Published July 13, 2026 · altengine team

What managed search really costs

When we launched, we said the usual search options are either a cluster you rent by the hour or a hosted service whose price climbs steeply past a few thousand documents — and promised a closer look. This is that post.

The prices below are public list prices as of July 2026, for the smallest realistic configuration of each service. Every one of these vendors has knobs — tiers, add-ons, regions, commitments — so treat these as the honest floor, not a quote. Links to each pricing page are at the end so you can check our math.

There are really only two ways search gets priced

Strip away the plan names and almost every managed search product bills one of two ways:

• You rent a cluster. A machine (or three) runs 24/7 and you pay for it by the hour whether or not anyone searches. This is Elastic Cloud, AWS OpenSearch, Typesense Cloud, and Meilisearch Cloud. The bill has a floor: the cost of keeping the lights on.
• You pay per record and per search. No cluster to think about, but you're metered on how many records you store and how many searches you run. This is Algolia — and it's the model closest to altengine.

altengine is in the second camp, with two deliberate differences: there's no monthly floor, and storage is billed by bytes, not by record count. That's the whole story of why the numbers below come out the way they do.

The floor: what you pay before serving a single query

For the cluster-based services, this is the smallest running configuration — one small node, single zone, no redundancy. It's what shows up on the invoice in a month where your app got almost no traffic.

*Algolia has a free tier (10,000 searches and 100,000 records/month); past that it's usage-based, no fixed floor.

The point isn't that a $22 node is expensive — it isn't. The point is that it's always on. A side project that gets ten searches a day still owes the full month, because you're renting hardware, not buying searches. With a usage-based model, a quiet month is a cheap month: an index you aren't querying costs only its storage, and under $3 it costs nothing at all.

Where per-record pricing bites

The usage-based services — Algolia and altengine — are the closest comparison. altengine charges $0.20 per 1,000 queries against Algolia's $0.50, so it already undercuts on search alone — but search rarely dominates a real bill. The bigger difference is storage.

Algolia meters your catalog by record count: $0.40 per 1,000 records beyond the first 100,000. altengine meters it by size: $1/GB-month for your documents plus $2/GB-month for their index, counting only the content you actually push. Those two models diverge fast as a catalog grows.

Take a growing app: 500,000 documents averaging ~2 KB each (about 1 GB of content), served with 50,000 searches in a month.

Storage is where it splits widest. altengine's 1 GB of documents and index runs $3; Algolia's 400,000 chargeable records (past its free 100,000) run $160. altengine is cheaper on the searches too — $10 against $20. Together that's an ~18× difference on the same workload, and it widens with every record you add, because record fees don't care how small your documents are — but altengine's storage bill does.

The honest caveat

Two things worth saying plainly, because a comparison that only flatters us isn't worth reading:

• A busy, steady workload can make a rented node cheap. If you're running constant heavy traffic against a modest index, a $22 Typesense node or a small OpenSearch instance can be very cost-effective — you're keeping a machine saturated, which is exactly what it's priced for. Usage pricing wins when traffic is spiky, seasonal, or small, and when you'd rather not babysit a cluster.
• These are floors, not full quotes. Production deployments add zones, replicas, bandwidth, and support tiers. We've compared the smallest honest configuration of each; your real numbers depend on your shape of load. The pricing pages are linked below — run your own.

How altengine prices search

One plan, pay-as-you-go, and the first $3 of usage every month is free for every organization:

• $0.20 per 1,000 queries
• $1.00 per GB-month for your documents, $2.00 per GB-month for their index — counting only the content you push
• $0.10 per 1,000 document deletes; indexing isn't charged per request, and deleting an index is free
• No always-on cluster, no monthly floor, no card required to start

An index you aren't querying costs only its storage. A month under $3 costs nothing. See the pricing page for the full rate card.

Start free in the console

Sources

List prices as of July 2026: Algolia, Elastic Cloud, AWS OpenSearch, Typesense Cloud, Meilisearch Cloud. altengine is an independent project and is not affiliated with any of these vendors.