Reflag (formerly Bucket) Review (2026)

What Reflag actually is

First, the name. Reflag is the tool that used to be called Bucket. It rebranded on 4 September 2025, and bucket.co now 301-redirects to reflag.com. Bucket started in 2021 as a feature-metrics and feedback product - the name referenced the “leaky bucket” of user drop-off - pivoted into feature management in June 2024, then became Reflag. Same company, current name. If you’ve read about Bucket anywhere, this is it.

Now the product. Reflag is feature flags purpose-built for B2B SaaS, on TypeScript. Its standout is native feature entitlements: you gate flags at the company or subscription level and map features to subscription tiers, not just to individual users. On top of that it does feature adoption metrics and user feedback. The whole thing is scoped tightly around one buyer - a B2B SaaS product team that needs per-account entitlements.

This is a narrow tool on purpose, and that’s the entire story - both the appeal and the risk.

The one thing it’s genuinely good at: entitlements

Most flag tools think per user. You roll a feature out to 10% of users, then 50%, then everyone. Reflag thinks per account. In B2B SaaS, the unit that matters is usually the customer company and the plan they’re on. Reflag lets you tie a feature to a subscription tier, so every account on that tier gets it, and gate access at the company level.

That maps directly onto how SaaS pricing and packaging actually works. If your real problem is “customers on the Growth plan should get this feature, customers on Starter shouldn’t,” a general flag tool makes you build that yourself. Reflag treats it as the core use case. It’s TypeScript-first, ships a toolbar and CLI, and has an MCP server for AI-assisted flagging - a modern, developer-friendly package for its niche.

Pricing: published and simple

Reflag lists its prices, and the quota unit is MTU - Monthly Tracked Users:

Plan	Price	MTUs	Notes
Starter	$0	1,000	3 environments, unlimited seats, no credit card
Pro	$100/mo annual ($125 monthly)	10,000	Unlimited environments and seats, user permissions
Enterprise	Custom	Custom	Migration support, SAML

Extra MTUs on Pro run around $120 to $150/mo per additional 25,000. It’s clean, and there’s no sales wall to see the numbers - a welcome contrast to the enterprise vendors in this category. The pricing fits its audience: small enough teams that $100/mo is reasonable, scaling by tracked users rather than seats.

The gotcha: narrow scope plus early-stage risk

Here’s what to weigh honestly. Reflag’s focus is a strength for the right team and a hard limit for everyone else. If you’re not a B2B SaaS on the TypeScript stack, this isn’t your tool. There’s no A/B experimentation at all. There’s no self-hosting and no open-source option. And the full multi-language SDK breadth isn’t even published - it’s deliberately TypeScript and JavaScript centric, so if your backend is Go or Python or your app is native mobile, the coverage may not be there.

Then there’s company risk. Reflag is small - around $3.7M raised - and it just rebranded from Bucket in September 2025. That’s a small team with a thin public track record against well-funded incumbents. It has real customers, including GitBook, Pleo, and Code Climate, which is a good signal for a company its size. But betting a multi-year roadmap on it is a different risk profile than picking an established vendor.

Reflag versus the field

The honest framing: Reflag wins on B2B-SaaS entitlements and loses on breadth. Against LaunchDarkly or DevCycle, it’s cheaper, simpler, and far better at tying features to subscription tiers - but it does dramatically less, and it’s a fraction of the size. Against experimentation platforms like Optimizely or Statsig, it isn’t even in the same category, because it doesn’t do experiments. It’s not trying to.

Should you use it?

Use Reflag if you’re a B2B SaaS product or engineering team on TypeScript, you need per-account feature entitlements tied to subscription tiers, and you value a focused tool with published pricing over a broad platform.

Don’t use Reflag if you need A/B experimentation, you’re not on the TypeScript or JavaScript stack, you need broad multi-language SDKs, you require self-hosting, or a small, freshly rebranded vendor is more longevity risk than you want to carry.

Bottom line: Reflag does one job - B2B SaaS feature entitlements - and does it cleanly. For that exact team it’s a genuinely good fit and priced fairly. For anyone outside that niche, the missing experimentation, limited SDK breadth, and early-stage size make it the wrong tool. Match it to the job, and don’t stretch it past its scope.

---

Pricing and features verified against reflag.com on 23 July 2026. Feature-flag pricing changes often - we re-verify regularly. Reflag was known as Bucket until its September 2025 rebrand.