Rule Builder．2025

Rule Builder for Non-Technical Users

I led the 0→1 UX for a B2B payment recovery product, turning complex backend logic into a no-code rule builder business teams could use without a developer. The result made rule creation faster and more confident for first-time users.

Disclaimer:
This project was completed under a non-disclosure agreement. I've kept the client's confidential data and proprietary details out of this case study, and the views here are my own.

Team

- Director of Marketing
- Principal Architect
- Co-founder

DURATION

14 weeks

Deliverable

- UX Strategy
- Affinity Map
- Workflow
- Persona
- Information Architecture
- User Flows
- Prototype
- Usability Testing

Tool

- Figma
- ChatGPT
- Perplexity
- Gemini
- Claude

Impact

~50% faster
Rule creation time
Higher confidence
Clear status & confirmation steps
Built UX foundation
Built a discovery-to-test workflow in a low-maturity org
Human and robot hands touching fingers through a glowing digital interface with icons and charts.

The Challenge

The team was launching a B2B web product and wanted business teams to manage complex payment recovery rules on their own.
The original framing was broad: make payment churn recovery more proactive and easier to manage.
Through discovery, I helped refine the challenge into a more specific product problem:
How might we help revenue teams edit payment-churn rules confidently without engineering support?

Building a UX Foundation

This team was new to working with UX, so I created a UX foundation deck to align on the strategy, goals, and deliverables. It established a shared language early and served as a reference point for key artifacts and decisions throughout the project.
Venn diagram titled UX Success showing three overlapping circles labeled Business Goals, User Needs, and Technical Constraints with UX in the intersection, explaining key questions for each area.

Understanding the Problem

I ran stakeholder interviews to understand why this mattered, who'd use it, and what would constrain delivery, then clustered the input into themes with an affinity map.
Workspace board layout with sections on user design, key workflows, adoption and learnability, system integration, design constraints, and success measurement, featuring clustered sticky notes in pink, yellow, blue, and purple.
Interview Synthesis:
  • The team explored turning backend expertise into a product offering that helps business teams manage recurring customer pain points.
  • Target users span SMB and enterprise teams who currently rely on existing tools to cover parts of the workflow.
  • Success meant proving value and viability fast, while aligning a team new to UX.

Workflow Transformation

Next, I mapped the current workflow and the desired future workflow.

1. Original Workflow

Workflow diagram showing original process stages from Issue Detection to Visibility & Governance with description and roles of Owner and Support at each stage.
A mostly linear flow with a default response, limited personalization, and manual exception handling later in the process.

2. Future Workflow

Future workflow diagram illustrating steps from Issue Detection to Visibility and Governance with areas of ownership and support in Automation, Revenue Ops, Data/Analytics, Marketing Ops, and Engineering Support.
A “classify & route” step was introduced to handle cases by scenario instead of pushing everything through one default flow.
Business teams could make routine changes through the UI, and engineering could concentrate on integrations and reliability.

Who I Designed For

This work also clarified who the first release should support. Since outreach is executed through a CRM, I prioritized Revenue Ops as the primary user and focused the first phase on their core tasks.
Portrait of a middle-aged man with short hair and beard wearing a light blue button-up shirt in a bright office, alongside text describing the role, pain points, workflow, design implications, and success criteria of a Revenue Ops Manager (Workflow Owner).

Benchmarking the Space

I reviewed publicly available tools via marketplace apps (Shopify ecosystem), Mobbin, and AI-assisted scanning to identify relevant competitors. I captured recurring patterns and standout approaches to inform our IA and core user flows.

01

Rule builders show results for each rule and each step to help teams tune.

02

Dashboards show both rates and impact, then break results down by scenario and reason so teams know what to fix next.

03

Templates give a reliable starting point and speed rule setup.

From Workflow to Product Structure

Once the future workflow was aligned, I translated it into product structure.
I created the information architecture based on core workflows and roles, then validated it against competitor patterns and technical constraints with engineering.
That process helped define the first release around two priority areas:
  • Rulesets
  • Rule Building
Website sitemap diagram with main sections Home, Overview, Recovery Flow, Customers, Analytics & Insights, Campaigns & Communications, Integrations, and Administration, each branching into detailed subsections.
This turned a broad product idea into a more buildable and focused experience.

