What it’s about
Being a cyclist myself and having worked in the cycling industry for years, I know the love and passion for bikes — from myself, my colleagues, and fellow bike peers. Being proud of “your ride” is as typical for the community as “n+1” being the perfect number of bikes to own. I’d had this community idea for a while and wanted to see how far I could take it.
It was a straightforward exercise in building with available tools — with self-imposed limitations.
My self-imposed challenge: from my startup experience, I know that in this kind of environment there are usually at least two constraints — time and money (you could add development resources as a third). So I posed myself the question: How far can you get with free or very limited financial investment, in as little time as possible, while staying customer-data sensitive from the start?
This app was an exercise in building first, and it became a minimal-data community alternative for people who aren’t on Instagram or are looking for one.
Building it was genuinely fun — trying out new things and learning a lot along the way. So I decided to give it its own domain and see where it goes from here.
Approach
- Started with Lovable on the free tier: first draft, first version of the backend, initial RLS, basic app layout.
- The first version got built out pretty far, but I also hit the limits often. It became clear that at some point a) the app would have to leave Lovable, and b) owning my own infrastructure would become a topic.
- Waiting times: Lovable’s daily credit counter resetting (5 credits a day is really, really limited) left waiting time I could use on other AI tools — e.g. working on design with Google Stitch, exploring alternate approaches, and discussing the future tech stack and migration options.
- Migration to my own tech stack: this(!) became a bigger task than expected. I learned a lot about Supabase setup, and how to prepare and finally migrate an existing, working Lovable project to my own infrastructure.
- Domain setup, email system setup.
Learnings
TL;DR: if you’re willing to spend money, it’ll save you a lot of hassle. I could have bought a domain via Lovable and published instantly. I could also have skipped the learning curve of running my own infrastructure (Netlify/Supabase) and setting up email (transactional as well as marketing).
- jump-start with Lovable, iterate and adapt — but quickly learn about the limitations
- dealing with tool limitations:
- financial (pay-as-you-go or upgrade your contract)
- limited control, every change costs credits
— even simple text changes eat up credits and feels clumsy to spend credits on simple things
- prioritization: dealing with Lovable’s credit limits: with only 5 credits to spare a day, you have to choose wisely what and how you build
- connect and control via GitHub, re-connect, switch accounts, naming and renaming conventions
- dealing with Lovable workspaces: account vs. project ownership, multi-account usage, editing, roles & rights management
- moving a project to your own infrastructure: domain settings, Supabase setup and RLS, Netlify setup, email services setup (transactional + customer)
- full project migration to own infrastructure (Netlify, Supabase, domain management, e-mail services, tracking)
- product scaling & strategy & growth
- from 0 to 1
- marketing and traction, seeding, access limitations, growth initiatives
- user management and customer support
- user database management and user role management
- technical management (bugs, problems, infrastructure)