Bucket vs LaunchDarkly: The Alternative for B2B

LaunchDarkly is one of the oldest and largest players in the feature management space. Their focus is on the big B2C ecommerce companies of the world. While there are plenty of LaunchDarkly alternatives, there was never one designed especially for B2B. 

That's precisely why we purpose-built Bucket for the B2B use case. This focus helped us create a powerful solution with all the core functionalities of LaunchDarkly that’s simpler to use, easier to implement, simpler to budget, and provides entitlements, feature adoption metrics, and feedback that B2B can actually use. 

This post will give you a deep dive into what makes Bucket different and which platform is better for whom.

TLDR; Bucket is the better option for B2B companies. It’s built to work at the company/organization level right out of the box, is easy to use, gives you feature adoption metrics and customer feedback to iterate faster, get closer to your customers, and is aligned with the needs of B2B (all features we make will be for you).

‍If you’re in B2C, LaunchDarkly is the way to go. Products or online shops where end-users are the primary entity will see the value of LaunchDarkly’s mature A/B experimentation capabilities and device-specific insights

Bucket is designed for B2B 

Bucket and LaunchDarkly do what you expect a feature management solution to do: help development teams roll out new features - gradually and without risk. Both platforms offer the essential feature flagging capabilities required for managing feature flags and test features. This includes targeting rules, segmentation, environments, percentage rollouts, kill switches, feature toggles, and data warehouse integrations. 

Because Bucket is purpose-built for B2B, it only has features that are valuable for B2B companies. This is most noticeable in how we segment users. Bucket focuses on companies while LaunchDarkly focuses on an abstract concept called “contexts”.

Native company targeting rules

Native company-level feature targeting rules let you gradually roll out features to companies rather than users. There are also release stages that signal which rollout stage a feature has reached. 

In Bucket, the concept of turning a feature on/off is built in whereas LaunchDarkly treats every flag as potentially having complex data types that aren’t limited to on/off. That means the targeting user interface is much simpler in Bucket. 

Company-level user aggregation

Bucket automatically aggregates users at the company level. You don’t need to create company segments and ensure you add the right users. You get a birds-eye view of each company on the Companies page.

Bucket also shows you which features are enabled for a given company along with the time at which they were enabled.

Powerful company-level filters

Bucket has powerful filters that let you filter companies by feature access, feature adoption, feature satisfaction, and more.

Custom company segments

Filters let you create and export lists of companies and manage feature access based on criteria ranging from subscription plan type to feature satisfaction. 

Simple company-level reporting

Feature reporting is simple with customizable columns that give you an overview of feature adoption and satisfaction without the need for an API or manually creating reports.

Overall, LaunchDarkly is a versatile and flexible platform. But, it’s built for B2C and ecommerce businesses, first and foremost. While powerful, it has many features that B2B companies can’t use, making it unnecessarily complex. 

Bucket is intuitive to use

Engineers and other technical users implement user-friendly interfaces. They deserve dev tools that return the favor. Bucket’s focus on the B2B use case allowed many unnecessary features and interfaces found in LaunchDarkly to be removed. The result is a well-thought-out, intuitive UI that provides a smooth experience for engineering teams.

For example, the feature tab acts as the mission control for feature releases.

Feature views are easily customizable, letting teams create as many views as needed.

If there are columns that engineers want to show or hide, the columns are easy to configure.

The simplicity and ease of use offer an excellent user experience that makes it much easier and faster to get up and running.

LaunchDarkly, on the other hand, has a complex UI that’s confusing to navigate when getting started. While plenty of advanced features are found within the various menus and tabs, there’s a steep learning curve for engineers without previous experience on the platform.

Bucket has simple pricing

Bucket’s pricing is based exclusively on Monthly Tracked Users (MTUs). An MTU is just a unique end-user who has sent at least one event to Bucket during a given month, making budgeting more predictable for engineering teams.

While both platforms provide unlimited apps, environments, seats, and metrics, LaunchDarkly's pricing model is based on a combination of contexts, service connections, and experimentation keys. 

You need to pay $12 per 1k extra monthly contexts (1k included), $12 per service connection, and $33 per month per 10k experimentation keys

Example

Acme Inc. is rolling out new features to their web app, android app, and iOS app with React, Javascript, Node.js, and Ruby SDKs. During that month, each device had 5k users who visited an average of 3 times across all the devices. Acme Inc. also collected feature adoption metrics for features that were accessed by 12k users. 

This is how much it would cost on each platform: 

In the end, Bucket costs 50% less.‍

This scenario only covers the publicly available pricing for LaunchDarkly. Data export events and larger volumes of contexts, service connections, and experimentation keys have no listed pricing. 

Overall, B2B companies find more value in Bucket because they only pay for features that they can use and they can reasonably forecast how much it will cost.

Bucket provides customer adoption metrics & feedback

LaunchDarkly is one of the industry leaders in A/B experimentation. They have powerful A/B testing and multivariate testing capabilities that are perfect for B2C products with 10s of thousands of users.

The problem for a B2B company is that A/B experiments are useless. They’ll never have the volume they need to get statistically significant sample sizes. It also doesn’t fit the holistic approach B2B products and engineering teams need to take. Online stores can design by metrics; B2B requires a level of taste that integrates different features into a polished package. In the end, B2B engineering teams end up paying for a tool that they can’t use.

Bucket takes a different approach to provide B2B companies with feature adoption metrics they can actually use and combines it with qualitative feedback that gives them the context that numbers alone can’t. 

After rolling out a feature internally, the next step is to release it to a segment of beta customers. This is when you want to start collecting adoption metrics and feedback.

LaunchDarkly lets you send events to collect custom metrics to measure conversion rate or sales value, for example. Buckets lets you easily collect real-time feature usage metrics and crunches them using an open-source framework called STARS (Segment, Tried, Adopted, Retained, Satisfied). It’s a B2B-optimized funnel that measures adoption and satisfaction specifically for B2B features. The power is in its simplicity:

• If a user tries a feature once, their company is in the Tried stage.
• If users at the company use a feature more than x number of times, they’re Adopted.
• If they continue using the feature at least x amount of times per period, they’re Retained.

Bucket can also automatically collect CSAT scores and feedback on features, but only after a customizable number of interactions, to give the adoption metrics context.

When put together, the adoption metrics and feedback let engineers uncover issues quickly instead of waiting for feedback to trickle down from customer success to product managers. They are able to iterate based on what they have learned and then roll out the feature to all customers.

Practically speaking, Bucket lets engineers turn themselves into Product Engineers who don’t just ship into the void but actively contribute and even lead the direction of product development.

Which is better?

Ultimately, it depends on a company's specific requirements and the industry in which they operate. Bucket and LaunchDarkly are both capable feature flag management platforms that offer the same core capabilities. Where they differ is their approach to feature management, segmentation, and experimentation vs adoption metrics and customer feedback.‍

Bucket is the better choice for B2B companies. It’s purpose-built for the B2B use case, working at the company level right out of the box. It’s also simple to get started with, more cost-effective, and provides engineers with usable feature adoption metrics and customer feedback that help them actively contribute to and shape product roadmaps. 

‍LaunchDarkly is better for B2C. It handles complex A/B testing scenarios across multiple B2C and ecommerce applications. Yes, its UI can be confusing when getting started and it can get expensive, but its experimentation capabilities make it the right choice for B2C companies.