Exploring the Rule Builder

The core design challenge was how to let non-technical users create and edit complex rules without introducing ambiguity.
I explored two directions before landing on a third.
Diagram of two connected blocks with text lines, representing a node structure or rule builder interface.
Node based flow builder
Powerful, but too complex for business teams to edit on their own.
Interface mockup depicting a search bar at the top, followed by a shaded content box with two lines of text, and a rectangular button aligned to the bottom right.
Free text to rule
Familiar and appealing, but risky. It would require guided input, strong validation, and extra engineering effort to keep rule translation consistent and safe.
To balance usability, correctness, and delivery constraints, I designed a block and dropdown based builder. This made available options explicit, reduced ambiguity, and guided users toward valid configurations.
A grid of seven rounded rectangular blocks arranged in three rows with empty gray placeholder bars inside each block.
This was the key design decision of the project: prioritize clarity and reliability over maximum flexibility.

Clarifying Steps & Edge Cases

Structured blocks kept rule creation deterministic and easy to validate, while free text added ambiguity, so I documented the block-based experience in a user flow, including key edge cases.
User flow diagram illustrating the process of creating, editing, reviewing, approving, testing, and activating rules with decision points and edge cases annotated.

Prototyping and Testing

After building a high-fidelity prototype in Figma, I ran moderated usability testing with five participants to see whether business users could create and edit rules confidently. Each person completed the same task on both the original design and the redesign, so I could compare the two directly.
In the first round, participants could finish the task, but they spent too long working out how to build a valid rule. Even when they said it felt easy, I watched the same pattern across the group: hesitation, backtracking, and uncertainty about whether they might publish something incorrect.

What I Changed

Based on testing, I redesigned the experience into a more guided, step by step builder.
I introduced:
  • Clearer progression through the flow
  • More explicit status and structure
  • Confirmation moments to reduce fear of mistakes
  • Stronger guidance around what users could safely do next
A horizontal process flow diagram with four white rounded rectangles connected by right-pointing arrows on a light gray background.
Instead of asking users to assemble logic more freely, the final concept scaffolded the experience so the product felt safer and easier to trust.
In follow-up testing with the same five participants, the guided builder cut rule-creation time by about half. They also reported feeling more confident, mainly because the clearer structure and confirmation steps reduced the fear of publishing something wrong.

Trust but Verify

This was my first 0→1 project using AI deeply, so I treated it like a research teammate rather than a decision-maker. I used AI to accelerate synthesis, market scanning, and ideation, then validated outputs through source checks, stakeholder review, feasibility constraints, and user feedback.
To reduce errors and bias, I cross-checked results across multiple AI systems and primary sources, and I used AI to critique my interview scripts and facilitation between testing rounds.
Diagram showing an AI framework with three stages: Generate by AI (draft interview guide, summarize notes, competitor analysis, UI variations), Verify by Human (source & date checks, stakeholder validation, constraint checks, user feedback) with AI cross-check, and Apply by persona + workflow (IA + user flows, edge cases, final screens + prototype), with a feedback loop from Apply to Generate labeled Learnings from Apply.

Impact

Faster rule creation
Follow up testing showed about a 50% improvement in time to create a rule.
Higher confidence for first time users
Users felt safer moving through the flow because progress, status, and confirmations were more explicit.
Stronger product foundation
The project translated a messy backend process into a clear structure for future product development and cross functional alignment.

Reflection & What's Next

This project reinforced how much confidence matters in workflow-heavy products. Getting users through the steps is one thing; getting them to trust what they're about to do is the harder part.
It also showed me how much impact design can have in 0→1 environments, where the work extends past the interface into workflow, scope, and the shared understanding behind a product.
Next, I’d define a lightweight measurement plan (adoption, rule error rate, time-to-change) and continue strengthening governance UX (approvals, versioning, rollback) to support sustainable business ownership.
Go to previous case study

AI-Native Design Workflow

Two features from concept to tested prototype

Scalable UX for Enterprise API Tools

Streamlined complex workflows for technical users
Go to next case study
Copyright ⓒ 2026, Sin Ko.
All Rights Reserved.
SubstackMediumLinkedInVimeo
Back to the top arrow