Why I Built It
WheelOfNames is great and the most popular randomizer on the planet* (citation needed). They also place ads and share user data with a couple hundred technology partners. I felt I could do better. Maybe not with their User Experience, but I could make a privacy-first ads-free version that I would rather use.
Any One Will Do
Any One Will Do is a privacy-first random selection web application. All the data that is needed for operating the application lives in the user’s browser and never hits a remote server. Even saving sessions is all done in the local browser. Full disclosure, We do use Plausible, an EU friendly privacy-first analytics company that aggregates data so we have some usage stats to evaluate existing and future features.
Users can select from multiple animations as well as create groups. The victory sound is even an optional setting.
What I Learned
Building towards a strong vision is hard work, even in the age of AI. A “simple” application like this took me at least 100 hours to get it to the current state. The wheel animation had to be just right. The “click” had to line up with imaginary pegs. What happens when someone puts 10,000 names in (spoiler: it works great with a little extra UI work). Even the decision to show demo data so new users can try it right away at the risk of making it a little harder to load your own list was a decision I went back and forth with.
Working towards 0 server requirements, other than the initial application load, was a new experience for me. Spending time understanding local storage and not rely on a database was a bonus learning opportunity.
This is my first project iterating with AI and many of the techniques I use today were born from this experience.
Maybe the hardest part was just getting eyeballs on the project. I don’t have a large following. My partner shared some links on Facebook and I posted a few comments on Reddit, and that was about it. Since we don’t monetize the site, we don’t have any paid promotions to drive users. Even so, we are pleased to have the traffic we get, about 500 uniques a month, with about 1/2 repeat users.
Features include auto-saved sessions, import/export, weighted entries, shared results, grouping, and support for adding metadata (like links) to entries.
See the 30-day launch check-in for how it’s done since launch.