Client: Confidential Global CPG Company
My Role: Lead AI Product Designer
Duration: 2 months

From MVP to Enterprise Platform: Designing for Scale and Adoption

A quick note: I’ve designed multiple AI-powered innovation engines for different global brands. While the MVP structures share a similar foundation, each evolution is tailored to the client’s specific needs, workflows, and innovation goals.

The Challenge

We first built an MVP AI innovation engine in a four-week sprint to prove the concept and secure funding for future development. With funding in place, we entered the next phase: refining the user experience, expanding functionality, and improving process efficiency. This case study covers that evolution, as we transformed the MVP into a scalable platform for multi-brand innovation and collaboration.

My Role

As Lead AI Product Designer, I evolved the MVP into an enterprise-ready platform centered on user trust, explainability, and multi-brand adoption. I redesigned the experience from a linear workflow into a dynamic, all-in-one workspace where users could generate, compare, and refine opportunity spaces and concepts in a single flow. I also introduced an image studio to enhance visualization and give the user more control in the creative process. The platform advanced from an MVP to a highly intuitive, multi-flow workspace, enabling faster, clearer, and more collaborative innovation across teams and brands.

Designing for Adoption: from MVP to V2

This phase transformed the early MVP into a multi-flow, multi-brand workspace built for real-world use. I focused on expanding functionality and refining the UI to make the system more transparent, flexible, and aligned with how teams actually work:

CORE FEATURE 1

Linear to Dynamic Flow

I redesigned the previous linear flow into a dynamic workspace, letting teams generate, compare, and refine concepts faster and in parrallel in one intuitive flow.
CORE FEATURE 2

Brand Expansion

Expanded the platform from single-brand to multi-brand functionality and introduced an Insight Studio that revealed more of the data and logic behind each opportunity space.
CORE FEATURE 3

Image Studio

Introduced tools to regenerate concept images aligned with each brand’s unique look, feel, and product formats for testing.

Feature 1:
From Linear to Dynamic Flow

The original MVP followed a rigid, single-path flow. The redesign transformed it into a dynamic workspace where teams could explore, compare, and refine multiple concepts simultaneously.

This shift mirrored how innovation teams actually work - jumping between ideas, testing variations, and iterating quickly.

The UI now supports parallel exploration, side-by-side comparison, and a smoother creative loop within one unified environment.

Feature 2: Insight Studio & Brand Expansion

Scaling the platform for multi-brand use required a major backend restructuring of how brand data was stored, vectorized, and parsed to surface brand-specific insights. This foundation powered the Insight Studio, which revealed the proprietary data and reasoning behind each opportunity space.

On the front end (my domain - thankfully), the challenge was designing a UI that made it easy for users to switch between brands, and ensure brand specific insights were displayed in a structured, modular way to later accommodate user input and interaction in future versions.

Feature 3:
Image Studio

To support faster testing and visualization, I designed an Image Studio that allows users to regenerate concept imagery based on each brand’s unique look, feel, and product formats.

The feature bridges creative exploration with brand consistency, helping teams generate visuals that reflect real packaging and category aesthetics, accelerating concept testing and cross-team communication.

Outcome & Purpose of MVP

Version 2 validated the platform’s ability to scale innovation across brands and workflows while maintaining usability and transparency. The introduction of multi-brand functionality, a redesigned workspace, and deeper data visibility proved that the system could handle more complex data and simultaneous flows without sacrificing accuracy.

This success secured leadership funding for V3, enabling us to deepen AI reasoning, expand data coverage, and move toward a fully integrated, enterprise-grade innovation ecosystem.

Related case studies