LaunchDarkly
LaunchDarklyExternal reviews
627 reviews
from
and
External reviews are not included in the AWS star rating for the product.
Highly Recommend LaunchDarkly
What do you like best about the product?
It is very easy to manage feature flags.
What do you dislike about the product?
Sometimes finding support on the website is difficult.
What problems is the product solving and how is that benefiting you?
Being able to manage feature flags.
Easy toggle management
What do you like best about the product?
What I really appreciate about LaunchDarkly is its ability to let us update and test new features directly without the need to constantly push out new code versions with changes or reverting changes. It's super useful because it means we can experiment with different features, see how they perform, and make necessary adjustments in real time. This way, we can make sure everything works perfectly for a small group of users before we roll it out to everyone.
What do you dislike about the product?
The pricing might be a bit on higher side, though it worth for a real CI/CD environment. The SDKs could be a bit better.
What problems is the product solving and how is that benefiting you?
LaunchDarkly helps to safely deploying and manage new features in real-time by providing a reliable toggle service. It help by enabling gradual rollouts, facilitating A/B testing, and allowing for quick reversions without needing to redeploy the entire application. This approach minimises risk, enhances deployment speed and enables personalised user experiences.
Extremely feature rich, intuitive, fast and customisable
What do you like best about the product?
Launch Darkly has a very robust and feature rich platform. Beyond just feature flagging, you have the ability to deeply control how and when features are delivered to your customers, with tools to automate stale flags and code cleanup. They have integrations with all the software that we use which has made integrating that much simpler.
Beyond that, we have a very custom implementation which takes into account cache affecting feature flags on our edge compute platform. The Launch Darkly team have been there to answer our questions and provide hands on support every step of the way. They even took on the task of training our experimentation team on how to use the software.
We have very strict requirements on our edge platform to return results in a tiny timeframe, the Launch Darkly API is incredibly fast, reliable and suitable for this usecase.
Beyond that, we have a very custom implementation which takes into account cache affecting feature flags on our edge compute platform. The Launch Darkly team have been there to answer our questions and provide hands on support every step of the way. They even took on the task of training our experimentation team on how to use the software.
We have very strict requirements on our edge platform to return results in a tiny timeframe, the Launch Darkly API is incredibly fast, reliable and suitable for this usecase.
What do you dislike about the product?
It's very difficult to find things I dislike, but if I had to dig deep I would say that it would be useful to have the ability to separately handle a "holdout" group percentage for each flag, along side an active and inactive state. The process to do this at the moment is farily manual and involves creating a prerequisite flag alongside the experiment.
Due to our custom implementation, it would be an advantage to be able to fetch the "tags" associated with the flags in the same fetch. Currently, we have to handle this ourselves (due to a strict requirement to return results within a tiny timeframe).
Due to our custom implementation, it would be an advantage to be able to fetch the "tags" associated with the flags in the same fetch. Currently, we have to handle this ourselves (due to a strict requirement to return results within a tiny timeframe).
What problems is the product solving and how is that benefiting you?
Our website receives over 400 million user sessions every year, so releasing features can be a risky endevour! Launch Darkly helps you reduce the risk and measure the impact of releasing features in production. Since metrics can be fed directly into Launch Darkly, you can setup automatic safety mechanisms which switch features off when a relevant metric falls below acceptable thresholds. We closely measure our DORA metrics, and Launch Darkly has helped us improve these, especially with Mean Time to Recovery.
Before we had implemented Launch Darkly, our experimentation team had little to no autonomy when it came to changing percentages for slow rollouts. Infact, it was a long and drawnout process which involved a lot of steps and a lot of waiting between those steps. Now that we have our Launch Darkly implementation in place, once an experiment has been created, the experimentation team need only log into the Launch Darkly dashboard and make the changes directly themselves which are then reflected instantly in production.
