frAIgrant - Search for the most fitting fragrance for your taste | Product Hunt

Hello, Product Hunt community!👋

We are thrilled to launch our first product here📌 The idea for frAIgrant came to me when I was doing an internship and had no way to try new perfumes when I wanted to try something new.

So my co-founder Johannes and I decided to build a platform that lets you search for a new perfume if you already know in which direction it should go➡️ Especially in today's day and age you don't have a lot of opportunities to find a new perfume that exactly matches your taste. Although most people interested in buying a new perfume go into stores to try them out firsthand. Still, when you decide to purchase a niche perfume, that fewer people are wearing, there is no real solution to finding a new perfume, based on your taste❌

So here we are!😁 All you have got to do is search for your perfume of choice. Either a perfume that you maybe already have worn for a long time or a perfume that you have your eyes on👀 Our algorithm looks at the different ingredients and the importance of top, heart, and base notes so you get a recommendation that matches your taste✅ A percentage display gives you an indication of how much the recommendations are overlapping with your input fragrance.

We don't have "AI" in our name without a reason!👆 This is our first product, but definitely not the last one. We are planning on integrating survey data, reviews, feedback loops, and demographic data into our product so we can use AI to optimize recommendations as precisely as possible🎯

We are also working on developing a B2B product, that customers will be able to integrate into their shops to let their users search for a fitting fragrance, that matches their taste. This will be first rolled out for Shopify users soon🔜

Thank you very much for reading until here and I hope that our product can help you out🙏 I'll be here to answer any of your questions if you have any. We care a lot about our users, so if you have any feedback or comments about our platform, feel free to contact me or leave a comment📩 Thank you very much!

Cheers, David

Congrats on the launch. Some sensible suggestions came out - I'm still sad that there's nothing similar to Vert D'encens from TF and they discontinued it!

I'd love to see this extended to recommendations - people who liked X also liked Y, but using your similarity score to help improve that recommendations model. Using the similarity should allow for collecting niche fragrances into similar segments so it's not just popular scents that get promoted.

I like quite a range of scents, but wonder if I like certain "types" of fragrance, a group that you'd never associate because of how they smell so different, but might turn up over and again amongst certain user segments.