What do you like best about the product?

It does what it says it does, and does it well. It's easy to add new flags. It's easy to configure how the flags are calculated, add custom test groups, percentage rollouts, etc. Performance is great, I've never seen the service down. Pricing is generous. In this age of build vs buy, LaunchDarkly is still a a Good Buy.

What do you dislike about the product?

Sometimes the UX is frustrating. Hard to remember where to go to do what i want.

What problems is the product solving and how is that benefiting you?

We need to roll out features slowly before turning them on to the full audience. This solves that, and does it very well.

What do you like best about the product?

What I like most about LaunchDarkly is how cleanly it decouples deployment from release. We can ship code to production behind a flag and then roll it out gradually—to internal users first, then a small percentage of customers, and finally everyone—without any redeploys. The targeting rules (by user attribute, segment, or environment) are powerful yet still straightforward to configure in the dashboard, so even non-engineers like PMs and support can flip flags safely.

The kill-switch capability has saved us more than once: when something misbehaves in production, we can simply toggle the flag off instead of rushing a hotfix or doing a rollback. The SDKs for all the major languages we use have been reliable, and flag changes propagate to clients in near real time.

What do you dislike about the product?

Pricing can get steep as you scale. Adding seats and unlocking certain features (like experimentation and advanced analytics) often nudges you toward an enterprise plan, which can be tough to justify for smaller teams. The UI also comes with a real learning curve: given the depth of features across projects, environments, segments, and targeting rules, new team members usually need some time before they feel comfortable navigating everything.

What problems is the product solving and how is that benefiting you?

LaunchDarkly solves the core problem of releases being risky, all-or-nothing events. By decoupling deployment from release, we can merge and deploy code continuously while keeping new features hidden behind flags until they're ready. This has dramatically reduced the stress around shipping — there's no more bundling features into big releases or scheduling late-night deploys.
The targeting and gradual rollout capabilities let us release to internal users, then beta customers, then a small percentage of production traffic before going 100%. If something breaks, we flip the flag off instantly instead of rolling back or hot-fixing. This has cut our incident recovery time significantly and given engineers the confidence to ship more often.
It's also empowered non-engineering teams — PMs control beta access, support can toggle entitlements, and marketing can time feature launches to campaigns without needing engineering involvement for every change.

What do you like best about the product?

We use LaunchDarkly to more confidently release new features. By feature flagging new features/logic, we can enable features for a subset of users/accounts before a full release; this gives us the confidence that 1. features can we rolled back at the flip of a switch if any issues arise and 2. that features have been tested in a production environment before making them visible to all users. LaunchDarkly's UX is simple and allows us to quickly enable features for specific users/accounts and easily track how the feature flags have been evaluating in a given environment. LaunchDarkly gives us peace of mind every time we release a new feature.

What do you dislike about the product?

AI features feel like an afterthought and somewhat unnecessary.

What problems is the product solving and how is that benefiting you?

Controlled rollout of new features, giving us peace of mind that we can instantly rollback anything broken and fully test new features before they're enabled for all users.

What do you like best about the product?

I like that LaunchDarkly is fast and works well with AI. The Slackbot makes it easy to open tickets from Slack conversations, and the snappy UI makes it effective to work with the product. Additionally, the initial setup was easy.

What do you dislike about the product?

not much

What problems is the product solving and how is that benefiting you?

LaunchDarkly is fast and integrates well with AI. The Slackbot feature allows easy ticket creation from Slack, and its snappy UI enhances productivity.

What do you like best about the product?

I appreciate how LaunchDarkly lets us differentiate between which models we serve to different customers. It allows us to transition seamlessly and smoothly whenever we have to do more complex migrations, enabling us to transition customers individually. I also value the bulk edits feature.

What do you dislike about the product?

More advanced bulk edits are not included in the standard paid plan.

What problems is the product solving and how is that benefiting you?

I use LaunchDarkly to gate certain features with feature flags, manage customer models, and transition smoothly during complex migrations.

What do you like best about the product?