Before we had implemented Launch Darkly, our experimentation team had little to no autonomy when it came to changing percentages for slow rollouts. Infact, it was a long and drawnout process which involved a lot of steps and a lot of waiting between those steps. Now that we have our Launch Darkly implementation in place, once an experiment has been created, the experimentation team need only log into the Launch Darkly dashboard and make the changes directly themselves which are then reflected instantly in production.
Intuitive platform, enables experimentation teams, great support from account execs
What do you like best about the product?
The platform is easy to use, and is very practical in allowing different teams to have controls over specific parts of the release process. Experimentation can have their roles separately from engingeering and these permissions allow multiple parts of the business to align on feature rollout. There looks to be loads of potential for metrics tracking in the web app, which will be very useful in taking the grunt-work out of analysis. The account team / customer support is great, all of the memebers are very helpful and keen to work together to help you get the most out of the platform and the relationship.
What do you dislike about the product?
LaunchDarkly is very dependent on initial integration and the choices made in set up. It is vital to ensure experimentation are involved when first setting up the platform, if they are to be end users, else the setup choices may hinder experimentation capabilities and create tech debt.
What problems is the product solving and how is that benefiting you?
LaunchDarkly allows the experimentation team to control the rollout of AB tests, as opposed to being bottlenecked by pipelines within the tech stack, creating a much shorter time-to-launch for expeirments - it is essentially the click of a button! Likewise, the control being handed over to the experimentation team allows us to intervene much quicker if a feature/experiment is not going in the direction expected - with an almost instant turn off. The platform makes it easy to create, manage, and have visibility of experiments as well as maintaining accountability through logging changes.
Easy to hit the ground running feature flag and configuration SaaS tool
What do you like best about the product?
In both a personal and professional capacity LD has been easy to integrate with. Their SDKs are plug'n'play.
What do you dislike about the product?
Using launch darkly for configuration can be a bit tedious especially when dealing with JSON variations. Navigating around the JSON input does not feel natural, and the review box (diff tool) does not hightlight field changes very well, if at all.
What problems is the product solving and how is that benefiting you?
Feature flagging and configuration management. Config management means less engineer input into certain features when other stakeholders in the business can own the config.
Comprehensive solution for feature flag management
What do you like best about the product?
LaunchDarkly offers a complete full-featured platform for managing our feature flags. The interface is straightforward, the documentation is complete, and the customer success support has been very helpful. Most importantly we haven't encountered any operational issues in production related to our use of LaunchDarkly.
What do you dislike about the product?
The broad scope of LaunchDarkly functionality also means it can be difficult to be sure we are using it to its full potential.
What problems is the product solving and how is that benefiting you?
LaunchDarkly allows us to activate technical preview features for specific customers only, and also allows us to hide features deployed through continuous delivery so that we can reveal them as part of a coordinated marketing release.
Feature flags made easy
What do you like best about the product?
I think one of the biggest advantages of using LaunchDarkly is the easy integration of libraries with application code. Simply register the LaunchDarkly client when bootstrapping the application, and make it available. All engineering levels can start working with feature flags in no time. Another great feature is the ability to target specific users and control how a flag's value behaves, which is excellent for piloting new applications in production. Even though we didn't use several integrations, seeing so many ways to connect LaunchDarkly to other apps is awesome. For us, the Slack integration was more than enough. Newer apps with no built-in feature flags would default to LaunchDarkly, increasing the frequency of use in the organization.
As far as I'm aware, we never reached out to Customer Support, so I don't have any useful comments to provide on that topic.
As far as I'm aware, we never reached out to Customer Support, so I don't have any useful comments to provide on that topic.
What do you dislike about the product?
Probably the website at the begining, but never was a dealbreaker. It does what it needs to do and after sometime I kind of like it.
What problems is the product solving and how is that benefiting you?
Streamlining the process for launching new features into production, controlling when to turn them on and off, increases engineers' confidence when deploying without affecting the running systems.
LaunchDarkly was and actually is a big game changer in our company.
What do you like best about the product?
