WORK
about
WORK
about
WORK
about

LOCAL GUIDE REDESIGN

Top10 Berlin

Restructuring discovery for a content-heavy local guide

Restructuring discovery for a content-heavy local guide

Research-led redesign of a local guide with 2,500+ locations. I connected behavioral evidence, content restructuring and information architecture to shape exploratory browsing and guided filtering for Berlin locals.

Product design

UX Research

information architecture

content audit

prototyping

top10 berlin case study hero image
top10 berlin case study hero image

01

·

context & research

The existing navigation conflicted with how users actually explored the city

A large content system had grown around a rigid hierarchy

Top10 Berlin contained more than 2,500 locations across 250+ curated lists. Over time, this content had accumulated within a three-level structure of topic → Top10 list → location, making orientation increasingly dependent on navigating predefined categories.

From e three-level menu to open exploration

Behavioral evidence showed friction in the navigation

Reviewing 200+ PostHog sessions showed low interaction, erratic navigation behavior and frequent exits after users opened the three-level menu. The pattern suggested a mismatch between the interface and the exploratory way users approached the content.

Research shifted the focus toward Berlin locals

Competitive analysis, analytics and seven interviews with Berliners pointed toward a stronger local use case. Users were often looking for relevant ways to spend limited free time rather than planning a traditional tourist itinerary.

02

·

discovery model

Different search intentions required different discovery paths

Open-ended exploration needed browsing, not deeper menus

Users often approached the platform without a precise destination in mind. Instead of asking them to understand the existing category hierarchy, the redesign supported exploratory browsing that surfaced options more directly.

From a 3 level menu to an open exploration

More specific intent needed progressive filtering

When users already had constraints — such as location or type of activity — a guided filter could narrow the available options step by step instead of exposing the full content hierarchy at once.

A behavioural model connected search intent to interaction design

Secondary research on information behaviour helped distinguish between exploratory and more directed search patterns. I used this distinction to define separate interaction models for general browsing and guided filtering.

turn_taking_diagram

Guided filtering

03

·

information
architecture

information architecture

The discovery model required restructuring 15 years of accumulated content

The audit exposed duplicates, outdated content and inconsistent categories

Years of editorial growth had produced duplicate, seasonal and outdated entries as well as inconsistent naming and categorisation. I mapped the existing content to understand where the structure was breaking down

Indexing and recategorisation created a more coherent system

I mapped and categorised 1,000+ entries using Google Analytics, Google Search Console and PostHog alongside the content inventory. Low- and no-traffic pages and sections were identified for removal or consolidation, while the remaining content was renamed, indexed and recategorised into a clearer structure.

The architecture had to support both current and future structures

Because the content restructure could not happen in one release, the IA needed to work during transition as well as support the intended future model. The solution was therefore designed for phased implementation rather than a clean-slate replacement.

04

·

Validation & Delivery

Conversion & Account Flows

Testing and phased implementation turned the concept into a buildable product direction

Moderated testing evaluated browsing and filtering flows

Interactive prototypes were tested with users to evaluate whether the proposed browsing and filtering logic supported easier orientation and decision-making

Seven key screens were developed from low to high fidelity

The work progressed from structural exploration into detailed interface design across seven key screens, covering the main browsing, filtering and location-discovery interactions.

The design was prepared for phased implementation

The final work included prototypes, interface specifications and an implementation direction that accounted for the gradual restructuring of the existing content system.

Designed for a wider view!
The remaining sections show detailed product flows and process artifacts. For the best reading experience, please continue on a larger screen.