# From rough wireframes to a full marketplace

> GTS Innovative's founder had the idea, rough wireframes and three user groups. In about three months they became an AI-assisted three-sided marketplace, with a six-step AI invention workflow, NDA-based permissions and admin tools.

- Client: GTS Innovative, led by a non-technical founder
- Type: AI-assisted three-sided marketplace
- Category: Marketplace build
- Duration: About 3 months
- Status: Live at gatherthespark.com
- Stack: Three-sided marketplace · Six-step AI workflow · NDA-based permissions · Admin tools
- Role: End to end, first call to launch
- Result: 50+ screens (Delivery scope in about three months, not adoption or revenue.)
- Result: 15+ core workflows (Delivery scope in about three months, not adoption or revenue.)
- Result: 6 steps in the AI invention workflow (Delivery scope, not adoption or revenue.)
- See the site: https://gatherthespark.com
- Built by: Luka Abramović (https://lukaabramovic.com/)
- Canonical: https://lukaabramovic.com/work/gts-innovative-marketplace/

---

## The starting point

A non-technical founder had the idea, rough wireframes and three user groups. Inventors would develop their ideas on the marketplace, and service providers and investors would discover those inventions and reach out to collaborate. The flows, permissions, product logic and AI output were all still open.

## Three sides, one first release.

Wireframes show screens. They don't settle who can do what, how the product behaves between screens, or what the AI should produce. All of that had to be decided before anything could be built.

### Roles

Three user groups, each with its own journeys and permissions.

### Logic

Product logic and AI output that nobody had defined yet.

### Release

A first release that could ship without waiting for everything that could come later.

## My role

I ran it end to end, from the first call to launch. I turned the wireframes into roles, journeys and a release plan, and led design and engineering with the Adamant Code team. I also cut what would have delayed the first release: mobile app, escrow, ads and recommendations.

1. First call
2. Discovery
3. Scope & proposal
4. Solution design
5. Build lead
6. Launch
7. Iteration

## From wireframes to a release plan.

1. Turn the founder's wireframes into roles and user journeys for all three sides of the marketplace.
2. Write a release plan, and cut what would delay the first release: mobile app, escrow, ads and recommendations.
3. Lead design and engineering together, through 50+ screens and 15+ core workflows.

<VideoFigure src="/media/gts-innovative-demo.mp4" poster={gtsPoster} label="GTS Innovative demo, 1 minute 51 seconds, captions burned in" tag="▶ GTS Innovative demo · 1:51" />

## A marketplace, not a mock-up.

### Three-sided marketplace

Each of the three user groups gets its own roles, journeys and screens.

### Six-step AI workflow

An AI-assisted invention workflow, broken into six steps.

### NDA-based permissions

Who sees what follows the NDAs in place.

### Admin tools

Tools for the team running the marketplace.

## 50+ screens in about three months.

In about three months the founder's wireframes became 50+ screens, 15+ core workflows and a six-step AI workflow. That is delivery scope, not adoption or revenue.

<MetricTiles
  size="m"
  items={[
    { value: '50+', note: 'Screens. Delivery scope in about three months, not adoption or revenue.' },
    { value: '15+', note: 'Core workflows. Delivery scope in about three months, not adoption or revenue.' },
  ]}
/>

> You guys blew past the timelines, but it wasn't sloppy. We didn't rush through and then have to go back and fix a ton of things.
>
> – Kalen Thompson, Founder, GTS Innovative

## What this proves

A non-technical founder doesn't need a finished spec. They need someone to turn the idea into roles, journeys and a release plan, and to cut what can wait.