Hacker News Reader

Comment by usr1106 | Hacker News Reader

100+ employees for 1200 customers. So every customer needs to pay at least 0.1 developers on average to break even. Doesn't look sustainable. So we should expect to see more changes of terms over time. Like we have seen it from others with free offers, e.g. Qt in the past, Redis or gitlab somewhat more recently. Not saying that those have become bad, just no longer as free (either speech or beer) as they once started.

View on HN

Isn't Hashicorp all B2B? So like, one of those 1200 customers could be a single government customer paying crazy amounts? I don't think this is a good metric just like "lines of code" isn't a good metric.

> 100+ employees for 1200 customers

That's only one of their products - many of the others don't have an enterprise version at all ( e.g. Packer, Vagrant, Waypoint, Boundary), so that skews your math a lot. But not every product needs to be profitable in itself, there can be loss leaders, reputation builders, integrations. IMHO the main money making products are probably Vault and in some point in the future Boundary when it grows enough (the alternatives in the space cost an arm and a leg, and are terrible in comparison).

It's been a few years since I looked into the enterprise HashiCorp offerings, but I remember there was an X seat minimum (where X was 10 I think). Larger customers will obviously have many more than 10 seat license/support agreements. The pricing was definitely "enterprise".

>100+ employees for 1200 customers. So every customer needs to pay at least 0.1 developers on average to break even. Doesn't look sustainable.

wasn't Discord computer-stuff related team even smaller?

They are cheap too! Like crazy cheap.. I expect a price increase sooner or later..

I must admit I have never looked at their pricing. We are a small company of geeks, we don't buy support.

I'm not sure what OP is referring too, because their prices for the supported services previously started in the mid six figure range for them to not ghost you.

We had a tech + commercial meeting with them like two months ago, so they could provide us pricing for a managed Vault (Hashicorp Cloud Vault), and we're still waiting for their offer.

Meanwhile, we got tired and we've deployed our own OSS Vault ¯\_(ツ)_/¯

I've used and enjoyed Vault in the past, but these days just reach for AWS / GCP secret managers.

When would you use Vault over those offerings?

I wasn't involved in the decision and can't recall everything, but I remember a couple things:

- We have A LOT of environments, so GCP Secrets pricing climbed really quickly.

- Interest in dynamic secrets

Maybe OP was being ironic?

Irony is not helpful when talking to someone who does not know the facts.

The main thing you buy are enterprise level features, not support. E.g. for Terraform that'd be Sentinel(Policy as Code), collaboration, auditing, etc. related features. Vault'd be multi-cluster active-active HA, integrations with HSMs, etc.

Though i guess someone might want to buy support only, the value proposition is limited.

> Vault'd be multi-cluster active-active HA

is there such a thing? actually there is only the replication feature the last time I looked, which effectively can be built by using s3/gcloud storage and vault without too much trouble.

That seems... Not too bad?

You can add Elasticsearch to that list too.

Reply on news.ycombinator.com