A lot of possibilities to manage your product. Now we are using it on a daily basis. It is really easy to use, and if you explore it more you will find more and more useful things. "Easy to learn, hard to master". It was really easy to implement it in our environment. Contact with support was always a pleasure (Aaron is the nicest person I ve ever met).
What do you dislike about the product?
Implementation process was a little tricky. Not from the technical side but organisational. We didn't know that we can create one project for all platforms etc.
What problems is the product solving and how is that benefiting you?
It allows us to proceed to CI/CD. Now every team can work on its part of the product. Timelines and limits of regular upgrades do not stop them, they can do it alone. Fuckups on the production can be "fixed" quickly by switching off the flag, then the team has more time to investigate and fix the problem with no time pressure.
Powerful feature flag tool that we rely on
What do you like best about the product?
Very configurable - for instance, you can target any flag by any of multiple facts you send to the LaunchDarkly tool, and can use custom names for the contexts you use (such as "User" or "Enterprise").
Great modern JS SDK means you can do all your feature flagging on the frontend if you want, allowing neat tricks like React updating in real time when you toggle a flag on.
Thoughtfully designed and well-documented API means you can also fetch flags completely manually on the backend and handle them yourself.
The API also allows you to do admin things such as creating flags and setting targeting. In our case we had a senior engineer make tooling and scripts to use their API to synchronize it with our old in-house feature flag system, allowing us to shift to using LD seamlessly with no manual labor or risk of outdated targeting.
Product feels very well designed and is reliable and performant.
We have never had even a hint of a reliability problem, as far as I know they have had zero downtime problems.
They have a lot of features, many we don't even use such as scheduling flags, approvals, and especially, A/B testing.
Great modern JS SDK means you can do all your feature flagging on the frontend if you want, allowing neat tricks like React updating in real time when you toggle a flag on.
Thoughtfully designed and well-documented API means you can also fetch flags completely manually on the backend and handle them yourself.
The API also allows you to do admin things such as creating flags and setting targeting. In our case we had a senior engineer make tooling and scripts to use their API to synchronize it with our old in-house feature flag system, allowing us to shift to using LD seamlessly with no manual labor or risk of outdated targeting.
Product feels very well designed and is reliable and performant.
We have never had even a hint of a reliability problem, as far as I know they have had zero downtime problems.
They have a lot of features, many we don't even use such as scheduling flags, approvals, and especially, A/B testing.
What do you dislike about the product?
They find ways to inconvenience you by withholding features on the non-enterprise plans. There are parts of the UI where they tease you such as the flag compare tool, where it will show you what flags don't match across environments, but you can't have it copy the settings if you aren't on Enterprise. Overall there are too many of these, but I admit that most of them aren't vital to my usage.
The (MAU-based) pricing can get expensive if you have a lot of users/entities that you are targeting separately. Likewise, the seat-based pricing can get expensive if you need a lot of licenses.
Overall it is expensive. If software is your core competency, your needs are actually very simple (meaning you don't use a lot of their many advanced features and tools), and your usage level is high enough to hurt, you may be better off considering either a much simpler tool, or something simple that you support in-house.
The (MAU-based) pricing can get expensive if you have a lot of users/entities that you are targeting separately. Likewise, the seat-based pricing can get expensive if you need a lot of licenses.
Overall it is expensive. If software is your core competency, your needs are actually very simple (meaning you don't use a lot of their many advanced features and tools), and your usage level is high enough to hurt, you may be better off considering either a much simpler tool, or something simple that you support in-house.
What problems is the product solving and how is that benefiting you?
We use LaunchDarkly to allow us to protect all the code we push until it's passed QA and we're ready to launch a feature. We also use it to hold back a feature sometimes for a customer who isn't ready to migrate.
Successfully controlled multiple FFs in my app
What do you like best about the product?
easy to use - just create FF and swith targeting + possibility to use precise targeting with rules usage.
What do you dislike about the product?
Stale flag state after 90 days not always enough to rollout the feature.
What problems is the product solving and how is that benefiting you?
I can easily rollout new features without any fear of failure as I can killswitch any feature with just one button
showing 231 - 240