Feature Flagging and Experimentation Release Notes
Release notes for feature flagging, A/B testing and experimentation platforms
Products (17)
Latest Feature Flagging and Experimentation Updates
- Oct 16, 2026
- Date parsed from source:Oct 16, 2026
- First seen by Releasebot:Oct 16, 2026
@vercel/[email protected]
Flags SDK skips connected projects when definitions can’t be fetched, preventing build failures.
Patch Changes
- #553 bc7a562 Thanks @vincent-derks! - Skip connected projects whose definitions cannot be fetched instead of failing the build.
- October 2026
- No date parsed from source.
- First seen by Releasebot:Oct 16, 2026
What's new, and why it matters
ABTasty shares its latest platform features, improvements, and updates, with context on their value for teams.
Discover the latest features, improvements, and platform updates, with the context behind every release and the value they bring to your team.
Original source All of your release notes in one feed
Join Releasebot and get updates from Vercel and hundreds of other software products.
- Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 15, 2026
v2.286.0
Flagsmith adds experimentation improvements, including removing flag reads from the dashboard and plan-gating bring-your-own warehouses, while fixing a self-hosted GitHub 404 issue and updating warehouse documentation.
2.286.0 (2026-10-15)
Features
- experimentation: drop experimentation flag reads from the dashboard (#8772) (4a92f99)
- experimentation: plan-gate bring-your-own warehouses (#8779) (03fba58)
Bug Fixes
- GitHub: "Add to organisation" opens a 404 page on self-hosted (#8782) (59523a6)
Docs
- experimentation: add Databricks bring-your-own-warehouse guide (#8708) (19ff3fe)
- experimentation: add experimentation sizing to self-hosting sizing guide (#8765) (11cf3f0)
- experimentation: Cover self-hosted warehouses (#8780) (77082ff)
- experimentation: Update the Helm chart docs for warehouse delivery (#8777) (4700912)
- Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 15, 2026
v2.285.0
Flagsmith releases updates to experimentation, including warehouse controls and improved rollout handling, plus a new Accordion component, clearer usage overage details, and fixes for flags, segments, and dark-mode setup.
2.285.0 (2026-10-15)
Features
- ds: add the Accordion component (#8714) (cfaf7ba)
- experimentation: hide the managed Flagsmith warehouse on self-hosted (#8761) (a630eb9)
- experimentation: only block internal warehouse hosts on SaaS (#8760) (f451ebc)
- experimentation: replace flag gates with one experimentation_warehouses flag (#8771) (0480eef)
- tokens: adopt the design system colour system (#8594) (70eee7a)
- usage: say whether an overage is covered or charged (#8653) (ff1f862)
Bug Fixes
- experimentation: Remove superseded experiment rollouts on start (#8770) (78d77d1)
- Flags: Enforce the segment override limit and run delete hooks in the flag API (#8729) (0934ae4)
- Flags: Segment overrides can be transplanted (#8769) (c6cfc65)
- Segments: Refuse system segment changes through the flag API (#8728) (75760e0)
- Segments: Refuse system segment changes through the legacy versioning APIs (#8730) (8777622)
- self-hosted: theme the first-run setup steps in dark mode (#8755) (cd2bf41)
Dependency Updates
- api: update dependency flagsmith-common to v3.16.0 (#8746) (6f41b05)
Docs
- Add documentation on how to add the ingestion server (#8676) (486d78e)
- experimentation: drop Enterprise-only beta notice (#8764) (7cf5933)
Refactoring
- Experiments: Write rollouts through the flag API (#8731) (1def437)
- Oct 14, 2026
- Date parsed from source:Oct 14, 2026
- First seen by Releasebot:Oct 14, 2026
Introducing the official GrowthBook Slack app
Growthbook releases its Slack app in beta, bringing AI-powered answers, experiment and feature flag alerts, approvals, and results into team channels. The app helps teams collaborate on experiments and make evidence-based decisions where they already work.
Great tools and data are critical to an experimentation program, but they’re not enough on their own. A strong program also needs a culture where teams experiment with rigor, share ideas, learn from results, and make decisions based on evidence.
For a lot of organizations, much of that culture lives in Slack. That’s where teams discuss what to build and decide what’s next. But experiment updates don’t always make it into those conversations, which makes it harder to act on what teams learn.
