Hub for self-hosted experimentation and feature flag platforms on GrowthBook, Unleash or Flagsmith, and how industry, release controls and data location change the setup.
For SaaS engineering and product teams: gradual rollouts, beta programs and A/B tests on self-hosted GrowthBook, Unleash or Flagsmith, which limits of the free editions bite first, SDK coverage and the cost per seat once Enterprise features are needed.
For banks and fintechs: treating flag changes as production changes, approvals and audit logs, keeping customer attributes used for targeting inside the bank's infrastructure, DORA change management and which editions provide change requests.
For online stores testing pricing displays, checkout steps and recommendations: server-side versus client-side experiments, avoiding flicker, revenue metrics from the order database or warehouse, sample sizes and how GrowthBook's warehouse-native analysis fits.
For iOS and Android teams: remote configuration and kill switches without app store releases, SDK caching and offline defaults, edge or proxy layers for latency, and rolling out to percentages of users on self-hosted Unleash, Flagsmith or GrowthBook.
Running A/B test analysis on your own warehouse with GrowthBook: connecting BigQuery, Snowflake, ClickHouse or Postgres, defining metrics in SQL, what leaves the warehouse (aggregates only) and the data engineering work needed first.
For platform teams decoupling deploys from releases: percentage rollouts, canary cohorts, operational kill switches, flag lifecycle and cleanup, and wiring flags into CI/CD and observability so a bad release is switched off in seconds.
For public companies whose flags change financially relevant systems: when a flag toggle counts as a production change, approval and segregation of duties, audit trails auditors test, and the edition features that provide change requests and logs.
For SaaS companies under SOC 2: how feature flag changes map to change management criteria, approvals, role-based access and audit logs on self-hosted platforms, and the evidence an auditor will sample.
For EU organizations that cannot send user attributes used for targeting to US SaaS: self-hosting flag evaluation in the EU, local evaluation in SDKs, what telemetry still leaves, and the transfer documentation to keep.
For defense, critical infrastructure and on-premises products: running Unleash or Flagsmith with no outbound internet, offline images and updates, local SDK evaluation, and shipping flag configurations between environments by export and import.
A requirements checklist for self-hosted feature flags and A/B testing: functional must-haves, hosting and data residency, identity and access, privacy and retention, integrations, operations and support, so a team can compare self-hosted options before choosing a platform.
Questions to put in an RFP or vendor brief for self-hosted feature flags and A/B testing: platform and edition fit, licence obligations, migration of existing data, hosting model, security evidence, support terms after go-live and how the fixed scope is defined.
Cost structure for self-hosted feature flags and A/B testing: subscription pricing published by SaaS vendors against self-hosted infrastructure, implementation, support and internal staff time, which drivers move each line, and when bringing in an implementation partner costs less than building alone.
Tell us what you're working on, whether it's a deployment, an audit, a security test or a cyber range. You'll speak with an engineer who can help you scope it.