0-to-1 product · Solo design + build
From an old idea to a live platform
A housing platform for Icelanders on the move: home swaps, homes for rent and rental requests in one place. I designed and built it alone, taking a recurring idea from concept to a live product.
Project details
01 · Overview
A recurring need became working software.
Flakkið brings three common housing needs into one platform: home swaps, homes for rent and rental requests. It was created for Icelanders living or travelling abroad who were already trying to solve these needs through scattered Facebook posts.
I designed and built the whole product myself, from the interface and responsive flows to accounts, listings, search and content management.
Why this case study matters
Flakkið shows that I can take a product from idea to deployed software and work across product structure, interface design, authentication, content management and responsive implementation.
02 · Background
Everyone was already asking, just in the wrong place.
Icelanders abroad constantly ask around in Facebook groups: who wants to swap homes this summer, who has an apartment near Alicante, or who is looking for a long-term rental in Reykjavík.
I had carried the idea for years: one dedicated place for those requests, with three clear listing types. When I finally had time between projects, I went straight to building.
PHASE 01
An old idea
A recurring need I had observed for years in Icelandic communities abroad.
PHASE 02
A solo build
I designed and developed the product at the same time, without a separate handoff.
PHASE 03
A live platform
The product launched and real listings were published.
03 · The build
Designed and coded at the same time.
There was no handoff because there was no one to hand off to. I worked between Figma and the browser, making product and implementation decisions together.
I built the experience end to end: accounts and authentication, listing creation for three listing types, search and filtering, a CMS-powered blog and responsive layouts.
This was also where I began using ChatGPT as part of my build workflow. It supported implementation and faster iteration, while I chose what to ship, reviewed the output and retained responsibility for product decisions and QA.
01
Structure
Defined three listing types and the information each one needed.
02
Design
Created the interface and refined it directly in the browser.
03
Develop
Built authentication, forms, listings, search and responsive behaviour.
04
Launch
Deployed a real product rather than stopping at a prototype.
04 · What shipped
The platform launched with real listings; impact was not measured.
The listings were real and specific: a summer swap from a flat on Óðinsgata, a penthouse in Villamartín, a sea-view apartment in Calabria and families offering their Reykjavík homes while travelling elsewhere.
These published examples show that the end-to-end flow reached production. I had not instrumented the product or defined a success metric, so I cannot claim scale, retention, conversion or broader market impact.
Three clear listing types
Users could publish a “House swap,” “For rent” or “Rent wanted” listing.
05 · Where it wandered
I had built the product before defining its destination.
I went from a long-held idea to implementation without first conducting user research, choosing a revenue model or defining a success metric. The product could launch, but I had no agreed evidence for deciding what should happen next.
Advertising, subscriptions and property services were all possible. Without criteria for comparing those options, none was a defensible next step.
The name became unexpectedly fitting
Flakkið means “the wandering.”
The name became a useful reminder: exploration helped me ship, but a product goal is what guides the next decision.
06 · Reflection
It changed how I start every project after it.
Flakkið gave me end-to-end implementation experience and exposed a planning gap. The next project, Dýrfinna, began with a client workshop, user sessions and a scoped agreement before high-fidelity design.
LESSON 01
Shipping exposed the missing question
Without a success measure, I could not evaluate whether the next release should focus on supply, discovery or monetisation.
LESSON 02
Instrumentation belongs in the plan
Analytics and feedback loops need to be chosen before launch if they are going to inform what happens afterwards.
LESSON 03
Solo building strengthened my range
The project taught me to connect UX decisions with authentication, content modelling, responsive behaviour and implementation details.