That’s why we’re so excited about the GrowthBook Slack app released in GrowthBook 5.1. About to launch an experiment? The right people see it in Slack and can weigh in before it goes live. Had a big win (or a loss worth learning from)? The results are automatically shared in your team’s channel.
By bringing experiments, feature flags, and AI-powered answers about your data into the conversations where teams already collaborate, the Slack app encourages broader participation and makes experimentation a more natural part of how your organization operates.
What the Slack app does
Chat with GrowthBook’s AI assistant directly from Slack
Tag @GrowthBook in any channel to ask about your experiments and get answers without opening GrowthBook. It's the same AI assistant you use in the app, now available in Slack.
Ask for a quick status check like “@GrowthBook how is the homepage hero test doing?" Or explore past experiments with “@GrowthBook give me a list of all the experiments that tested our checkout flow.” You can even ask broader questions like “@GrowthBook what are the key Learnings we have so far this year?”
Get real-time, granular notifications from GrowthBook
Stay on top of experiments and flags without constantly checking GrowthBook. Get immediate alerts for important events, including SRM issues, significance results, and pending approvals. Filters by project, environment, or tag keep each channel focused on what’s relevant.
What changed: Activity and change trackingKeep teams informed as experiments and flags evolve, from experiment launches and edits to flag revisions and ramp schedule changes.
Potential issues: Health and data qualitySurface issues that need attention, including sample ratio mismatch (SRM), multiple exposures, missing data, underpowered results, failed updates or publishes, and failing guardrails.
What we learned: Results and decisionsBring significance results, decision points, and bandit variation weight changes into the conversation so teams can triage and decide what to do next.
Someone needs to act: Reviews and follow-upsKeep approval requests for flags and experiments visible so they are more quickly actioned. Get updates when experiments are ready for approval, ready to launch, or have been running longer than 90 days.
Building a culture of experimentation
A strong culture of experimentation is one where teams challenge assumptions, test ideas, and let evidence guide decisions. It requires getting the right people involved, designing experiments that maximize learning, making insights accessible so teams can make decisions collaboratively in the moment, and ensuring those insights guide future decisions to shape the longer-term strategy. It also means giving evidence more weight than the most senior opinion in the room.
Bringing experimentation into the Slack channels where teams already collaborate helps make these practices part of everyday decision-making across the organization
Bring more people into experimentation
Experimentation works best when it extends beyond just the teams running tests. Product, engineering, design, growth, marketing, and leadership all have a role in identifying opportunities, challenging assumptions, and deciding what to do next.
But involving those teams at the end of an experiment, if at all, limits their ability to contribute. Bringing relevant updates into Slack gives stakeholders visibility while there's still time to offer context, raise concerns, or influence a decision. This helps keep people involved throughout the process, rather than bringing them in only when results are ready.
Make data-driven decisions together
Experiments often reveal trade-offs that aren't obvious from a single metric. A change that lifts conversions might also increase cancellations a month later. Weighing that trade-off well takes input from across the business. For example, customer success can share what they're hearing about churn and finance can speak to what a cancellation costs.
The GrowthBook Slack app gives teams a natural place to discuss and work through those trade-offs. When results are posted to a shared channel, people can share their perspectives in the moment, add context, and weigh in on conversations, helping teams make informed decisions together.
Turn experiment insights into better business decisions
Experiments can inform broader business decisions. For example, an online grocery store might test different ways of featuring products and find that one approach drives fewer sales but shifts customers toward higher-margin items. That insight could influence everything from inventory choices to how much to spend on marketing.
Sharing those findings across the organization helps teams build on what they've learned and identify new opportunities to test. The GrowthBook Slack app helps drive collaboration by making results and insights accessible in the channels where teams already work. Combined with GrowthBook's Learnings, it helps teams use those insights in future experiments and strategy.
What's next
The GrowthBook Slack app is currently in beta and we’re continuing to improve it. Future enhancements will include more proactive AI capabilities, like regularly summarizing your top experiments and new learnings, surfacing what needs your attention, and flagging issues before you have to go looking for them.
We’d love to hear what's working well for you and what you'd like to see next. Join our Slack App GitHub discussion to share your feedback and help shape what’s ahead.
