Reviews from AWS Marketplace
0 AWS reviews
-
5 star0
-
4 star0
-
3 star0
-
2 star0
-
1 star0
External reviews
External reviews are not included in the AWS star rating for the product.
LaunchDarkly: an awesome tool for running experiments
What do you like best about the product?
LaunchDarkly is great for experimentation. It has helpful features like waterfall rules which allow you to carefully define groups for testing and create allowlists for who can see a given feature. It has options to target segments, individuals, mobile vs desktop, or build a custom rule
What do you dislike about the product?
Saving your changes can be a bit unituitive, when you request a review and someone approves a LD flag change, the change does not go into effect immediately, rather you have to apply the change. It would be nice to have functionality to be able to setup a review so that an approval of changes automatically applies those changes
What problems is the product solving and how is that benefiting you?
LaunchDarkly enables us to run A/B tests on new product launches
- Leave a Comment |
- Mark review as helpful
Good enough
What do you like best about the product?
It get the job done and let's us update things live in production. A recent example was where we added the ability to change what region of a service we hit due to downtime on their end without having to do a deploy. I want to see us make better use of A/B testing and similar features in the future. I think we will find even more value there.
What do you dislike about the product?
The biggest is that it seems to have stability issues. We have a Slack channel the updates as our services encounter and recover from issues and I see LaunchDarkly more often than most other services we use.
What problems is the product solving and how is that benefiting you?
It allows us to test and release features independant of doing deploys. It also lets us update configuration without doing deploys.
It is a really good way to update production quickly, the UX is intuitive
What do you like best about the product?
With the team we use it mainly for kill switches, it is an awesome feature that allows to turn on and off experimental code that we would like to proceed with caution.
The backend client for Java is awesome and it already uses the streaming feature, so no need to poll LaunchDarkly for changes in FFs, it is already integrated.
The backend client for Java is awesome and it already uses the streaming feature, so no need to poll LaunchDarkly for changes in FFs, it is already integrated.
What do you dislike about the product?
At first it was bit overwhelming all the posibilities of contexts since I was planning on using it as a kill switch, but then once one gets familiar, it is not difficult.
What problems is the product solving and how is that benefiting you?
Mainly it helps us with the Continuous delivery approach by having feature flags turned off in production and being able to coninuously merge to master and test the features before going forward in production without slowing down the development lifecycle.
The killswitch is awesome, it allows to turn off features that otherwise we would need to ask DevOps team and a lot of bureaucracy.
The killswitch is awesome, it allows to turn off features that otherwise we would need to ask DevOps team and a lot of bureaucracy.
Simple and effective tool
What do you like best about the product?
LaunchDarkly has made controlled feature releases easy. It is our default tool to use for releases, and making rules and segments of users is intuitive. I love the feature to compare flag differences between environments.
What do you dislike about the product?
When looking at contexts in LaunchDarkly, I couldn't find any documentation on the difference between LDRelay source vs JSClient.
What problems is the product solving and how is that benefiting you?
Targeting individual users and segments of users based on specific properties. This makes feature releases targeted and easy to control and turn off if necessary.
Full control of enabling/disabling product features at your fingertips
What do you like best about the product?
The ease and simplicity of rolling out features/services to customers enhances the product rollout and provides control of the where/when/how.
What do you dislike about the product?
It used to be I could see the notes/comments on the history without searching so I could easily find the specific account should an edit need to be made. Now that's been added to a Details tearsheet which means I have to search and click through each update
What problems is the product solving and how is that benefiting you?
Helps me manage product beta releases
Works well, but the cost is very high
What do you like best about the product?
Easy ability to manage flags across enviornments
What do you dislike about the product?
The cost is too high, so my company is migrating away from it.
What problems is the product solving and how is that benefiting you?
Across environment flag configuration
Launch Darkly For Complete App rewrite
What do you like best about the product?
