member-console/design/documents/reference-open-source-billing-platforms.md

Open-source billing platforms: who's genuinely FOSS?

Of the three billing platforms examined — Billabear, Lago, and Kill Bill — only Kill Bill ships under a genuinely permissive open-source license with no strings attached. Lago uses the OSI-approved AGPLv3 but operates an open-core model that gates meaningful features behind commercial licenses. Billabear is the least open of the three, using a source-available Fair Core License that prohibits competitive use entirely. All three monetize through some form of proprietary layer or premium gating, but the degree of openness varies dramatically. For teams evaluating these platforms on FOSS principles, the differences matter — and the devil is in the licensing details.

The license spectrum runs from permissive to restrictive

Kill Bill uses the Apache License 2.0, one of the most permissive OSI-approved licenses available. It has maintained this license without interruption since its founding in 2010 — a remarkable record of stability given the wave of relicensing that swept the infrastructure software industry (MongoDB → SSPL, HashiCorp → BSL, Elastic → SSPL). Apache 2.0 includes a patent grant and imposes no restrictions on commercial use, modification, or redistribution. The admin UI components use the MIT License, equally permissive.

Lago uses the GNU Affero General Public License v3 (AGPLv3), a strong copyleft license that is also OSI-approved. AGPL's key requirement is that anyone running modified Lago code as a network service must make their modifications available to users. Lago chose AGPLv3 deliberately from day one to prevent competitors from forking and commercializing the code without contributing back — a rationale they documented in a detailed 2022 blog post. Client SDKs use the MIT License. Importantly, Lago has never changed its license, despite initial assumptions to the contrary.

Billabear uses the Fair Core License (FCL), a source-available license created by Keygen. This is not an OSI-approved open-source license. The FCL includes a non-compete clause that prohibits using the software to build a competing commercial billing service. However, each version converts automatically to Apache 2.0 on its second anniversary — a "Delayed Open Source Publication" (DOSP) mechanism. Billabear previously used the Business Source License (BSL 1.1) and switched to FCL around mid-2024. The founder himself describes the project as "source available" rather than "open source."

Platform	License	OSI-approved?	Non-compete clause	Auto-converts to FOSS?
Kill Bill	Apache 2.0	✅ Yes	No	N/A — already FOSS
Lago	AGPLv3	✅ Yes	No (copyleft instead)	N/A — already FOSS
Billabear	Fair Core License	❌ No	Yes — prohibits competing use	Apache 2.0 after 2 years

Every project gates features, but the scope differs enormously

All three platforms employ some form of feature gating, but what gets locked away reveals their true priorities.

Kill Bill keeps the most in the open. Its core engine includes RBAC with fine-grained permissions, comprehensive audit trails, native multi-tenancy, webhook/event bus systems, and analytics plugins — all Apache 2.0 licensed. The proprietary layer, branded Aviate, focuses on operational acceleration: API-driven dynamic catalog management (replacing XML configuration), built-in metering, wallet/prepaid credits, coupons, tax computation, and pre-configured cloud infrastructure. The distinction is meaningful — Kill Bill gates convenience features behind Aviate, not security or compliance features. The tradeoff is time: self-hosting the OSS version takes 3–6+ months to production versus weeks with Aviate.

Lago takes the open-core approach more aggressively. While the core billing engine (metrics, usage ingestion, plan configuration, invoice generation, payment integrations, webhooks) is AGPLv3, several practically important features require a commercial premium license: customer-facing portals, branded invoices with custom SMTP, progressive billing, tax integrations, CRM/CPQ/accounting integrations, automatic dunning, data pipeline integrations, and revenue sharing. Notably, the latest pricing structure restricts self-hosted deployment itself to the Enterprise tier — meaning teams wanting to run Lago on their own infrastructure must negotiate a commercial agreement. The Helm chart configuration includes a global.license field for activating premium features via license key.

Billabear embeds license-key-gated functionality directly in its single codebase. The FCL license explicitly states users "must not move, change, disable, or circumvent the license key functionality." The exact scope of locked features is not publicly documented, creating transparency concerns. Additionally, the self-hosted version receives feature releases only every six months, while the cloud version gets continuous updates — a soft gate that nudges users toward the paid SaaS.

None of the three platforms appear to impose an "SSO tax" in the traditional sense. Kill Bill's RBAC is fully open-source. Lago lists RBAC on its pricing page without clear tier restrictions. Billabear references a "powerful permission system" but details are sparse. None of the three projects were found to require a Contributor License Agreement (CLA), which means relicensing community contributions would be difficult — a positive signal for FOSS sustainability.

