Statsig Pricing, Decoded - Why Flags Are Free and the Bill Lives in Events (2026)

Most pricing pages for Statsig miss the one thing that makes it different, so let me lead with it. Statsig does not charge for feature flags. At all. On any tier. Flag and config checks are unlimited and free, forever. What you pay for is analytics events. That single fact flips the usual feature-flag math on its head, and it’s the reason this decode is short and clear where a LaunchDarkly one is long and painful.

This is a neutral walkthrough of the model: what Statsig bills on, why flags are free, how to estimate your own number, and where the cheaper picks are if events - not flags - are your cost center. Every figure here is from our Statsig review.

What Statsig bills on

Start with what it does not bill on: flag checks. Statsig lets you evaluate flags and configs an unlimited number of times, on every tier, at zero cost. Compare that to LaunchDarkly, which meters client-side MAU and climbs with your traffic. With Statsig, traffic against flags is free.

The meter moves on analytics and experiment events instead - the tracked events and experiment exposures you send. That’s the whole model in one line: flags are free, events are the bill.

Statsig was built by ex-Facebook engineers to mirror Facebook’s internal experimentation stack, so the analytics and experiment engine is the center of gravity, and that’s exactly what it charges for.

The pricing ladder

Three tiers, and the free one is genuinely generous with no credit card wall.

Tier	Price	What you get
Developer (Free)	$0	2M events/mo, unlimited flag checks, 50k session replays, no card
Pro	$150/mo flat	5M events included, then $0.05/1k events, unlimited flags, 100k replays
Enterprise	Contact sales	Custom event or experiment contracts, warehouse-native, SSO, RBAC, HIPAA

Two things stand out. First, Pro is a flat $150/mo, not usage-based on flags - a predictable number you can drop in a budget, which is rare here. Second, the free tier’s 2M events and 50k replays are enough to run a small product in production for nothing.

The one limit to note: Pro is one project per subscription. Run multiple products and that adds up, one subscription at a time.

Why the flag-free model matters

Think about what “flags are free” actually buys you. A high-traffic app might check flags millions of times a minute. On an MAU-billed tool, that traffic is your bill. On Statsig, it’s zero. If your use case is mostly feature flags with some experimentation on top, Statsig is one of the cheapest serious platforms you can pick.

The trade is that the meter lives on events. If you send a lot of analytics - heavy event tracking, lots of experiment exposures - that’s where cost accrues. Flags free, analytics paid. Know which side of that line your usage sits on.

How to estimate your own bill

Here’s the sequence to get a real number:

1. Ignore your flag volume entirely. It’s free. However many times you check flags, that line is $0.
2. Count your monthly analytics events. Every tracked event plus every experiment exposure. This is the number that matters.
3. Compare against the tier includes. Under 2M events a month with light needs? The free Developer tier may cover you. Need more, or want the flat predictability? Pro includes 5M.
4. Model the overage. Past Pro’s 5M, it’s $0.05 per 1,000 events. So 10M events a month is 5M free-in-tier plus 5M overage: 5,000 x $0.05 = about $250 on top of the $150 base, roughly $400/mo.
5. Add a subscription per project. Pro is one project each, so a two-product team is two Pro subscriptions.

The output is your real number - and because flags don’t factor in, it’s far easier to forecast than an MAU-based tool.

The one asterisk: who owns it

Before you commit for the long haul, price this in. OpenAI acquired Statsig on 2 September 2025 for a reported ~$1.1 billion in an all-stock deal, and founder Vijaye Raji became OpenAI’s CTO of Applications. The product still ships independently from its Bellevue office and serves existing customers.

The risk isn’t the price today - it’s roadmap uncertainty. Does the standalone product stay a priority inside a company as large as OpenAI? That’s unanswerable from the outside, and adopting Statsig now is a multi-year bet on it. Not a dealbreaker, but a factor to weigh alongside the friendly pricing.

Cheaper alternatives, and when they win

Statsig is already cheap for flags. But if analytics events are your real cost, or you want a different shape of bill, two tools attack it differently:

• PostHog - if you want analytics and flags in one meter. Flags and experiments share a free tier of 1M requests a month, then fractions of a cent per request. Its experiments read straight from your product analytics, the same as Statsig’s pitch. The watch-out is that each product (analytics, replay, surveys) bills separately, so the total climbs across product lines even while flags stay cheap. Good if analytics is your center of gravity and you want it bundled.
• GrowthBook - if you want to stop paying for events entirely. The self-hosted MIT edition runs unlimited users for $0, and being warehouse-native means it queries your own data warehouse instead of ingesting events you’d otherwise pay to send. Cloud Pro is $40 per seat per month. The cost isn’t dollars, it’s a properly instrumented warehouse and someone who understands the statistics.

What none of these fully match is Statsig’s experimentation engine, which is the strongest of the group. That’s the actual trade: engine depth versus how the bill is shaped.

The bottom line on Statsig pricing: it’s the friendliest flag billing in the category because flags are free, and Pro’s flat $150 is a rare predictable number. Count your events, not your flags, model the overage honestly, and price in the OpenAI-ownership question for a multi-year commitment. Do that and Statsig is excellent value.

---

Pricing verified against statsig.com on 26 July 2026. Feature-flag pricing changes often - we re-verify regularly. The OpenAI acquisition figure (~$1.1 billion, 2 September 2025) is from third-party reporting and vendor confirmation, attributed as such.