A to Z today

Personal side project, built and designed solo.

I'm obsessed with daily word games: 400+ Parle matches and counting. I wanted to experiment with new tools and build something from scratch, end-to-end. The result is A to Z Today: a daily word game where you guess the word by narrowing down the alphabetical range.

No brief, no PM, no handoff. Just a product idea and the question of how far I could take it alone.

I used Lovable for vibe-coding the experience. Supabase handled auth, database, and storage. It was a deliberate exercise in using today's tools to compress the gap between idea and working product.

The first version shipped fast and rough. The moment I shared it, feedback came in immediately. This became the real design brief.

Onboarding was the first thing that needed fixing. People were confused before they even started playing. I added clearer instructions, a GIF showing the core interaction, and a persistent ‘how to play’ link. You should be puzzled by the word, not by the game.

The leaderboard started hidden: my instinct was to use it to gate signups. I quickly reconsidered. A visible leaderboard motivates people to earn their place on it. One user hit a 124-day streak. That's the kind of engagement a hidden leaderboard would have killed.

The dictionary broke early when the free API I relied on got blocked. I rebuilt it with my own dictionary: better performance and no external dependency. Sharing also grew from a copy-to-clipboard shortcut into a proper social feature.

Numbers were modest but meaningful: 1,000+ matches played and 50+ registered users. The quality of engagement was the real signal: people came back daily and organically recruited others. LinkedIn drove the most traction; Twitter did nothing. Reddit was okay-ish but expensive.

The biggest takeaway wasn't about the game. It was about end-to-end ownership: making product decisions, breaking things, fixing them, and shipping without a team to catch your mistakes.