back to work

UX RESEARCH·PRODUCT DESIGN·CONCEPT

Nitefind

Nightlife discovery without the decision fatigue.

Mobile product · 10 weeks · Sole designer

01 / why this exists

Choosing where to go out shouldn’t take longer than actually going out.

People already have plenty of ways to find bars, events, and venues. The problem I found wasn’t access to options. It was piecing together enough information to trust one.

Interviews showed people bouncing between platforms to check atmosphere, crowd level, price, photos, events, and whether a place was actually worth going to. Nitefind became an exploration of how to bring those signals together without creating another overwhelming discovery app.

4 interviews→recurring needs→1 product direction

02 / the signal

The interviews pointed to three recurring needs that mattered more than demographics: speed, confidence, and discovery.

01 speed

What’s happening right now?

Spontaneous plans made stale information especially costly. People wanted current crowd levels, events, specials, and nearby options without checking several sources.

Research synthesis excerpt: vibe, crowd size and price as top factors, participants using multiple platforms, and demand for real-time crowd and event updates — research synthesis · excerpt

02 confidence

Will I actually like this place?

Photos weren’t enough. Vibe, crowd size, price, noise level, and atmosphere helped people decide whether a venue was worth committing to.

Research synthesis excerpt: frustrations about dead or overcrowded spots, too many apps, and untrustworthy information about what is happening right now — research synthesis · excerpt

03 discovery

Show me something that fits, not another top-ten list.

Participants described places through qualities like chill, chaotic, romantic, queer-friendly, hidden gem, photo-worthy, and speakeasy rather than conventional venue categories.

Research synthesis excerpt: notable findings listing speakeasy, queer-friendly, photo-worthy, chill versus chaotic and hidden gem filters, atmosphere over popularity, and subcultural filter interest — research synthesis · excerpt

the insight

Different behaviors pointed to the same underlying need: reduce the number of decisions between “I want to go out” and “I’m somewhere good.”

03 / key decisions

The research narrowed the product before I designed it.

Three recurring needs became three product priorities.

01research signal

Vibe before category

People described the night they wanted more naturally through atmosphere than venue type.

product response

Filters centered on qualities like romantic, chill, live music, crowd level, price, and niche tags rather than relying only on traditional venue categories.

Nitefind advanced filters screen showing multiple vibe and tag filter chips — advanced filters

02research signal

What’s happening now

A good venue wasn’t useful if the information about it was stale.

product response

Crowd indicators, current events, specials, distance, travel time, and venue activity became first-class information.

Nitefind venue detail screen for The Loft showing crowd and vibe information, price, distance and tonight’s events — venue detail · the loft

03research signal

Reduce the choice

Sometimes people didn’t want better search tools. They wanted help making the decision.

product response

Quick Pick surfaced a small set of recommendations matched to location and preferences rather than returning another long list.

Nitefind Quick Pick screen with a featured recommendation and additional matches — quick pick

from requirements to structure

Structure before polish.

With the core signals defined, I mapped the product around three high-value moments: discovering somewhere nearby, evaluating whether it fit the night, and getting to a decision quickly.

Early paper sketch of the Nitefind map and discovery screen

Early paper sketch of the Nitefind venue detail screen

Early paper sketch of the Nitefind Quick Pick recommendation flow

Midnight Grid over Alt Scene

Alt Scene captured nightlife more literally: eclectic, expressive, and energetic. I liked it more in isolation.

But Nitefind was supposed to reduce cognitive load. Adding another loud visual layer worked against that goal.

Midnight Grid won because the interface needed to organize nightlife, not compete with it.

Midnight Grid mood board: black and violet, sleek, sophisticated, futuristic — Midnight Grid · chosenQuieter direction. Better product fit.

Alt Scene mood board: rebellious, authentic, eclectic, warm reds and oranges — Alt Scene · rejectedStronger personality. Weaker product fit.

Three tests. Four changes worth making.

3 participants · in person · mobile prototype

observationdesign response

People hesitated to share location with no context.: Added a plain-language explanation and a way to continue without sharing location.
Quick Pick didn’t clearly explain why a venue was being recommended.: Added visible preference and filter context so the recommendation felt less arbitrary.
Saving a venue lacked clear feedback.: Added confirmation and a dedicated destination for saved venues.
Advanced filters weren’t specific enough.: Expanded the vocabulary toward more descriptive and niche qualities.

the fix, in the product

Nitefind location permission screen with a plain-language explanation and an option to continue without sharing location — Context before permission.

Detail of the Nitefind Quick Pick recommendation card showing rating, Romantic and Ambient tags, and tonight’s programme — Show why the recommendation fits.

Nitefind bookmarks screen listing saved venues — Confirm the action and make it recoverable.

Detail of the Nitefind filters screen showing the expanded vibe and tag vocabulary — Match how people actually describe a night out.

04 / the product

Less searching. More deciding.

The final product organizes discovery around the information people said they needed before committing to a place.

find

Start with what’s nearby.

Location, crowd level, price, vibe, and distance make the map useful before opening a venue.

filter

Describe the night, not the category.

Vibe and niche filters let people narrow options using the qualities that surfaced during research.

Nitefind advanced filters screen with a readable vibe and niche filter vocabulary

evaluate

Put decision-making information up front.

Atmosphere, crowd level, pricing, travel time, events, specials, and imagery answer questions people were already checking elsewhere.

Nitefind venue detail page for The Loft with gallery, atmosphere and crowd information, pricing, events and specials

decide

When choosing is the problem, reduce the choices.

Quick Pick turns preferences and location into a short recommendation set instead of another long results list.

Nitefind Quick Pick screen with a featured recommendation and a short set of alternates

save

Keep “maybe later” easy to recover.

05 / responsive design

Same decision hierarchy, more room.

Larger breakpoints created room for imagery, events, and supporting information without changing what mattered first. The hierarchy stayed centered on answering the same question: is this somewhere I want to go?

The Loft venue page shown across mobile, tablet and desktop breakpoints

06 / looking back

What I’d do differently now.

Nitefind taught me that research doesn’t need to produce more artifacts. It needs to make the next decision clearer.

If I continued the project, I’d recruit a broader participant group and test the core recommendation model more deeply, particularly how much people trust crowd data and why Quick Pick recommends a venue.

I’d also move into rough sketches sooner. I spent too much time refining early layout directions before proving the underlying structure.

The biggest takeaway wasn’t a feature or screen. It was learning to ask whether every design decision reduces uncertainty or adds another thing for the user to process.

Early rough layout iterations: two sketch passes across three jobs to be done — early layout iterations

next project

SmiteForge

One brand, two very different platforms.

cross-platform UI · information architecture

view case study ↗