Overall great Experience but some highlights are ,
LaunchDarkly: A Comprehensive Review
Here’s a detailed breakdown of LaunchDarkly's key aspects, focusing on the features and benefits that users most frequently highlight.

UI / UX: Intuitive and Empowering
One of the most consistently praised aspects of LaunchDarkly is its user interface. Instead of just being "easy to use," the platform's design has a direct impact on workflow efficiency.

Centralized Flag Management: The main dashboard provides a single source of truth for all feature flags. This clarity is invaluable in a complex microservices architecture where tracking feature states could otherwise be a nightmare. For example, a developer can instantly see that the "new-checkout-flow" is active for 5% of users in the UK, while the "beta-reporting-engine" is only on for internal staff. This eliminates ambiguity and the need to dig through code or config files.

Targeting and Segmentation: The user targeting rules engine is incredibly powerful yet simple to operate. You can create complex segments with a few clicks. For instance, you could roll out a new feature to "users in Germany on an iOS device who are part of the 'Pro' subscription tier." This granularity allows for precise, safe testing and reduces the risk associated with big-bang releases.

Toggle and Flag Status Visibility: The visual toggles and clear status indicators (e.g., "Active," "Inactive," "Launched") mean that even non-technical team members, like product managers or marketers, can understand the state of a feature and even control its release. This democratizes the release process and frees up engineering resources.

Integrations: A Connected Workflow
LaunchDarkly's value is significantly amplified by its extensive ecosystem of integrations, which embed feature flagging directly into the existing development lifecycle.

CI/CD and Code Repositories: Integrations with tools like Jira, GitHub, and Slack create a seamless feedback loop. A practical example is creating a Jira ticket that automatically generates a corresponding feature flag in LaunchDarkly. When a pull request is merged in GitHub, the flag can be automatically enabled in a staging environment.

APM and Observability: The integrations with platforms like Datadog, New Relic, and Dynatrace are a game-changer for performance monitoring. When a new feature is rolled out, you can overlay LaunchDarkly events (like "flag turned on") onto your performance graphs. If you see a spike in latency or errors that coincides perfectly with a feature release, you can immediately disable the flag with a single click in Launch_Darkly, effectively "killing" the problematic feature without a frantic rollback or hotfix deployment.

Unexpected Benefit: The Slack integration provides more than just notifications. It allows teams to manage flags directly from a Slack channel. For instance, during an incident, an engineer could type a command like /ld-kill-flag production new-api-integration to immediately mitigate an issue without ever leaving the incident response channel.

Performance: Negligible Overhead
A common concern with any third-party SDK is performance overhead. LaunchDarkly is engineered to minimize this.

Streaming Architecture: Instead of making a remote call for every flag evaluation, the LaunchDarkly SDKs establish a streaming connection (Server-Sent Events) to receive flag updates. This means that once the initial connection is made, all flag evaluations are performed in-memory at near-zero latency. For a high-traffic e-commerce site, this ensures that the user experience is not degraded by waiting for a feature flag service to respond.

Resilience and Fallbacks: The SDKs are designed with resilience in mind. If the connection to LaunchDarkly's servers is ever lost, the SDK will continue to serve the last known set of flag values. This ensures that your application continues to function predictably, even in the event of a network partition or an issue with LaunchDarkly's service.

Pricing / ROI: From Cost to Investment
While LaunchDarkly is a premium product, the return on investment is often justified by risk reduction and increased development velocity.

Decoupled Deploy and Release: The core value proposition is the ability to decouple code deployment from feature release. This means developers can merge and deploy code to production continuously, even if the features aren't ready for users. The code sits dormant behind a flag. This eliminates the stress and risk of "big bang" release days. The ROI here is measured in reduced deployment-related incidents, fewer rollbacks, and less developer time spent on managing complex branching strategies.

Saved Engineering Hours: Consider the time it takes to build a robust, in-house feature flagging system with a user-friendly UI, audit logs, and complex targeting rules. This is a significant engineering effort. By using LaunchDarkly, that time is instead spent on building core product features that deliver direct value to customers. The subscription cost is often a fraction of the cost of the engineering salaries that would be required to build and maintain a comparable internal solution.