We have used launch darkly to completely revamp our front end app using LaunchDarkly flags. We have
What do you dislike about the product?
The ordering of segments can be a little confusing. We ran into an issue where if we write an opt in segment using a query param, it HAS to be first in the rule set to work properly. Individual targetting tends to break things up.
What problems is the product solving and how is that benefiting you?
We were able to rewrite our app page by page using LD.
Pretty straightforward, but lacking in API reporting
What do you like best about the product?
It seems relatively straightforward to and helps streamline the feature flag process.
What do you dislike about the product?
The backend API is lacking for reporting purposes. It's very difficult to find out what segments have the feature flag enabled for example.
What problems is the product solving and how is that benefiting you?
It helps deploy new features selectively to segments of users. We can also immediately turn off a feature if there is a problem.
A flexible UX for controlling new features dynamically
What do you like best about the product?
When rolling out a new feature, whether customer-facing or not, it's important to dynamically enable that feature, on varying environments, and then, incrementally on "alpha", "beta", and then the entire customer base, in small batches at a time. We use LD for these scenarios -- the features can be defined as simple booleans or integers, or as JSON hash-values deliverying varying combinations of values in one query.
This LD-based incremental approach is especially useful for transitioning "old" code to new code, say where the new code is making network calls to a new microservice, which increases both performance and complexity, especially with error management.
We have successfuly mananged several such transitions with LD controls as we address tech-debt with modern, domain-based code, including increased use of varying languages across many internal microservices.
That these LD controls can be partitioned by team ownership helps with cross-functional collaboration.
Finally, using LD feature controls also enables stakeholders from non-Engineering treams, such as Product Managers, Customer Success, Developer Success, etc. to participate in the rollout and control of these new features, as needed.
This LD-based incremental approach is especially useful for transitioning "old" code to new code, say where the new code is making network calls to a new microservice, which increases both performance and complexity, especially with error management.
We have successfuly mananged several such transitions with LD controls as we address tech-debt with modern, domain-based code, including increased use of varying languages across many internal microservices.
That these LD controls can be partitioned by team ownership helps with cross-functional collaboration.
Finally, using LD feature controls also enables stakeholders from non-Engineering treams, such as Product Managers, Customer Success, Developer Success, etc. to participate in the rollout and control of these new features, as needed.
What do you dislike about the product?
The downsides to using LD controls for new features is that these controls can sometimes become "permanent", in that they work for many stakeholders, but are not actually within our own product controls, and thus incur a permanent cost.
These controls are also network-based, which in some circumstances adds latency to the simple act of controlling a feature.
These controls are also network-based, which in some circumstances adds latency to the simple act of controlling a feature.
What problems is the product solving and how is that benefiting you?
LD helps us transition from "old" code to new code gradually, incrementally, across multiple operating environments from the same UI.
We can deploy new code, without worrying about breaking existing production through the use of LD feature flags around the new code, and then, incrementally enable and test the new features in various environments.
This includes the eventual migration of processing from the old code to the new code, incrementally across customers.
Having LD features owned by various teams is also a very useful feature that helps our teams manage our own feature flags.
We can deploy new code, without worrying about breaking existing production through the use of LD feature flags around the new code, and then, incrementally enable and test the new features in various environments.
This includes the eventual migration of processing from the old code to the new code, incrementally across customers.
Having LD features owned by various teams is also a very useful feature that helps our teams manage our own feature flags.
Easy to use, full fledged system that works across tech stacks
What do you like best about the product?
It is reliable; the client frameworks have efficient caches built in and as a developer, I dont have to worry if adding a feature flag will have a performance impact.
What do you dislike about the product?
Obviously it is a cost that could be solved with an open source solution. But when your scale is so large that managing a feature flag system is a cost in itself, it makes sense to consider the already built solution.
What problems is the product solving and how is that benefiting you?
Managining features flags, gradual rollouts, etc.
showing 161 - 170