What auth actually costs at 10,000 users

Auth pricing is hard to compare because no two vendors bill for the same thing. So we took one specific scenario and priced it across seven platforms, with every number taken from the vendor’s own public pricing page and checked on the day this was published.

The scenario. A consumer-facing product with 10,000 monthly active users. Passkeys for sign-in. Your own domain on the login page. No vendor branding. Three enterprise customers who want SSO into their own identity provider. That is an ordinary B2C product with a few business customers, and we picked it because it is common.

Amounts are in each vendor’s own currency. We didn’t convert them, because an exchange rate would add a number nobody published. Sources: Stytch, ZITADEL, Clerk, Hanko, WorkOS, Auth0. All checked 11 September 2026.

We are not the cheapest

Stytch is free in this scenario and ZITADEL comes in a fraction under us. We are third. If price is all you care about, the table tells you where to go. Better you hear that from us than find out afterwards.

What we claim is narrower. Nothing is gated behind a tier. There are no per-connection fees, so your third enterprise customer costs the same as your first: nothing. We don’t charge to remove our branding, because we don’t put it on your login page in the first place. And the number on the pricing page is the number on the invoice.

Where the price differences come from

Every platform in that table can do everything the scenario asks for. The spread from $0 to $2,100 comes from packaging: which capabilities each vendor includes in the product, and which it bills for on top.

Enterprise SSO is the clearest example. It is a standard protocol, implemented once, and it behaves the same for your fiftieth customer as for your first. Charged per connection, it becomes a monthly cost that grows every time you win a B2B deal. At $125 each, twenty enterprise customers is $2,500 a month for code that has not changed since it was written.

Auth0 goes one step further. On its consumer plans, enterprise connections are listed as unavailable at any price, so our scenario can’t be bought on those tiers at all. Three SSO connections means moving to the B2B product, and that move is what produces the $2,100.

Two billing metrics worth understanding before you sign

Clerk bills in monthly retained users. An MRU is a user who returns 24 hours or more after signing up, so someone who signs up and never comes back is not counted. That is a reasonable thing to measure, and arguably fairer than raw MAU for a product where many signups never return. But no other vendor uses it, so it is hard to forecast from your existing analytics, and hard to compare the resulting bill with anyone else’s quote without modelling it first.

ZITADEL bills in daily active users summed over the month. If 1,000 distinct people use your product every day, that is roughly 30,000 DAU-sum, not 1,000 MAU. The metric favours products with occasional users over ones people use every day. That is a legitimate choice, but it makes the comparison above approximate, which is why the table says so. Overage rates beyond the included allowance are not published, so to find out what happens when you exceed it, you would have to ask them.

Neither metric is a trick. But with either one, the quote in front of you can’t be compared directly with the one next to it, and you have to do the arithmetic at the moment you are choosing a vendor.

What happens to the bill as you grow

This month’s invoice is rarely the problem. The problem is what the invoice does when you succeed. Per-connection pricing means every enterprise deal carries a recurring line item you did not quote for. Proprietary metrics mean you cannot model next year. Feature gating means the capability you need at the moment you need it sits one tier up, and you find that out under deadline with a customer waiting.

One fairness note, since it gets repeated carelessly: when Auth0 repriced in 2023, existing customers were grandfathered on their plans unless they changed subscription. Nobody’s bill jumped overnight. Our point is about where that pricing structure leads as you grow.

Two honest comparisons

Hanko is German-owned, in Kiel, hosted in AWS Frankfurt, and lands within a euro or two of us in this scenario. If you want a European, passkey-first provider, they are a real alternative. We package things differently: their SAML is $49 per connection where ours is included, and we could not find SCIM. Which of those matters depends on what you are building.

Stytch is free here, and if that settles it for you, take it. Two things are worth knowing first. Its own documentation states that it operates exclusively out of US servers. Some buyers won’t care. If your customers ask where identity data lives, it is the answer they will care about most.

The other is that Stytch is no longer an independent identity company. Twilio closed its acquisition in November 2025, so you would be adopting one product line inside a communications business. The software is good. The question is whose roadmap your login sits on, what happens to it the next time that portfolio is reviewed, and who picks up the phone when something specific to identity goes wrong. Nobody wants to migrate their authentication twice.

Our own numbers are €0 up to 1,000 monthly active users, €99 to 10,000, €199 to 20,000, and a conversation above that. Every product feature is in every one of those, including the free one. The same scenario, priced the same way, is on the pricing page, and we re-check the competitor figures when we touch it.