Scaling consistency across complex B2B AdTech platform
TL;DR
I led the definition and rollout of a scalable design system for a complex B2B AdTech platform at Publica by IAS. The goal was to reduce inconsistency, improve cross-functional alignment, and accelerate product delivery across design and engineering teams.
What began as an effort to standardize components evolved into a foundational system that reshaped how teams built and shipped product.
Impact
Reduced time spent recreating or searching for components
Improved design–engineering alignment through shared standards
Increased consistency across core product workflows
Established a scalable foundation for future product growth
Context
When I joined Publica, there was no formal design system in place.
Design decisions were made locally by teams, resulting in:
inconsistent UI patterns across similar workflows
duplicated or re-created components
heavy reliance on tribal knowledge
friction between design and engineering during implementation
As the product and team scaled, these issues became more costly and visible.
The problem
There was no reliable “source of truth” for product design.
Teams struggled with:
identifying existing components
knowing which patterns were approved or reusable
maintaining consistency across features
aligning design decisions with engineering constraints
Over time, this created:
fragmented user experiences
slower product delivery
increased rework in implementation
Why this mattered at scale
As the product evolved:
Multiple versions of similar components began to emerge
Maintenance became increasingly difficult
Onboarding new designers and engineers slowed down delivery
The design system itself had become a bottleneck rather than an accelerator.
My role
Product Designer, Publica by IAS
I led the design system initiative end-to-end, partnering with product and engineering to define structure, establish standards, and create a system that could scale with the product.
Strategic approach
Instead of rebuilding everything upfront, I treated the design system as a product with users, constraints, and phased adoption.
The goal wasn’t perfection, it was adoption and scalability.
Key Decisions
1. Start with adoption, not completeness
Rather than building a fully exhaustive system, I prioritized usability and immediate value to increase adoption across teams.
2. Focus on high-impact components first
We prioritized components that appeared most frequently in core workflows, ensuring early wins and visible impact.
3. Reduce, don’t expand
Instead of adding more variants, we actively reduced redundancy and simplified existing patterns to improve clarity and maintainability.
4. Design for flexibility within structure
Components were designed with clear rules but flexible properties, balancing consistency with real product needs.
5. Treat the system as evolving infrastructure
The system was not a one-time initiative, it was designed to evolve alongside active product development.
Execution
The rollout was incremental:
Audited existing components and patterns
Identified redundancies and inconsistencies
Rebuilt system structure iteratively, page by page
Introduced standards gradually to avoid disruption
This ensured the system improved without blocking product delivery.
Key challenges
1. No existing baseline
Without a shared system, even defining “correct” required alignment across teams.
2. Balancing speed vs. structure
Teams needed faster delivery, while the system required careful consolidation.
3. Adoption across disciplines
The system needed to work not just for designers, but also engineers and PMs.
Outcomes
Speed & efficiency
Reduced time spent recreating components
Faster iteration cycles for designers
Consistency
More consistent UI across core product workflows
Reduced divergence between similar features
Collaboration
Improved alignment between design and engineering
Established a shared reference point for product decisions
System foundation
Created a structured, scalable design system with:
standardized components
defined usage guidelines
centralized documentation
What changed beyond the system
This work shifted how teams approached product development:
From ad-hoc design decisions → shared standards
From isolated workflows → cross-functional alignment
From inconsistency → system-driven consistency
What I’d improve next
Strengthen integration with engineering via tools like Storybook
Improve governance to prevent future component drift
Expand documentation into decision-driven guidelines (not just usage rules)