1 vs. round 2
63% → 92% Task completion after redesigning the brief
48% Of started solutions reached Test, first GA cohort

The brief

Leadership's ask was blunt: enterprise clients want to go from business idea to working app without waiting six weeks for a requirements document. Can AI collapse that timeline — and can we make people trust it enough to actually use it?

The second half is what this case study is about. Building an AI that drafts a data model in nine seconds is a research problem. Getting a risk-averse ops manager to click "Approve" on that model without personally re-deriving it is a design problem.

Inside FieldWorrk

High-fidelity, working recreations of the four screens — not screenshots. Try them the way the research below describes them being used.

fieldworrk.app — discover
1Discover
2Reimagine
3Design
4Build
5Test
A
"...Society Manager should be able to view the list of requests and approve or reject, with a comment..."
11:31 AM
I've captured the initial structure. Check each field on the right — correct anything that's off.
11:31 AM

Discovery Snapshot

Updated
Use case Club Management System
Business outcome Eliminate paper-based records
Systems CustomerFacingApp, ClubAdminApp
Data MembershipRequest, Event
Teams SocietyMember, SocietyManager
Try it — send a message, or click ✎ on any field

This is the round-2 fix: five falsifiable fields instead of one AI-written paragraph. Editing one doesn't touch the others.

fieldworrk.app — reimagine
Discover
2Reimagine
3Design
4Build
5Test

I've updated the process with the sub-steps and automated notifications you described. Compare both versions below.

3 steps·manual, ~4 days
Try it — switch tabs above

Same process, before and after. A process owner sees exactly what the AI proposes to automate before approving anything.

fieldworrk.app — design
Discover
Reimagine
3Design
4Build
5Test
0 of 5 designed Assumptions logged: 0
Try it — click any step to expand its sub-steps

Every expansion is where a judgment call could hide. That's what the assumptions counter is tracking as each step gets designed.

fieldworrk.app — build
Discover
Reimagine
Design
4Build
5Test
Try it — filter, then retry a failed build

Errors explain themselves in place and offer a next action — no separate error console to go find.

01Discovery

Talking to the people who'd actually use this

Before touching a screen, I ran 14 contextual interviews with business analysts, process owners, and citizen developers across our banking and retail pilot accounts, focused on how they currently move a process from idea to deployed system, and where it breaks.

By the time I've written the requirements doc, had it reviewed, and handed it to the dev team, the business has usually changed their mind about half of it.— P3, Business Analyst, Banking
I don't trust a system that just tells me "done." I need to know what it assumed, because I'm the one who gets asked questions six months later when something breaks.— P7, Process Owner, Retail Ops
Every low-code tool I've used either treats me like I can't understand the technical side at all, or dumps raw JSON on me. There's no middle ground.— P11, Citizen Developer

Three principles came out of this and shaped every decision after:

Principle 1

Never say "done" without saying based on what. The system has to show its reasoning, not just its output.

Principle 2

Keep the conversation and the record of truth separate. No one should have to re-read chat scrollback to know what was decided.

Principle 3

Meet people at their actual technical level — readable summaries by default, technical detail on demand, never the reverse.

02First usability test

A single AI-written paragraph wasn't good enough

The first prototype covered only the Discover phase: chat on the left, one free-text AI summary on the right. We ran a moderated test with 8 participants, each describing a real process from their own job.

Task completion5 / 8 participants
62.5%
System Usability Scaleindustry average is 68
58.4 / 100

Three participants independently said some version of not being able to tell what the AI actually understood versus what it was just repeating back.

This just looks like it rephrased what I said. I don't know if it actually extracted anything or if it's just... summarizing.— P4, usability session

This was the most important finding of the project. The problem was never the AI's accuracy — it was legibility of its reasoning. Users couldn't distinguish "the AI understood this" from "the AI echoed this."

03Design response

Turning one paragraph into five falsifiable fields

We broke the free-text summary into a Discovery Snapshot — discrete, independently editable fields. Not a visual redesign so much as an epistemic one: forcing the AI's understanding into named slots a user could individually confirm or correct.

Use caseClub Management System
Business outcomeEliminate paper-based records
SystemsCustomerFacingApp, ClubAdminApp
DataMembershipRequest, Event
TeamsSocietyMember, SocietyManager
Spotlight — the assumptions registry

Built from a single interview line

P7's comment about being "the one who gets asked questions six months later" is a liability problem wearing a UX problem's clothes. So every time the AI made a judgment call rather than working from explicit input, it logged it — and the running count surfaced right in the chat.

Tested with 6 participants in governance-heavy roles: 5 of 6 said the visible count alone increased their willingness to approve a design without opening every assumption. But one gap surfaced too:

I want to trust the number, but numbers I can't check aren't audits, they're just... vibes.— P2, usability session
Shipped without drill-down — prioritized next
04Retest

The structured brief, retested

Same task, 8 returning plus 4 new participants (n=12), three weeks later.

Task completionround 1 → round 2
62.5%
91.7%
SUS scoreround 1 → round 2
58.4
76.2

Median time to a positive trust comment moved from 3:40 (negative sentiment) to 6:55 — and the sentiment flipped from suspicion to "oh, I can just fix this field."

Users didn't need the AI to be more accurate. They needed a surface for disagreement. Editable, discrete fields gave them that surface.

05In the wild

What 312 real solutions told us

Six weeks into general availability, we pulled a phase-completion funnel across every solution started by early-access clients.

Discover
100%
Reimagine
89.4%
−10.6%, mostly one-off explorations, not friction
Design
78.2%
Build
63.5%
−14.7%, largest drop in the funnel
Test
48.4%

Build also had 3.4× longer average dwell time than any other phase. Qualitatively, this was where the interaction paradigm silently switched from conversation to dashboard, with no transition to prepare users for it.

Build is fundamentally an async, multi-system status board, and that doesn't finish in the same conversational rhythm as Discover or Reimagine. I don't think we designed the seam between Design and Build carefully enough.— Engineering Lead, project retro

One decision I'd defend, one I'd reverse

Would defend

Separating chat from artifact

Every phase pairs an ephemeral chat with a persistent, structured record. When we asked participants to find a decision made moments earlier, 12 of 12 went straight to the structured panel, not the scrollback — even though the decision was made in the chat. That's strong evidence the split matches how people actually think about "discussed" vs. "decided."

Would reverse

Shipping Build with no bridge screen

Given the funnel data and the retro comment above, I'd insist on a small transitional moment between Design and Build — even one message reframing the shift. We treated it as a visual polish problem when it was really a mental-model transition, and no amount of icon or color work fixes that.

Known gaps going into next release