Support / Onboarding: A True Partnership
Users frequently report positive experiences with LaunchDarkly's support and documentation.

Comprehensive Documentation: The developer documentation is clear, with copy-and-paste examples for every supported language and framework. This drastically reduces the time to get started. A developer can typically have the SDK integrated and their first feature flag operational within an hour.

Responsive and Knowledgeable Support: When issues do arise, support is noted to be responsive and staffed by engineers who understand the product deeply. This is a significant step up from basic first-line support and is crucial when dealing with a service that is so integral to the production environment.

AI / Intelligence: The Future of Flagging
LaunchDarkly is beginning to incorporate intelligence into its platform to move from reactive to proactive feature management.

Experimentation and A/B Testing: The platform's experimentation features allow you to tie feature flags to business metrics. For example, you can roll out a new "Add to Cart" button design to 10% of users and measure its impact on the conversion rate compared to the old design. The platform handles the statistical analysis and tells you if the new feature is a winner, a loser, or inconclusive.

Future Direction: While not fully "AI" in the generative sense, the direction is towards intelligent automation. This includes features that can automatically detect performance regressions caused by a feature release and potentially even automatically disable the flag. This moves towards a self-healing system where the platform itself helps ensure application stability.

What do you dislike about the product?

Nothing so far—everything looks good to me.

What problems is the product solving and how is that benefiting you?

Problem 1: Risky, Stressful, All-or-Nothing Releases
Traditional “release day” tends to be a high-stakes, high-stress event. A huge batch of new code goes live at once, and if anything breaks, it turns into a frantic scramble to roll back, which can mean downtime and frustrated engineers.

LaunchDarkly’s Solution: It decouples code deployment from feature release. Your team can continuously deploy code to production servers, while the new functionality stays dormant behind a feature flag. The feature only becomes visible to users when you flip the switch in the LaunchDarkly dashboard.

How That Benefits You:
Fearless Deployments: You can deploy code on a Tuesday afternoon with confidence, knowing it won’t affect users until you’re ready. This removes much of the stress and ceremony around “release day.”

Instant Rollbacks (Kill Switch): If a feature causes a bug or performance issue after release, you don’t need to redeploy or roll back code. You simply flip the flag off. Your Mean Time To Recovery (MTTR) drops from hours to seconds, giving you a strong safety net.

Problem 2: Slow Development Velocity and “Merge Hell”
Long-lived feature branches are a common bottleneck. They’re hard to merge, they delay feedback, and they make parallel work more difficult, which slows the whole team down.

LaunchDarkly’s Solution: It supports Trunk-Based Development. Developers can merge small, incomplete pieces of work into the main codebase (the “trunk”) every day, with those changes safely wrapped in a feature flag.

How That Benefits You:
Increased Development Speed: Features reach a “done” state faster because the team integrates continuously, instead of dealing with the painful merge of a massive, month-old branch.

Improved Code Quality: Small, frequent merges are easier to review and test, which helps reduce the number of bugs introduced into the main branch.

Problem 3: The Gap Between Staging and Production
Even with a strong staging environment, it rarely matches the scale, data, and messy reality of production. As a result, some issues only show up after you’ve released to real users.

LaunchDarkly’s Solution: It enables safer testing in production. You can expose a new, higher-risk feature to a tiny, controlled audience—such as the internal engineering team, a single test account, or 1% of your user base.

How That Benefits You:
Ultimate Confidence: You can validate that a feature works with real production data and infrastructure without putting all users at risk. You catch “production-only” bugs when they can affect only a handful of people (or just you).

Example: You can roll out a new database query or API integration to your internal team first. If it performs well under real-world conditions, you can progressively expand the audience with confidence.

Problem 4: Making Business Decisions Based on Guesses
You ship a feature you believe will improve engagement or conversion, but without a solid testing framework, it’s hard to know whether it actually helped. In that situation, you’re essentially guessing about the business impact of engineering work.

