Experience Builder
A web-based tool for creating and managing step by step experiences, enabling non-technical users to build interactive content without coding.
Role:
Product Designer at CareAR® (contract)
Activities:
Research, UX&UI design, Prototyping, User testing
All designs, concepts, materials, and solutions presented in this case study are the exclusive property of CareAR® and are protected under applicable intellectual property and copyright laws.
The problem nobody was talking about yet
It started with a different project. While working on CareAR's core AR remote assistance product, a pattern kept surfacing in conversations with enterprise clients: the content problem. Field technicians were showing up to jobs with printed PDFs, outdated manuals, or nothing at all. The knowledge existed somewhere - in someone's head, in a Word doc, in a training video from years ago - but there was no way to turn it into something a technician could actually use on-site.
The opportunity was too strong to ignore. What if we built a tool that let anyone - technical or not - create rich, interactive, AR-powered work instructions? No developers required. No specialized software. Just a content creator and a blank canvas.
That's how Experience Builder was born. And I was brought in to design it from the ground up.
Understanding the people in the room (and in the field)
Before I touched a single wireframe, I needed to understand two very different worlds.
The creators - technical and non-technical content writers responsible for translating expert knowledge into instructional content. They were used to working in tools like PowerPoint, Word, or basic LMS platforms. Some were highly technical, others had never built anything interactive in their lives.
The consumers - subject matter experts and in-field technicians who needed to act on that content, often in challenging environments: loud factory floors, tight server rooms, customer homes. They needed clarity fast. No room for ambiguity.
I ran discovery interviews with both groups. With content creators, I wanted to understand their existing workflow: How do you create instructional content today? What does the handoff look like? Where do things break down? With field users, I focused on context of use: What does a bad instruction experience cost you? What makes you trust a guide?
The research surfaced a sharp tension. Creators wanted flexibility and control. Field users wanted simplicity and speed. My job was to design a platform that could satisfy both - without letting either side compromise the other.
Interview findings
Service blueprint
Processes comparison
The core design challenge: power without complexity
Building a no-code tool for non-technical users is one of the harder UX problems to crack. Give people too little control and it feels like a toy. Give them too much and the learning curve becomes a barrier that kills adoption.
The Experience Builder needed to support a rich content palette - text, images, video, 3D models, AR, forms, object detection, custom navigation flows - while remaining approachable enough that someone who had never built a digital experience could publish their first one in an afternoon.
My approach centered on three principles that I held throughout the design process:
Progressive disclosure. Core actions - add a page, add content, connect pages - are front and center. Advanced capabilities like machine learning verification and iframe embeds live one layer deeper, accessible when you need them without cluttering the default experience.
Template-first creation. Rather than starting with a blank canvas (which research showed caused anxiety and early drop-off), we designed a template system that gives creators a working starting point. Templates aren't rigid - they're editable scaffolding that communicates "this is how a good experience is structured" while leaving room for customization.
Constrained customization. One of the most deliberate decisions in the design was how much styling control to expose. Too little and enterprise clients couldn't build on-brand experiences. Too much and we'd become a design tool - which nobody on either user group wanted to operate. We landed on a system where brand parameters (color, logo, basic layout) could be set at the organization level, then inherited across all experiences. Creators focused on content. Brand consistency handled itself.
From clicks to confidence: validating with real users
Designing in the abstract only gets you so far. Throughout the MVP design phase, I ran 1:1 usability testing sessions online with clickable prototypes - putting designs in front of actual content creators and watching where they hesitated, where they got lost, and where they moved confidently.
These sessions were where the real design happened. Early iterations of the workflow builder - the feature that lets creators connect pages into branching flows - were technically correct but visually overwhelming. Users would understand the concept after explanation, but couldn't intuit it on their own. We simplified the connection model and introduced visual affordances that made the logic feel more like connecting dots than writing logic. Subsequent test rounds showed a marked improvement in first-time task completion without guidance.
The result: adoption that kept climbing
We measured adoption for five months post-launch, tracking daily active users, session behavior, content output, and return usage. The numbers told a clear story.
User growth held steady at +18% month over month for five consecutive months - not a launch spike, but consistent organic pull driven by word spreading within enterprise teams. By the end of the measurement period, creators had published 448 experiences in total, with a 43% completion rate from started to published - meaning nearly half of everyone who opened the builder followed through to shipping something real. In a no-code tool aimed at non-technical users, that's a meaningful signal that the UX removed enough friction to get people across the finish line.
The usage pattern was the most telling metric of all. Return users were coming back an average of 3 sessions per week, each lasting around 17 minutes. That's not someone exploring a new tool - that's a professional integrating it into their weekly workflow. Experience Builder hadn't just launched successfully. It had become a habit.
What I'd take forward
This project sharpened my thinking on designing for creator tools specifically - a category where the user's output is the real product, not the tool itself. When the end goal is a field technician successfully replacing a printer component or setting up a smart modem without a service call, every design decision in the builder has a downstream consequence you never directly see. That invisible accountability made this one of the most challenging and rewarding projects I've worked on.