SafeBikely - 12x user growth through a ground-up redesign

SafeBikely - 12x user growth through a ground-up redesign

(Role)

(Role)

UX Designer

UX Designer

(Client)

(Client)

SafeBikely

SafeBikely

(Year)

(Year)

2020

2020

SafeBikely had a real product and a real market — smart bike lockers deployed across Norway, targeting commuters, workplaces, and cities. The app connecting users to those lockers was holding the business back. Low adoption, poor performance, and a UX that had never been properly designed from the ground up. I came in to fix that.

SafeBikely had a real product and a real market — smart bike lockers deployed across Norway, targeting commuters, workplaces, and cities. The app connecting users to those lockers was holding the business back. Low adoption, poor performance, and a UX that had never been properly designed from the ground up. I came in to fix that.

Before

Before

After

After

The Challenge

This wasn't a cosmetic redesign. The existing app had structural problems — unclear navigation, a registration flow that created unnecessary friction, and no coherent information architecture underneath it. The goal was to rebuild the experience around the user's actual job: find a locker, book it, get in, get out.

Process:

The redesign ran through weekly design sprints in a cross-functional team — Product Owner, UX Director, and developers in the loop throughout. I started with market and competitor analysis to understand what was working elsewhere in mobility apps, then rebuilt the information architecture before touching a single screen.

The central design tension was familiarity versus improvement. Users who already had the app had built habits around its quirks. A redesign that moved too fast would alienate them; one that moved too slowly wouldn't solve the underlying problems. Sequencing what to change — and what to leave alone — shaped most of the early decisions.

From there: sketches, team feedback, iteration, and high-fidelity prototypes for both iOS and Android before handoff.

The redesign ran through weekly design sprints in a cross-functional team — Product Owner, UX Director, and developers in the loop throughout. I started with market and competitor analysis to understand what was working elsewhere in mobility apps, then rebuilt the information architecture before touching a single screen.

The central design tension was familiarity versus improvement. Users who already had the app had built habits around its quirks. A redesign that moved too fast would alienate them; one that moved too slowly wouldn't solve the underlying problems. Sequencing what to change — and what to leave alone — shaped most of the early decisions.

From there: sketches, team feedback, iteration, and high-fidelity prototypes for both iOS and Android before handoff.

What changed:

The redesign introduced a streamlined booking flow directly from the map, phone-number authentication replacing a friction-heavy registration process, and three new navigation features — parking history, saved locations, and an in-app help section with live chat — each addressing a distinct gap in the original experience.

The redesign introduced a streamlined booking flow directly from the map, phone-number authentication replacing a friction-heavy registration process, and three new navigation features — parking history, saved locations, and an in-app help section with live chat — each addressing a distinct gap in the original experience.

Onboarding

Onboarding

Onboarding

First-time users needed to understand a physical product — bike lockers — through a digital interface. The onboarding flow was designed to close that gap fast, without over-explaining.

First-time users needed to understand a physical product — bike lockers — through a digital interface. The onboarding flow was designed to close that gap fast, without over-explaining.

Registration & Login

Registration & Login

Registration & Login

Replaced a form-heavy registration process with phone number authentication via SMS — fewer fields, less cognitive load, faster path to first use.

Replaced a form-heavy registration process with phone number authentication via SMS — fewer fields, less cognitive load, faster path to first use.

Navigation & Search

Navigation & Search

Navigation & Search

Added three features absent from the original app: parking history for managing past reservations, saved locations for frequent spots, and a help section with live chat for in-locker issues. Search was rebuilt to surface nearby availability without extra steps.

Added three features absent from the original app: parking history for managing past reservations, saved locations for frequent spots, and a help section with live chat for in-locker issues. Search was rebuilt to surface nearby availability without extra steps.

Booking

Booking

Booking

Reduced the reservation flow to the minimum viable number of steps — directly from the map, with real-time visibility into duration and cost before confirming. No surprises at checkout.

Reduced the reservation flow to the minimum viable number of steps — directly from the map, with real-time visibility into duration and cost before confirming. No surprises at checkout.

Impact

Following launch, the registered user base grew from 2,000 to 25,000 across Norway — a 12x increase. Engagement and retention metrics improved significantly, with users adopting the app as a primary parking solution rather than a last resort.

Following launch, the registered user base grew from 2,000 to 25,000 across Norway — a 12x increase. Engagement and retention metrics improved significantly, with users adopting the app as a primary parking solution rather than a last resort.

Project takeaways

Project takeaways

Project takeaways

The process was sprint-driven and team-validated, but leaned heavily on internal feedback rather than continuous input from real users. The results justify the approach, but I'd instrument earlier-stage user testing more deliberately today — particularly around the booking flow, where assumptions about "minimal steps" could have been pressure-tested before high-fidelity work began.

The process was sprint-driven and team-validated, but leaned heavily on internal feedback rather than continuous input from real users. The results justify the approach, but I'd instrument earlier-stage user testing more deliberately today — particularly around the booking flow, where assumptions about "minimal steps" could have been pressure-tested before high-fidelity work began.

25 000+

25 000+

25 000+

Users registered

Users registered

130 000+

130 000+

130 000+

Rentals

Rentals

25+

25+

25+

Cities covered

Cities covered

Always open to interesting conversations — about design, work, or whatever's on your mind.

Nikola Topalovic

© 2026

Always open to interesting conversations — about design, work, or whatever's on your mind.

Nikola Topalovic

© 2026

Always open to interesting conversations — about design, work, or whatever's on your mind.

Nikola Topalovic

© 2026