Featureflip

Overview

Featureflip is feature-flag infrastructure for changing application behavior without deploying new code, with rollout controls, experiments and code cleanup in one toolkit. It suits developers and small teams who work through SDKs, APIs and development tools; the main trade-off is a free tier capped at one project and no visual editor.

Its most compelling detail is the path from a dead flag to a reviewable pull request that removes its code path, making cleanup part of development workflow rather than a separate chore. Teams that need broader project capacity or longer audit history will need a paid plan.

Key features

Percentage rollouts, targeting rules, kill switches, A/B tests and isolated environments cover gradual releases, audience selection and rapid rollback. That mix is useful when experimentation and operational controls belong in the same system. Server-side testing and mobile SDKs widen the range of applications it can serve, but teams that rely on a visual editor should look elsewhere.

When a flag is dead, a GitHub Action can open a reviewable pull request to remove its code path across thirteen languages; after merge, the flag is archived. This is a practical safeguard against stale flag logic, particularly for teams that regularly ship experiments.

Thirteen SDKs and OpenFeature providers, a REST Management API and an Evaluation API offer several integration paths. An MCP server supports agents including Claude Code, Cursor, Copilot and Cline; Terraform and GitHub Actions are also supported. Server SDKs evaluate from in-memory configuration and retain their last configuration if Featureflip is unreachable, helping avoid a live service dependency for evaluation.

Featureflip says its service and PostgreSQL database run on Hetzner Cloud in the EU. Evaluation context is not stored: server SDKs do not transmit it, while client SDKs send it for evaluation and it is then discarded. Two-factor authentication is available on every plan; the security page also describes Google sign-in, hashed passwords and temporary lockouts after repeated failed sign-ins.

Pricing

The Free plan costs nothing, billed Free forever. It includes one project, ten flags per project, two environments per project, up to three team members, seven-day audit logs, one webhook endpoint and community support. It is a useful starting point for a small project, but the project, flag and environment caps can constrain growing teams.

Team costs $49 a month, or $39 /mo billed annually. It removes the flag cap and allows three environments per project, three projects, up to five team members, 30-day audit logs and three webhook endpoints, with email support. This is the clearest step up for a small team that has outgrown the free project limit, though the five-seat cap may matter as it expands.

Pro costs $99 a month, or $79 /mo billed annually. It provides unlimited flags, five environments per project, ten projects, up to ten team members, flag owners, 30-day audit logs and three webhook endpoints, plus email support. It fits teams managing more projects or needing named flag ownership; it does not extend the audit window beyond Team.

Business costs $249 a month, or $199 /mo billed annually. Its limits rise to 25 projects and 25 team members, with unlimited flags, environments and webhook endpoints, 90-day audit logs, OIDC SSO and priority support. It is aimed at larger teams needing broader capacity and longer audit history.

Enterprise has custom pricing negotiated directly, with custom scale and requirements, custom data retention, a contracted SLA, on-premise deployment and SAML/SCIM. There is no free trial. The Solo entry also shows free, billed Free forever, with one developer, one project, ten flags per project, two environments and seven-day audit logs; for teams, the Free plan has the same project, flag and environment caps but allows up to three members and includes one webhook endpoint.

Platforms

Featureflip supports Android, iOS, macOS, web, API and self-hosted deployments. The SDK and API options make it relevant to both client and server applications, while local server-side evaluation offers a fallback when the service cannot be reached.

Who it's for

Small teams, indie developers, startups and consultants are the stated audience. The Free plan can suit an individual or compact team validating a first workflow; Team becomes more practical once multiple projects or more than three collaborators are needed. Business and Enterprise are better aligned with larger project portfolios or specific identity, retention, deployment and SLA requirements.

Pros and cons

• Pros: Dead-flag cleanup can remove code through a reviewable pull request across thirteen languages, reducing reliance on manual cleanup.
• Pros: Server SDKs evaluate locally and keep the last configuration if the service is unreachable.
• Pros: Multiple SDKs, OpenFeature providers, APIs, Terraform and agent integrations give development teams several ways to operate flags.
• Cons: Free is limited to one project and ten flags, so it is a constrained long-term option for teams with several applications.
• Cons: There is no visual editor, which makes Featureflip a poor fit for teams that want to build experiments through a visual interface.
• Cons: The paid plans impose team and project caps until Business, and audit history remains at 30 days on Team and Pro.

Alternatives

For a broader comparison of experimentation products, see A/B Testing Tools; for flag-focused options, browse Feature Flag Software.

Experimently is worth considering if self-hosting on your own AWS account and an all-features-included free preview are priorities. Apex may suit teams looking for a free-forever starter plan with a free trial and a stated allowance of three live experiments and 10,000 tested visits.

Confidence by Spotify is a possible alternative for teams who value a free plan with unlimited seats and feature flags, 75,000 monthly credits, up to five surfaces and EU or US hosting. DevCycle is another freemium option, with its free plan expressed in client MAUs and monthly cloud configuration requests.

Firebase Cloud Messaging is a no-cost alternative in this category. GrowthBook may appeal to teams that want an open-source, self-hosted plan with unlimited flags, experiments and traffic for one project.

LaunchDarkly Experimentation offers a free-forever Developer plan with one project, three environments and A/B tests and experimentation. Mida may fit teams seeking a free, no-time-limit sandbox with stated monthly tested-user and daily tracked-event allowances.

Verdict

Featureflip is a strong fit for development-led small teams that want feature flags, experiments and a practical code-removal workflow, with local server evaluation as a useful resilience measure. Choose it if those tools and its free single-project plan match your needs; look elsewhere if you need a visual editor, or choose a paid tier only if its project, seat and audit limits suit your team.