GrowthBook vs Statsig

GrowthBook

Visual editor and URL redirects, server-side experiments through feature-flag rules or inline SDK code, custom assignment analysis, holdouts and namespaces for mutual exclusion, 24-plus SDKs spanning server, client, mobile and edge, and a documented edge app on Cloudflare and other providers that evaluates flags before HTML ships. We found no public information on rule-based personalisation that can itself be tested against a control, nor a searchable library of past experiment results, which is what holds this back.

Statsig

The capability list names server-side delivery through 30+ open-source SDKs, feature flags tied directly to product data, layers and holdouts, switchback and non-inferiority tests, a no-code editor and marketing experiments — broad by any standard. But what the captures show is a feature grid: we found no public information on edge or CDN delivery, on personalisation that can itself be tested against a control, or on a documented experiment lifecycle, so I hold it in the middle of the range.

GrowthBook

Both engines are on the record with their assumptions — Bayesian with a documented default improper prior and an optional proper Normal prior, frequentist as two-sample t-tests on relative change — alongside sample ratio mismatch detection, CUPED, guardrail metrics, minimum data rules and a power calculator, with an open-source implementation a result can be checked against. My reservations: the multiple-comparison corrections exist only for the frequentist engine while Bayesian is the default, sequential testing appears as a listed test type without the peeking mathematics documented, and adjusted p-values can be exported only as CSV, not through the API.

Statsig

The pricing page names the techniques I read for first: CUPED for variance reduction, Bonferroni and Benjamini-Hochberg for multiple comparisons, winsorization, sequential testing against peeking, plus a power analysis tool for sizing. That is more than one method by name — peeking protection and multiple-comparison correction are both called out. But we found no public information on sample ratio mismatch detection or guardrail metrics that stop a harmful test, and the assumptions behind the named procedures are not documented, so it stays short of the top of the range.

GrowthBook

The consent mechanics are documented in implementable detail: a first-party identifier cookie the edge server does not write by default, a documented setting plus a JavaScript event that delay cookie persistence until consent, and hash-based assignment that works with no cookie at all. Less complete: we found no published cookie lifetimes, nothing on IP anonymisation or consent management platform integration, and the privacy notice states DNT signals are not honored.

Statsig

The only captured statement on device storage is Amplitude's site-wide privacy notice, which lists essential, performance, functionality and advertising cookies and web beacons, names consent among the legal bases, and honours Global Privacy Control where required by law. We found no public information on what the Statsig script stores or does before consent, on consent management platform integrations, on a cookieless mode, or on how testing maps to §25 TDDDG.

GrowthBook

To its credit the edge documentation states plainly that the default mode makes a blocking API call on every user request, then documents two ways to remove it — a push model into an edge key-value store or a just-in-time payload cache — plus a self-hosted proxy with your own CDN and payload endpoints scoped to a single environment or project; edge and hybrid delivery is documented as eliminating redraw flicker for visual and redirect tests. What is missing for a higher mark: no published script sizes or Core Web Vitals figures, only a relative claim of fifty percent smaller SDKs.

Statsig

The one published figure is post-init evaluation latency under 1ms, which measures SDK evaluation speed, not what a snippet costs the page it tests. We found no public information on script size, anti-flicker handling and its trade-off, impact on Core Web Vitals, or a self-hosting option for the script.

GrowthBook

Warehouse-native is the design, not a bolt-on: metrics are defined in SQL and computed in the customer's own Snowflake, BigQuery, Databricks or Redshift, every query is visible, results are described as reproducible, and a full REST API with versioned endpoints covers experiments, results, fact metrics, segments and feature toggles. The nits: adjusted p-values are exportable only as CSV rather than through the API, and we found no stated API rate limits or a CDP audience integration.

Statsig

Warehouse native deployment — running the platform in the customer's own warehouse — is described on the pricing page alongside incoming data integrations, warehouse ingestion and a data export API, and that is the claim that makes an experiment result checkable in the customer's own data. But outgoing integrations and warehouse imports sit in the enterprise tier, and we found no public information on visitor-level export detail or documented API limits.

GrowthBook

A California contracting entity under California law, processing described as held in the United States with visitors consenting to that transfer, US AI providers named as subprocessors for optional features, no published subprocessor list and no stated transfer safeguard. The sole mitigation on the vendor's pages is the self-hosted, air-gapped deployment where no data leaves your own system; everything captured about the hosted offering points away from European sovereignty.

Statsig

The contracting and processing chain is American: the terms are governed by Washington law, and the privacy notice names Amplitude, Inc. of San Francisco as controller of visitor personal data, with transfers to the United States covered by standard contractual clauses and the Data Privacy Framework. 'EU hosting' appears on the pricing page only inside the enterprise warehouse-native deployment, and the captured pages name just Google and Stripe as subprocessors in passing — we found no published subprocessor list or a Statsig-specific residency statement, so this sits at the level of an EU option on a non-EU contract.

GrowthBook

Starter is free without a credit card, Pro is published at $40 per seat per month for up to 30 users, and — unusually for this category — traffic is published as unlimited, so the invoice hinges on seats and the metered extras. Those extras carry published limits and overage figures for managed warehouse events, CDN bandwidth and CDN requests, but Enterprise is a sales conversation and we found no public information on minimum term or VAT treatment.

Statsig

For the self-serve tiers a buyer can compute: Pro is $150 /mo with 5M events included and then $0.05 per 1K events, the free tier's 2M events per month and its downgrade behaviour are stated, exposure deduplication windows and the no-charge rule for 0% and 100% rollouts are documented, and the terms state taxes are the customer's with one-month auto-renewal and 30 days' notice on renewal price increases. The enterprise tier is Custom, with event- or experiment-based contracts, and warehouse native plus standalone analytics sit there — so a large deployment's annual invoice is a sales conversation, one step below full computability.