Hi Product OG's,

Super stoked to share what I've been working on.

Searching for software on Google has become a little hit or miss these days. So I rely on ProductHunt Search but every time I use it, I wished it had Google-like natural language support for finding products and that's the reason I built ProductSearch.

Product search is a natural language search engine that helps you find the right software you need in seconds and it's powered by ProductHunt's products data.

Here's how it works There are a couple of things going on inside that search engine, but I'll try to boil it down to simple points.

ℹ️ We collect product info from ProductHunt using their API 🦀 We crawl the internet and look for articles that mention the products from ProductHunt 👩‍💻 We then extract appropriate keywords that are relevant to the product 📜 We build the index out of keywords that are collected

Note: Our Index size is small, so you might get some irrelevant results. But I've been working hard to scale the index size!

Ranking I've been experimenting with a couple of ranking ideas but one that I've found to be particularly effective is this

🗂️ We only consider links that are organic, meaning they aren't paid for or incentivized ⬇️ The engine sometimes penalizes the products for some keywords, if it has too many sponsored backlinks

You'll see on the top right of the site that it says "Built by Grep" Yes, I've been experimenting with a new kind of search engine, but this seemed like a fun problem to solve first.

I've been using this site for the past couple of years. I learned a lot from other fellow makers, so consider this my small contribution to ProductHunt and I hope it helps users find products faster.

Please feel free to leave any type of feedback. I am happy to answer any questions you may have.

Vignesh