LaunchDarkly’s Solution: The Experimentation framework. You can wrap different versions of a feature in flags, assign them to different user groups, and tie the rollout to a key business metric (e.g., “add to cart” clicks, time on page, subscription sign-ups).

How That Benefits You:
Data-Driven Decisions: You can move from “we think this is better” to “we have proven this new checkout flow increases conversion by 8%.” That makes it possible to measure the ROI of development efforts objectively.

Product-Engineering Alignment: It creates a shared language between product managers and engineers. The focus shifts from simply shipping code to improving key business metrics, helping ensure the team is working on what delivers the most value.

What do you like best about the product?

I like how LaunchDarkly helps with production releases. We can just flip a flag and release a new feature to a beta group or to everyone. If there is an issue, we can easily just turn the flag off to hide the feature. I also find the UI easy to use, and it's clear what the state of the flag means and what environment I am applying the flag change to.

What do you dislike about the product?

Not much. I only use it for feature flags though.

What problems is the product solving and how is that benefiting you?

LaunchDarkly helps with production releases by allowing us to flip a flag and release features to a beta group or everyone. Issues can be managed easily by turning the flag off to hide the feature.

What do you like best about the product?

What I like best about LaunchDarkly is how easy it was to implement in our system. We can manage feature flags consistently across both backend and frontend use cases, which made adoption straightforward for our team.

It is also fairly easy to use once set up. Defining targeting rules for specific contexts, such as project-specific configurations or user segments, is intuitive and gives us good control over gradual rollouts and experiments.

From a performance perspective, we have not noticed any issues. The feature flag evaluation has been reliable and has not introduced any visible impact on our application experience.

What do you dislike about the product?

At first, learning how to set up the targeting rules can be a bit challenging.

What problems is the product solving and how is that benefiting you?

We get an easy to use feature flag management tool, which allows us to easily run expermients with customers to get early feedback.

What do you like best about the product?

I like the target individual section. And the progressive rollout functionality, even though it's not super polished

What do you dislike about the product?

We can not use custom rules because we have to make sure that user and organization IDs are matched. For example, we can not match a user on a flag by their name because we can not guarantee that the name will be part of the metadata at all times. The target individual section helps with this, but it doesn't always propagate with all of the information, so sometimes, if it's a newly created account, for example, we can not find it there. We'll have to trust that the ID is right.

What problems is the product solving and how is that benefiting you?

We're using it to handle feature flags mainly. It's great for us because we can turn on and off features for our clients. For better or worse, we use it as a sort of management tool rather than a pure roll-out tool.

What do you like best about the product?

The biggest advantage of LaunchDarkly is the way wherein it's ability to separate deployment and release in a real prod environment. We use it in our stack (Node.js backend + React frontend in most cases) to control feature rollouts without redeploys.

We can turn on features for internal users first, then for certain customers all e and finally roll out in percentage based releases. This has made biweekly releases a lot safer, especially for high impact changes.

The integration of the SDK was simple and once implemented it became a core part of our release workflow. It also gives product and QA teams control over feature exposure without relying on engineering, which increases overall velocity.

What do you dislike about the product?

Feature flag management can become or becomes a problem if not managed as per best practices. In our case, stale flags started piling up after multiple releases, and without strict cleanup which we have every quarter helps us fix this

Pricing is another concern, it scales quickly as usage grows. During renewal, we didn’t receive the same level of discount that was initially discussed during onboarding, which made cost planning a bit more harder.

What problems is the product solving and how is that benefiting you?

In reality, when we ship a feature, we start it with a disabled feature flag. First, we activate it for our beta users or certain customers through the means of domain targeting, then slowly roll it out to others users periodically all users in production.

We are also dependent on feature flags as kill switches. For instance, in the case where we observe any latency or error issues due to some downstream API or new business logic that we have introduced, we do not need to apply a hotfix but just disable the flag through the dashboard.

We also use segmentation rules (user attributes, environments) to test features with specific cohorts before a full rollout