Last year I started creating rapid prototypes using tools like Vercel V0, Bolt and Cursor, helping me to get ideas across and get quick customer feedback. I’ve since heard good things about Lovable, so let’s give this AI tool a try by creating a simple theatre ticketing app. On the Lovable homepage I enter the prompt that I generated and refined through Claude since I want to provide Lovable with the right level of detail to start creating my theatre ticketing app.
Similar to Vercel V0, Lovable offers a split screen, with an explanation of its design and build approach whilst the code is being generated.
The first version of the app is generated quickly and it looks ok. It’s easy to see that data is missing in the back-end, with only one stock image of theatre displaying on the “Shows” page.
The “Venues”, “Special Offers” and “Gift Cards” pages all return a 404 error. Similarly, whilst the event and seating selection process looks as you’d expect from your average theatre ticketing app, I’m not sure where the data is coming from 🙂
Lovable has a native integration with Supabase, an open-source backend solution, which can be used for things like user authentication, seating layout information, payment processings and ticket purchase management. I start with creating simple database tables in Supabase, which I can then populate with CSV data.
Once I’ve connected my Supabase account to my app in Lovable, I cab create user authentication and login functionality.
When the code throws up an error, the AI assistant explains in the chat what I need to do to resolve the error:
I follow the suggested steps and the login and authentication functionality is generated, using Lovable’s integration with user management platform Clerk.
It would be good to have more guidance on the “knowledge” section in Supabase to make sure that the right context and instructions are applied to any edits that I make. I apply a similar approach to how I’d create a personal assistant — providing, role, context, criteria, instructions and examples — but I’m not sure how this context will apply to any app changes that I make.
Through my prompts I can make changes quickly in Lovable, particularly on the front-end. For example, adding genre filters is pretty straightforward.
I then publish my app with the intention of putting in more work to make the different capabilities of my theatre ticketing app fully usable.
Main learning point: Lovable clearly accelerates prototyping but doesn’t eliminate the need for human expertise. Similar to other AI prototyping tools, Lovable isn’t a magic bullet that will eliminate all human involvement in product development. While Lovable rapidly transformed my concept into a functional app skeleton, I can see the need for more hands-on human supervision when I start adding more complexity or making more changes across the app. This doesn’t take away from the power of Lovable to really quickly experiment and validate different product ideas before making significant resource investments.