Bring experimentation into the conversation
Experimentation has the greatest impact when insights reach the people who can act on them. The GrowthBook Slack app brings launches, results, and AI-powered answers into the conversations where teams make decisions.
Original source - Oct 13, 2026
- Date parsed from source:Oct 13, 2026
- First seen by Releasebot:Oct 14, 2026
Version 4.42.1
ABTasty fixes a redirection-loop issue triggered by page URL updates, helping keep user journeys uninterrupted.
AI Generated Release Note
Bug Fixes
- Redirection Loop Prevention: Fixed an issue where users experienced a redirection loop when the page URL updated, ensuring uninterrupted user experience.
- Oct 13, 2026
- Date parsed from source:Oct 13, 2026
- First seen by Releasebot:Oct 13, 2026
v2.284.0
Flagsmith adds a Databricks warehouse connection form, hides unavailable experimentation UI, and improves setup guidance and signup conversion tracking.
2.284.0 (2026-10-13)
Features
- experimentation: add the Databricks warehouse connection form (#8725) (caa24ed)
- experimentation: hide experimentation UI when the experimentation infra isn't deployed (#8749) (bb9cd36)
- Signup: Track signup experiment conversion on activation (#8674) (09f3764)
Bug Fixes
- experimentation: order ClickHouse setup SQL and show experiment snippets for all supported SDKs (#8737) (d5147f4)
- Oct 13, 2026
- Date parsed from source:Oct 13, 2026
- First seen by Releasebot:Oct 13, 2026
v2.283.0
Flagsmith adds Databricks experimentation support, project and environment usage breakdowns, and improvements to API errors and flag writes.
2.283.0 (2026-10-13)
Features
- experimentation: add a Databricks statement client (#8700) (a1ace4a)
- experimentation: add the Databricks warehouse provider (#8703) (c3cff21)
- experimentation: gate experimentation on EXPERIMENTATION_ENABLED (#8747) (6337405)
- usage: break usage down by project and environment (#8654) (8ccdded)
Refactoring
- API: Render APIErrorDetail bodies from a common APIError base class (#8726) (c121827)
- Flags: Write flags through the flag API with AuthorData (#8727) (c092cd6)
- Oct 12, 2026
- Date parsed from source:Oct 12, 2026
- First seen by Releasebot:Oct 12, 2026
v2.282.0
Flagsmith adds updated_at to Trait and refactors experimentation results queries with a SQL dialect.
2.282.0 (2026-10-12)
Features
- traits: add updated_at to Trait (#8665) (027ff84)
Refactoring
- experimentation: extract a SQL dialect from the results queries (#8690) (a47f91b)
- Oct 12, 2026
- Date parsed from source:Oct 12, 2026
- First seen by Releasebot:Oct 12, 2026
v2.281.0
Flagsmith releases expanded feature management and experimentation tools, adding a Dependencies tab, easier flag prerequisite removal, managed feature segments, and richer usage visibility. The update also brings bug fixes, scheduled change corrections, and warehouse-backed experiment reporting.
2.281.0 (2026-10-11)
Features
dependencies: add the Dependencies tab (#8637) (7ca6598)
experimentation: read experiment results and exposures through the environment's warehouse (#8683) (fa52035)
Flag Dependency: Remove a feature's prerequisite (#8650) (e588efa)
Segments: Expose managed feature segments (#8699) (1419461)
usage: expose API limit restriction, grace period and overage billing state (#8641) (fa8a60c)
usage: group usage data by project or environment (#8645) (76c4802)
Bug Fixes
Change Requests: Scheduled feature changes inconsistent with actual state (#8680) (618f0fb)
Feature Lifecycle: Release for public use (#8694) (6b0a8d1)
features: lock the initial state and value when defaults are prevented (#8664) (da773ed)
Dependency Updates
docs: Bump serialize-javascript to 7.1.2 [security] (#8672) (a21c612)
update dependency flagsmith-common to v3.15.1 (#8239) (3166780)
CI
frontend-deploy: name what is being deployed (#8431) (68ce670)
Docs
scim: Document that Entra's aadOptscim062020 flag is not supported (#8621) (a1522da)
Refactoring
experimentation: introduce warehouse providers and port ClickHouse (#8661) (edfb91d)
experimentation: move managed warehouse reads into the Flagsmith provider (#8677) (7dfcd9a)
Original source