Comment by jabo | Hacker News Reader

Comment by jabo | Hacker News Reader

If I'm reading this correctly, it looks like search requests are now charged separate from the number of documents.

So if I have 5M records and 5M searches per month, in the old pricing scheme I would have paid $5K per month.

In the new pricing plan, I would pay $0.40 per 1K records and then $0.50 per 1K searches. So that's $2K for record storage and $2.5K for search requests for a total of $4.5K.

This blog post [1] claims they've slashed prices by 50%, which may be true at the unit level, but then it sounds like they've separated out charges for searches and records, so you still end up paying close to what you used to pay previously...

Am I misreading something?

[Disclaimer: I work on Typesense, an open source alternative to Algolia with a SaaS version, so I'm looking to understand this better myself.]

[1] https://www.algolia.com/about/news/algolia-introduces-new-de...

I think you are correct.

My biggest issues with Algolia were always:

- search requests should be separated from number of documents

-- think of a geoname service where there are 10 mio. documents vs. 500k search requests -- seems to be solved now

- it is crazy to require a new index for each sort direction

-- this is still the case

imho they should introduce cpu cycles + storage. until then self hosted Typesense, Meilisearch, Elasticsearch or hosted Typesense, Elasticsearch are still superior. I am leaving out Meilisearch here as their entry level is also nuts at 1.2k/month for hosted.

Hello, I'm the Meilisearch CEO. I think you're also correct, Jabo.

I just want to clarify. Meilisearch's pricing doesn't start at 1.2K/month, but at 0/month. We have a usage-based pricing that is basically 0.25/1000 documents and searches. And, funny thing, we are thinking about splitting the searches and documents, too, but we wanted to have more data to be sure to select the right unit price for each. :)

Sorry, this fine print so fine printed that I could not even see it.

You are correct. For < 1k documents it should be included on the hosted list.

Also, please really split documents and search request. In a real life scenario there will be 100k documents stored and only the top 1% will be searched for.

The one thing that always held us back from further investigating your product was searching in facets which was on the roadmap back then and imho crucial to any serious search ui. No idea if you implemented it now.

Reply on news.ycombinator.com