Personalizing experiences based on user segmentation

As Product Design Lead, I was responsible for setting the design direction and collaborating with leadership, product managers, engineering, and research to balance user experience with business constraints. I led a team of [N] designers and drove the process from discovery through delivery.

My involvement spanned defining the research strategy, synthesizing findings into actionable design opportunities, and aligning stakeholders around a coherent product vision.

[Company] is a [description — e.g., fast-growing food delivery platform] operating across [markets]. Known for its [distinctive quality], the company built a reputation for [key differentiator] while competing in a heavily-contested market.

At its peak, [Company] served [X] million users with [key metric] orders per day.

The business faced a structural tension: growth driven by promotional spend was unsustainable. As funding conditions tightened, the cost-per-order ceiling became a hard constraint. The challenge was to maintain conversion and order frequency while dramatically reducing the promotional subsidy.

The underlying issue: the product was trained to serve promotional behavior, making it difficult for users with genuine intent to find what they actually wanted without relying on discounts.

Analysis of NPS and CSAT data surfaced a recurring tension: users with genuine intent felt overwhelmed by the promotion-first experience. Key findings:

• Users who were in a hurry skipped promotions entirely and went straight to search
• Repeated promotional messaging caused "offer fatigue" — reducing click-through rates over time
• High-frequency users (3+ orders/week) were more likely to churn when their preferred restaurants were buried beneath promo content

Usage data revealed that the majority of sessions began with one of three behaviors: search, category browsing, or nearby restaurant filtering. The home screen — heavily invested in promotional real estate — was largely skipped.

A structured heuristic evaluation identified four core design failures:

• Overloaded home screen — too many competing priorities with no clear hierarchy
• Promotion-first IA — the information architecture trained users to expect discounts, penalizing non-promotional sessions
• Hidden utility categories — everyday use cases (reorder, favorites, nearby) were buried 2–3 taps deep
• No guidance for habit-forming — no onboarding or progressive disclosure for building regular ordering patterns

Direct competitors had doubled down on aggressive discounting, creating a race-to-the-bottom dynamic. This opened a differentiation opportunity: personalization as an alternative value proposition to price competition.

Users who valued convenience, reliability, or diet-specific options were underserved by every player in the market.

We used a mixed-methods approach combining qualitative depth with quantitative validation. The progression moved from hypothesis formation through behavioral data confirmation:

Through 24 in-depth interviews, a 400-participant survey, and behavioral data analysis, we identified six distinct ordering personas:

Survey data revealed the Bargain Hunter was the largest segment by count but the Quality Seeker and Health Enthusiast showed the highest order value and lowest churn — making them the strategic priority for profitability.

We framed three design challenges as hypotheses to test:

The redesign restructured the home screen around need-state navigation rather than promotional hierarchy. Key structural changes:

• Compact promo module — consolidates all promotional content into a single scannable row, reducing visual noise without hiding value
• Curated sub-homes — personalized landing experiences per segment, surfacing contextually relevant content based on ordering history
• Reorder module — prominent placement for repeat-order behavior, dramatically reducing time-to-conversion for Convenience Seekers
• Utility filters — quality signals (ratings, prep time, dietary tags) surfaced at the browse layer, not buried in restaurant detail pages

Post-launch data showed the redesign maintained conversion rates despite reduced promotional traffic — validating the core hypothesis that need-state personalization could substitute for discount-driven engagement.

The project validated the personalization thesis — that users with non-promotional intent could be served effectively through need-state architecture. However, the broader strategic shift required a [1.5–2 year] timeline to reach profitability, which [company] did not have runway to reach.

The core lesson: design can solve the product problem, but it cannot overcome a structural business constraint. The work demonstrated what was possible; the market timing determined what would happen.