300K
Users reached in validation research
20%
Traffic in A/B test before full adoption
5
Design principles co-defined across teams
2
Age groups tested, full acceptance in both
I was part of the core team leading Hurb's visual language overhaul, a cross-functional initiative across product design, marketing, branding, research, and engineering. I co-facilitated the design principles workshop, led the audit of existing product surfaces, and owned the application of new fundamentals to product screens and components.
Hurb is expanding internationally, and the design team is in the early stages of building a design system. But product design and marketing are working from separate visual languages. The result shows up where it hurts: inconsistent platform experiences, brand identity applied differently across touchpoints, and usability problems that trace directly to the lack of a shared visual foundation. Before building a component library on an unstable base, the team decides to go back to first principles.
Visual inconsistency as a trust problem
Hurb is expanding internationally — and the design team is in the early stages of a design system. But product design and marketing operate with separate visual languages. The result shows up where it hurts: inconsistent platform experiences, brand identity applied differently across touchpoints, and usability problems that stem directly from the absence of a shared visual foundation. Before building a component library on a shaky base, the team decides to go back to first principles.
Seven methods across qualitative and quantitative research
A 4-stage cross-team workshop (introduction, ideation with do/don't exercises, principle naming, and Maslow prioritisation) defined 5 shared design principles with both product and marketing designers. The new visual direction was then validated at scale: a first impression test using Microsoft's Product Reaction Cards surveyed 300K users split into two age groups. Card sorting tested interface variations with the same profiles. An A/B test exposed the new direction to 20% of traffic across the two highest-revenue regions, tracking revenue, session value, abandonment rate, and product page engagement.
Phase 01
Principles
4-stage workshop with the full design team — brand ideation (do/don't), naming, and Maslow prioritisation dynamics.
Phase 02
Audit
Review of all purchase flow screens, ad pieces, and banners — analysing colour, type, spacing, photography, and tone of voice against the new principles.
Phase 03
Inspiration
Moodboard exercise bringing in references from outside travel — fintechs, marketplaces, real estate — to avoid converging on category defaults.
Phase 04
Co-creation
Hands-on homepage redesign exercise applying the principles — exploring new directions, identifying visual patterns that emerge naturally.
Phase 05
Validation
Online survey (300K users), card sorting, and A/B test via Optimizely against the top 2 revenue regions at 20% traffic exposure.
Phase 06
Delivery
Each designer applies the new fundamentals to their team's screens. Design system reorganised for flexibility and universality.
Principle alignment
"Objective", "Simple", and "Clean" rank top 3 across both age groups, directly mapping to principles defined at the workshop. The visual direction and the team's intentions are aligned.
Trust signal
Cleaner interfaces read as more trustworthy, which mattered a lot for a brand handling high-value travel purchases where purchase anxiety runs high.
Older audience
The new, more contemporary approach is well accepted by older users — a key concern for Hurb's existing customer base. Contemporary doesn't mean inaccessible.
Photo-led layouts outperform
Photo-led layouts show stronger engagement for key user profiles over illustration-led ones — directly shaping the primary visual direction of the new language.
FOMO mechanics generate anxiety
Countdown timers and urgency mechanics are perceived as misleading — a finding that goes beyond aesthetics and informs content and conversion strategy.
A/B test result: satisfactory
No rejection in key business metrics, with positive movement in session value and abandonment rate, justifying full adoption of the new visual language.
01 — Principles before pixels — and cross-team, not top-down.
Defined collaboratively with both product and marketing designers, using Maslow's hierarchy as a prioritisation frame. A shared language only works if both teams feel they shaped it. It's the same logic applied to design culture: agree on the invariants before touching a single frame.
02 — Look outside the travel category for inspiration.
The moodboard exercise deliberately excludes travel competitors — pulling references from fintechs, marketplaces, and real estate to avoid designing toward the travel industry's visual clichés.
03 — Validate aesthetics like product decisions.
First impression test + card sorting + A/B test. Aesthetic choices earned the same evidence standard as UX decisions, rather than being settled by internal preference.
04 — Photo-led hierarchy over illustration.
Research finding, not an aesthetic opinion: photo-led layouts show stronger engagement for key user profiles, shaping the primary visual direction of the new language.
05 — Distribute the first delivery across the team.
Each designer applies new fundamentals to their own product area — documenting new components as they go. Speed without losing ownership.
300K
Users reached in validation research
20%
Traffic in A/B test before full adoption
5
Design principles co-defined across teams
2
Age groups tested, full acceptance in both
Changing too many variables at once makes it impossible to learn from the test.
The A/B test changed colour, typography, spacing, and layout simultaneously, which made causal attribution impossible. A more sequenced approach would have built clearer understanding of what was actually driving the improvement.
A principle without a worked example is just a label.
The principles were well received, but without worked examples showing how each resolves a real design dilemma, they risk becoming decorative labels rather than decision tools. Documenting them with concrete filters, while the project is still active, is what makes them last.
"A travel brand going global can't afford to look like it was built by two separate teams who never talk to each other. Visual inconsistency is a trust problem."