A mobile app that helps people relocating to a new city understand the real vibe of a neighborhood — not just the stats, but how it actually feels to live there. Built at the intersection of urban planning knowledge and UX research.
Before I became a UX designer, I spent a decade as a licensed architect working on urban planning projects — zoning analysis, walkability studies, neighborhood density, how built form shapes the way communities feel. When a friend told me she'd moved to a new city by choosing a neighborhood based entirely on apartment price and Google Street View, I recognized the problem immediately: the tools people use to research neighborhoods don't speak the language of lived urban experience.
This case study sits at the direct intersection of my architecture expertise and UX craft. I didn't need to research what makes a neighborhood feel safe, walkable, or vibrant — I've spent years measuring it. What I needed to learn was how to translate that spatial intelligence into a product people would actually use on a phone.
Urban planners use tools like Walk Score, PTAL (Public Transport Accessibility Level), and Eyes on the Street analysis to evaluate neighborhood quality. But these are professional tools built for professionals. No one moving to a new city knows what a "transit score of 82" means for their Sunday morning. My design challenge was translating expert spatial metrics into human-readable signals.
I recruited 6 participants aged 27–42 who had relocated to a new city in the past 3 years, or were actively planning to. I ran semi-structured interviews over video call, asking about their research process — not their wishes. I wanted to understand the actual tools and behaviours, not an idealised version.
"I spent three weeks on Idealista looking at apartments, and maybe thirty minutes thinking about the actual neighborhood. And the neighborhood is the thing that matters most."
I assumed people used mapping tools extensively and that the problem was information quantity — too many neighborhoods, not enough to go on.
People didn't have a shortage of data. They had a translation problem. Niche.com gives them a crime score and a school rating. Neither tells them whether the streets feel alive at 8pm, whether the Sunday market is the kind they'd go to, or whether they'd feel out of place. They were making life decisions based on metrics that measured the wrong things.
What it does well: Comprehensive data scores, school ratings, demographic breakdowns.
The gap: Optimised for parents with school-age children. No sense of daily texture. "A-" grade means nothing if you're a remote worker who wants walkable coffee shops and evening culture.
What it does well: Map integration, housing data, some neighborhood descriptions.
The gap: Neighborhood info is an afterthought to the listings. Description text is marketing copy, not honest characterisation. No resident voice.
Neither tool asks: what kind of person are you, and what kind of place would make you feel at home? Both assume the user wants to optimise for objective metrics. But neighborhood fit is fundamentally subjective.
The critical gap is Phase 3 → Phase 4. People have a shortlist but no reliable way to feel the difference between their options. That's where this app intervenes.
Freelance designer moving for lifestyle — lower cost of living, better weather, slower pace. She knows what she wants experientially (walkable, culturally active, not tourist-saturated) but can't find a tool that speaks that language.
Young couple expecting their first child, relocating for Tiago's new job. Their priorities are shifting from "cool neighborhood" to "good neighborhood to raise a child in" — but they're not ready to admit it's purely about schools. They still want to feel like themselves.
I am a person relocating to a new city. I need to understand the true character of neighborhoods — not just their stats — because every tool I try translates lived urban experience into numbers that don't tell me anything about whether I'd feel at home there.
This was the most architecture-specific challenge of the project. In urban planning, we talk about legibility — Kevin Lynch's idea that a good city is one you can read, one where the structure makes sense to someone encountering it for the first time. I asked myself: how do you make a neighborhood legible to someone who's never been there?
Lynch identified five elements that make cities legible: paths, edges, districts, nodes, and landmarks. But what makes a neighborhood feel right is harder to name — it's the density of activity, the presence of third places (cafés, parks, markets), the mix of uses, the scale of the street. I used this knowledge to build the vibe tag taxonomy: not arbitrary adjectives, but categories grounded in spatial typology.
User-generated + curated tags like "dog-friendly," "weekend market," "late-night cafés," "family footpaths." Voted on by residents. Each tag links to a count and a confidence score.
Neighborhood character shifts dramatically from 7am to 11pm. A "quiet morning coffee" district and a "loud Friday night bar street" can coexist. Show both.
Not star ratings. Structured prompts: "What do you love about living here?" / "What surprised you after moving in?" / "Who would love this neighborhood?" Short, honest, current.
Rather than showing metrics and asking users to interpret them, I designed a brief onboarding flow — 6 questions about lifestyle, social preferences, and priorities. The app then surfaces neighborhoods ranked by match, with plain-language explanations: "This neighborhood matches you because you said you want walkable coffee, and Pinheiros has 23 specialty cafés within a 10-minute walk."
I prioritised four flows for wireframing: the lifestyle quiz onboarding, the search + vibe filter, the neighborhood deep-dive, and the comparison view. Paper sketches first, then lo-fi Figma.
Round 1 had 5 participants who matched the primary personas — people who had recently relocated or were planning to. Round 2 added 5 participants who were not planning to move — I wanted to test whether the app was only useful in a crisis, or whether it could be useful for exploratory browsing.
3 of 5 Round 1 participants interpreted "vibe filters" as aesthetic preferences (like Airbnb style tags) rather than lifestyle-quality signals. They were looking for "instagrammable" and "industrial" — not what I intended. Redesigned the filter labels after Round 1 to use concrete behavioral anchors: "I walk everywhere," "I work from cafés," "I have kids under 5."
"94% match" made users stop questioning whether it was right for them. Two participants chose a neighborhood based purely on the match score without reading any of the resident reviews underneath. Iterated to show the score alongside the 3 reasons for it — not just a number, but a narrative.
Only 2 of 5 participants in Round 1 found the morning/evening/night toggle on the deep-dive screen without prompting. Moved it above the fold and made it more visually prominent in Round 2. Reduced missed interactions to 1 of 5.
Every participant cited resident quotes as the most valuable feature — more than the match score, more than the vibe tags. "This is what Reddit is trying to be, but organised." Expanded the resident voice section in the final design and added a structured prompt format to improve quote quality.
Added a persistent toggle between map view (spatial overview, good for understanding geographic relationships between neighborhoods) and list view (ranked by match, better for decision-making). As an urban planner, I know that both spatial and comparative information are essential — but different users need them at different times.
Vibe tags use both colour and iconography — never colour alone. Match percentages have alt text reading "94 out of 100." All interactive elements meet WCAG AA contrast ratios. The onboarding quiz supports voice input for users who prefer not to tap through multiple selections.
After Round 1, the deep-dive screen was restructured around a core insight from the research: people want an emotional preview before they want data. The final design opens with a resident quote and the top 3 vibe tags, then reveals detailed information progressively. Data is available, but it doesn't lead.
Kevin Lynch's legibility principles translate directly to app navigation. If a user can't build a mental map of the product — where they are, where they can go — they disengage. I applied the same thinking I'd use to evaluate a city block to evaluate a screen flow.
Ray Oldenburg's concept of the "third place" (not home, not work — the café, the park, the market) became the backbone of the vibe tag taxonomy. Tags aren't arbitrary: they map onto real spatial typologies that urban planners use to evaluate neighborhood quality.
I was so confident in my knowledge of what makes neighborhoods good that I almost didn't listen when users told me what they actually wanted to know. My biggest learning: domain expertise makes you a better designer, but only if you keep doing the research. It's a foundation, not a replacement.
Niche.com has more data than I could ever build into an app. But data without interpretation is just noise. My architecture background gave me the interpretive framework — the ability to say which data actually predicts lived experience. That's the differentiator, not the data itself.
"Buildings taught me that space isn't neutral — it shapes behaviour, mood, community. A good neighborhood doesn't happen by accident, and neither does a good product."