Three distinct monetization philosophies

Lago follows the venture-backed open-core SaaS model. With $22M in funding and YC S21 backing, Lago monetizes through its hosted cloud platform (Business and Enterprise tiers with "contact us" pricing), self-hosted premium licenses, and premium add-ons. They emphasize that they do not take a percentage of customer revenue, positioning against Stripe Billing's 0.5–0.8% cut. The "contact us" pricing across all tiers has drawn criticism for opacity.

Kill Bill takes a consulting-and-product hybrid approach. The team explicitly refuses to offer Kill Bill as a SaaS, stating it "goes against our value proposition" of data ownership. Revenue comes from Aviate (proprietary enterprise product with fixed pricing), consulting at $800 per four-hour block via private Slack channels, GitHub Sponsors for priority support, AWS Marketplace listings, and an implementation partner ecosystem. This model is unusual — it prioritizes the open-source core while building proprietary accelerators on top.

Billabear operates as a solo-founder SaaS business under Humbly Arrogant Software Limited (UK). The cloud offering uses revenue-based pricing: 1% of revenue processed on the Starter tier (with $50,000 free), flat $250 per $200,000 on Professional, and custom Enterprise pricing. Self-hosted users can purchase license keys for premium features and paid support contracts. The company has also offered lifetime deals (£500 GBP during the v1.1 release).

Revenue stream	Kill Bill	Lago	Billabear
Hosted SaaS	❌ Explicitly refused	✅ Primary model	✅ Primary model
Proprietary enterprise product	✅ Aviate	✅ Premium license keys	✅ License keys
Consulting	✅ $800/4-hour blocks	✅ Solutions engineers	✅ Contact sales
Support contracts	✅ Via GitHub Sponsors	✅ Tiered support	✅ Paid support
VC funding	❌	✅ $22M raised	❌

Community trust varies with transparency

Kill Bill commands the most community respect, largely through longevity. Active since 2010 and used by Square, Groupon, HP, Rakuten, and Equinix Metal, it has earned credibility through durability. A substantial Hacker News thread (435 points, 166 comments) in 2022 revealed both admiration and frustration. The main criticism came from a multi-year user who described the software as "an industrial tool with a lot of quirks and sharp edges" and raised transparency concerns about the core team's priorities between open-source and proprietary work. Documentation gaps and steep learning curves were recurring complaints. Notably, no one accused Kill Bill of bait-and-switch tactics or questioned the Apache 2.0 commitment.

Lago generates the most active community discussion, with 9,200+ GitHub stars and multiple Hacker News front-page appearances. Sentiment runs positive on the core technology but critical on pricing transparency. One HN commenter captured a common frustration: "It's a massive shame that simple table stakes features are behind 'get a quote.'" The decision to restrict self-hosted deployment to the Enterprise tier particularly rankles developers who assumed "open source" meant they could freely self-host. Default opt-out telemetry on self-hosted instances has also drawn minor criticism. That said, Lago's consistent AGPLv3 licensing — with no changes or threats of changes — provides genuine stability.

Billabear has the smallest community footprint at ~714 GitHub stars with minimal external contributions. Its June 2023 Hacker News launch (107 points) surfaced licensing debates, pricing opacity complaints, and code quality criticism. Community members initially bristled at the BSL license, though some warmed to the delayed open-source model. The switch to FCL in 2024 received a more positive reception. The project's transparency about being "source available" rather than "open source" is commendable, though some marketing pages still position it as an "open-source alternative" — a tension the founder acknowledges.

What this means for choosing a platform

The choice between these three platforms ultimately reflects what kind of "open" matters to your organization. Kill Bill is the safest bet for FOSS purists: Apache 2.0, no license changes in 15 years, and core features like RBAC, audit logs, and multi-tenancy fully in the open — but you'll pay in complexity and implementation time. Lago offers the most modern developer experience with genuine AGPLv3 licensing, but its aggressive open-core gating means the free version lacks features many teams consider table stakes. Billabear is honest about being source-available, not open-source, and best suited for teams that simply want affordable billing software rather than a FOSS-principled foundation.

The broader pattern here reflects an industry-wide tension: billing is complex enough that maintainers need sustainable revenue, but the mechanisms they choose to generate it — feature gating, proprietary layers, source-available licenses — erode the very openness that attracted users in the first place. Kill Bill's 15-year Apache 2.0 commitment proves it's possible to sustain a genuinely open billing platform, even if it requires accepting slower feature velocity and higher implementation overhead as the price of real freedom.