The two meters that matter
Flagsmith’s Start-Up plan is US$45/month and it’s decent value at small scale. But it carries two meters that both point at growth. Requests: 1 million per month included, then US$50 per additional million. Seats: three members included, and the next tier charges US$50 per member per month. Hire two engineers and cross the request cap in the same quarter, and a US$45 plan quietly becomes several hundred dollars.
RocketFlag charges A$79/month with neither meter. Unlimited seats, unlimited evaluations, unlimited flags. One customer serves millions of evaluations a month and pays what a team of three pays.
Where Flagsmith genuinely wins
Flagsmith is open source and self-hostable, and that’s not a small thing. If your compliance posture, data residency or plain preference says “run it yourself”, Flagsmith is the better choice and no flat price changes that. They also do multivariate flags and remote config, which RocketFlag deliberately doesn’t.
We’re the opposite trade: managed only, flags only, one price. Evaluations served in milliseconds from regions close to your users, an API that keeps serving flags even if your billing lapses, and Caretaker keeping your flag list honest with AI-agent removal prompts.
Run the numbers at your actual scale
Price both at the team you’ll be in a year, not the team you are today. Count the seats, estimate the requests, then look at the two invoices. Ours is on one page with no calculator. If the comparison still favours Flagsmith, take it; they’re good software. If what tipped it is a meter, that’s exactly the thing we removed.
