23 comments

[ 3.8 ms ] story [ 61.1 ms ] thread
[flagged]
I don't follow, was $9/user/mo (presumably what the Team plan was) low enough for it to be "effectively free" and invite problem customers?

I would have thought those sorts of people just go for the $0/mo option here.

Presumably.

It's also possible that some organizations incur more than $9 per month in admnistrative or other costs to bill a single customer.

Do they use another $9/per month/per customer SAAS subscription to achieve that?
You'd be surprised how many firms have subscriptions to billing insights, know-your-customer, fraud detection and other similar services. Many of the "AI" companies that are emerging do such things.
That wouldn't surprise me. What would suprise me, though, if any of this (individually or in aggregate) is such a huge expense on a per-customer basis
Maybe if you're billing with a paper and pencil, mailing out invoices, and you have to drive a bag of checks over to the bank.

BuildKite is an orchestration SaaS, they don't even own or run the hardware the agents are on. If it costs them $9/mo/user to manage that most of the C-suite should be fired.

I see the edit but I think you might not be too far off. I get the impression that businesses whose customers are engineers can be some of the most difficult to satisfy. Perhaps its easier to focus on the larger customers going forward.
Engineers as customers are either great ("hey, found a bug, <detailed report with suggested fix>") or terrible ("hey, found a bug, <long rant about how they could have written this themselves and how terrible your developers are and how they are going to leave for competitor X and ...>").
So? Businesses change their pricing all the time. This isnt some sort of conspiracy
This is a strange reaction. Nowhere in the title does OP imply that it's "conspiracy theory", they're just sharing the change.
compare "Buildkite has quietly removed their $9 "Team" plan" vs "Buildkite removed their $9 "Team" plan"
so could you point to a recent blog posts of theirs anouncing the pricing change that would make the "quietly" part false?
Headlines with "quietly" are one of my pet peeves. It's a general purpose go-to to make anything seem insidious.
It's customary to announce such changes ahead of time. BuildKite made an announcement the last time they made a major change to their plans, per another commenter.

I haven't found any announcement for this change, hence I think "quietly" isn't an unfair description.

One thing to note with Buildkite is that they are a "bring your own infrastructure" company. All of your builds and CI tasks happen on your own machines, with their SaaS as an orchestrater. This means that their infrastructure costs shouldn't be as high as other CI/CD platforms.

That said, they are absolutely the best CI platform if you're doing anything that requires special hardware. They're absolutely fantastic for GPU testing. They even provide a cloudformation stack with autoscaling, so if you're using AWS and aren't running builds the infra costs drop down to almost nothing. For the companies I've worked at that have used Buildkite it's definitely been worth the money.

"Managed 21st century jenkins" is how I'd describe it. We're (very) happy buildkite customers at our work.
I'm on a $15 per month legacy plan, this means I have unlimited build minutes (I think?), but for what it is worth, I can still switch to the $9 "Team" plan in their interface.

The interface tells me that a Team plan has:

> 20,000 Job minutes included per month

> + 5,000 Job minutes per user

> 250,000 Test executions included per month

> + 50,000 Test executions per user

Compared to a "Business plan" which has:

> 40,000 Job minutes included per month

> + 12,500 Job minutes per user

> 500,000 Test executions included per month

> + 50,000 Test executions per user

I love Buildkite! :)

Hi I’m Keith, Buildkite CEO.

You caught us! Well spotted.

We had such little success with the $9 plan. When people upgraded, they generally chose the more expensive option. Knowing this, we removed it to see what happens to upgrade rates. We might even increase the free offering to balance it out, not sure yet.

Pricing is dark art of science, economics, research and “making shit up”. I dunno if this iteration is any better, but we’ll see!