The most recent content from our members.
I am creating a template for our experiments dashboards I would like it to be as self serve as possible, making stakeholders changing the fewest things as possible. They have the ability to change the properties values but NOT the exposure key, which is a problem as they have to go in every chart and change the key, which…
We would like to request the capability to conduct multi-armed bandit experiments on your platform. Our payment team requires this method of experimentation, as it is a valuable and statistically robust method for identifying better-performing variations and optimizing traffic allocation. MAB testing is a type of A/B…
We are running an experiment, and in charts seeing a higher number of exposures than assignments. When we drill into users who have an exposure and not an assignment, they do end up having an “Assignment” event in their user event stream. What is preventing them from being included in the roll-up assignment calculation?…
Hey everyone, I wanted to share a few lessons I’ve learned running A/B tests using Amplitude Experiment and open up a conversation around how others are approaching it in real-world settings. We’ve been using Amplitude for product analytics for a while, but recently moved some of our experimentation flows over to Amplitude…
Hi everybody, I am running an experiment using the visual editor. I have made some copy changes to our subheading. However, when I land on the testing page, I can see the change from version A to version B happening in real-time. I already have a few colleagues complaining that it does not look good. Is there anything I…
The redirect from Control to Treatment does not work automatically under the following conditions: * When the user lands on the Control page via a link. * When the user lands on the Control page via a window.location.href redirect. * When the user lands on the Control page via a header('Location: <target_page>') redirect…
Hi! We currently utilize the NodeJS SDK to handle server-side analytics and experiments. We were wondering if there are any plans to build an exposure tracking provider for NodeJS, similar to the JS SDK implementation
I currently have one mutual exclusivity group setup divided 80/20 split between 2 experiments running on our site. Test currently being evaluated for 20% of traffic has shown leading indicators of a successful experience so we want to increase this to 50% of our traffic meaning that the MutEx split has to change from 80/20…
If I turn on an experiment with distribution, let’s say, 50%/50% control vs treatment, and 50% rollout, and then I set some testers in the testers tab for treatment, will the testers all be valid treatment? i.e. testers are overriding my distribution with higher priority?
I’m wondering what is the correct approach of forcing a variant and tracking an exposure event for it. Let’s say I added a feature flag with two variants - “control” (default one) and “new-onboarding”. On the app’s launch we call fetch with a small timeout in order not to block users for a long time before showing an…
It looks like you're new here. Sign in or register to get started.