Designing Togo AI Lab's Website

Client

Agence Togo Digital in partnership with MESPTN & CEGA (UC Berkeley)

Roles

  • UX Designer

Timeline

Sept 2025 - May 2026

Overview

Togo AI Lab is Togo's national ecosystem for data science innovation and AI adoption in the public sector. Built to bridge advanced research with government policy, the platform unifies over 70 digital initiatives, sectoral digitization sprints, and talent programs into a centralized digital hub for data-driven governance.

Live site

Context

Togo AI Lab (formerly Togo Data Lab) was launched as a national strategic initiative to spearhead AI adoption, data science research, and public-sector digitization in Togo.

As UX Designer at Agence Togo Digital (ATD), I was tasked with building the initial digital presence (v1) from the ground up to establish credibility and communicate the lab's core mission. As the lab expanded rapidly — scaling to cover over 70 projects across 10+ government ministries, executing national digitization sprints, and launching educational workshops — the original site needed to evolve.

During the v2 redesign, the project scope expanded dramatically. I led the retrospective on v1, conducted research for the upgraded architecture, and presented a design direction alongside alternative proposals from the design team. The team ultimately adopted a blended direction incorporating a teammate's visual proposal, which I supported through alignment and usability validation.

The Problem

The v1 Challenge: Establishing Initial Credibility (0 to 1)

Zero Brand Presence: The lab needed an immediate, authoritative web presence to announce its partnership with UC Berkeley's CEGA, Google.org, and the Togolese Government.

Undefined Audience: The initial site had to serve foreign partners, local researchers, and government officials all at once, without yet knowing which group would drive the most traffic.

The v2 Scale Challenge: Scope Creep & Structural Bottlenecks

Within months of launch, the lab's operational scope exploded, rendering v1 insufficient:

Information Architecture Overload: v1 could not cleanly organize 70+ active projects, multiple sectoral digitization sprints (Mines, Tourism, Foreign Affairs), and community programs (Masterclasses, Hackathons).

Competing User Personas: The platform suddenly needed to serve four distinct audiences simultaneously, each with completely different goals:

  • 1. Government Leadership & Policymakers: Seeking proof of impact, ROI, and policy dashboards.
  • 2. International Partners & Donors: Seeking institutional alignment, governance, and research rigor.
  • 3. Students & Developers: Looking for hackathons, training registration, and learning resources.
  • 4. Public & Media: Looking for high-level announcements and press releases.

Content Friction: High-value updates (such as the Kara regional center expansion) were buried deep within generic page layouts.

The v1

I designed v1 of the website.

The Redesign

The design journey unfolded across four distinct phases, starting with the v1 Launch, where we established the brand baseline and delivered a 0-to-1 MVP. As the platform experienced rapid Growth & Friction in Phase 2, integrating over 70 new projects exposed critical information architecture bottlenecks. This prompted Phase 3, focused on v2 Alignment, which involved a multi-proposal sprint and a rigorous evaluation of design trade-offs. This collaborative effort culminated in Phase 4's Unified Direction, achieving team consensus and delivering a robust, scalable system equipped to handle the lab's expanding scope.

Research & IA Audit of v1

We audited v1's user flows and gathered feedback from internal stakeholders at ATD and Lab leadership:

Key Finding

Visitors were dropping off because strategic initiatives (e.g., agricultural crop yield prediction via satellite data) were mixed in with basic news updates.

Architectural Fix

Proposed separating the site into distinct functional pillars: Research & Impact, Government Initiatives, Community & Upskilling, and News/Events.

During the v2 internal review at ATD, two distinct design directions were evaluated.

Proposal A (My Concept)
Proposal B (Teammate Concept)

Final Verdict

Leadership selected Proposal B — it responded more directly to user needs and communicated information more clearly thanks to its card-based layout.

Impact & Reflection

70+

Projects spread across more than 10 different government agencies.

35+

Research and innovation projects actively turning data into tangible public solutions.

50+

Professionals in the public and private sectors upskilled in AI and data science.

500+

Students introduced to AI to build a sustainable talent pipeline.

Measurable & Organizational Outcomes

  • Scalable Foundation: Successfully created a design framework capable of hosting the lab's expanding repository of 70+ projects and 35+ active research initiatives.
  • Unified Public Presence: Delivered an authoritative digital hub for major stakeholders, including MESPTN, ATD, Google.org, and UC Berkeley.
  • Streamlined Engagement: Created clear, frictionless access points for over 500 students participating in AI workshops and developer programs.

Personal Key Takeaways

  • Designing for Product Evolution: Building v1 taught me the value of speed-to-market, while evaluating v2 highlighted the necessity of flexible design systems that can adapt to rapid organizational growth.
  • Ego-Free Design Leadership: Advocating for a proposal that ultimately isn't selected is part of mature product design. Championing team alignment and ensuring core UX principles — IA, usability, scalability — were merged into the final direction resulted in a stronger end product for ATD and Togo AI Lab.