I’m building a small experiment called Rankiwiki, a site where every input becomes a ranking item.
The core rule is intentionally simple:
Anyone can type any word
That word instantly joins the ranking
No predefined options, no forms, no signup
This has been interesting, but it created a UX issue I didn’t fully anticipate.
As more people participate, long-tail entries (one-off words, spelling variants, synonyms) slowly dilute the percentages.Nothing is wrong with the data, but the top ranks start to feel weaker than they actually are.
I don’t want to delete or ignore minority inputs, openness is kind of the point,but I also want the rankings to feel meaningful at a glance.
Right now I’m considering a few approaches:
Before I implement anything, I’d really appreciate outside perspective:
If you were a user, what would make an open, free-text ranking feel clear and trustworthy, without killing the “anyone can say anything” spirit?
Happy to share more details if helpful. Thanks!
How I Finally Made Reddit Work for My SaaS (After 2 Months of Failure)
First you have to do your best to correct at the time of entry. So if for Best Anime I enter "Frieren: Beyond Journey's End" and someone has already entered "Frieren" then you have to ask me if I meant "Frieren" and if so it's just like I clicked "Frieren" to begin with.
So that way the duplicate simply never gets entered - which is your best outcome.