hi! I see this issue when i'm trying to upload screen. Just design issue)) May be there is another place, where I can share this little bugs? or slack is good?

Hi, I remember I used to find the button to start a new phase of the experiment, but I forgot the exact step to trigger the new phase.. any hints?

One question related with feature conditions, does it support to use nested json field or even array? Like we have the user attribute as

{ address: {state: 'CA'}}

And we can add

address.state

in the JSON rule

I also want to share, that hint isn't showing at guardrail metric

Hello everyone!! Does anyone already designed an experiment with TV Advertising as subject of analysis? It would be hard to isolate control from the TV Advertising effect so I wonder how we can minimize the bias. Any tips?

Hi, can we change the feature flag value type after creation? we try to change from number type to string type. but not find the way to change it.

Hi, I have seen the hint about download as a jupyter notebook, but I can’t see it in the 3 dot menu. Is it a bug?

Another issue for us is we use different experiment id compared with the feature flag id. So I have to change the tracking key in the experiment to use the experiment data. And then the feature flag can’t find the matching experiment id as it’s different. Any solution for this? like manually associate feature flag to experiment.

Hi, any tips on how would I go about configuring a holdback group of users that would be excluded from all experiments for a given time period? The idea is that at a later stage this holdback group will be divided to a treatment group receiving all changes at once and a control group -> we could measure first each experiment on it's own and then later measure the combined effect of multiple changes

Hi, a general question. what’s this

suspicious result

means? Is there a explanation doc for these kind of labels?

Hi everyone! 🙂 I have a question: if we create an experiment linked to our own feature flag tool, and we change the split of an AB test there, would it be automatically reflected in GB results? thanks a lot!

We just released version 1.2.0 🔥 - which includes a lot of requested upgrades: • Customizable environments • Mutually exclusive experiments adjustable through the UI • Per-environment rules • Tag manager • Custom tag colors • Tag quick sorts • Built-in SQL schemas for GA4, Segment, Snowplow, RudderStack, and Amplitude. If you're on our cloud version, you already have this latest version. If you're self hosting, be sure to update: https://docs.growthbook.io/self-host#updating-to-latest Please let us know if you have any issues or feedback.

I’m new with the concept

Mutually exclusive experiments

, can anyone give some real use examples to explain when and how we should use such feature?

guys, thanks for awesome update! I see some issue at the experiment panel, metric's names moved out

Hi there 👋 we are trying to use "in the list" targeting condition on the cloud version of GrowthBook Forced Value, but it doesn't seem to be working, targeting works however if using the "equals"

hello guys! Is it possible to to somehow share to mail some experiments statistics like: running experiments, ended experiments (and what result), how many experiments starts and etc.?

I see that when you try to add new experiment, there are no color difference between groups of metrics (easy fix I guess)

<!channel> what is the tracking function for amplitude. I see the example only sites segment.io like this but i want rto do it for amplitude

analytics.track("Experiment Viewed", {
      experimentId: experiment.key,
      variationId: result.variationId,
    });

Hey all, is my understanding correct that we need to handle the metrics tracking ourselves? And we then connect Growthbook to our DB for analysis of the experiments?

@future-teacher-7046 Do I need to do any specific configuration in my react app for growthbook to track the user that views an experiment and later does the action that translates into data in my db? For example, if i want to know if the random user who viewed the pay button went ahead to make a payment in the db, is there any specific configuration I need to do using the sdk in order to track this user's payment in growth books

I've set up my datasource (bigquery), and managing to get to grips with how Experiments and Metrics are tracked. However, for experiments, it looks like the identities table is required (with user_id and anonymous_id). We can't get any result experiments without that table existing. Is this expected behaviour?

Hi guys. I'm trying an A/B experiment using Google Analytics 4. Trying connect with BigQuery, cause is this way I understand the solution, but I can't understand what json file needs here

can I easily move experiments between my projects? or I should add experiments again to new project?

Does the JS SDK have a listener/hook for when devtool has changed? Trying to refresh our angular app

I have a test experiment that I'm running with seed data. However, I'm getting an error of

A Sample Ratio Mismatch (SRM) was detected with p-value of 0.

I inserted 2000 experiment_viewed records, split 50:50 across the 2 variations. From the 1000 users that saw variation 0, 300 went onto convert. Of the 1000 users that saw variation 1, 700 went onto convert. 1000 users did not convert at all. I'm trying to understand why this is giving a SRM.

Hi, I got an error when running the experiment. Not sure why this happen. My guess is I have some variation to be

0

. but it should be take care accordingly. Also found a number precise issue. (the split 55 but shown as 56)

hey all - QQ - we just recently broke our front-end off into a Nextjs app. We're about to run our first experiment since, and I'm just trying to work out all the pieces in my head. Presumably what we have to do now if we want to carry an experiment between the FE/BE is drop some sort of cookie with a reference to the version they should see, and then put them in that version in the backend. Is that right? Follow up Q - I know we can temporarily force a user to see a version, which we've been using for debug purposes - is that how we'd implement the forcing into a version requirement?

Quick question: SRM with p-val = 0 means there's no SRM, right? So this message shouldn't be displayed?

Hi there! I’m working on an implementation of GrowthBook where a logged-out user sees a Pricing page, and after they login or sign up, we want to show the same version of the pricing page they saw before - to accomplish this we set a cookie that forces the correct AB Test version between those two states. That all works fine. My other requirement is that I track revenue of the new pricing and old pricing versions via Mixpanel at invoice pay time, in which only logged-in users are tracked via the trackingCallback, which is their logged in ID - which also seems to work fine. However, (and correct me if I’m wrong) each of these pieces work by setting the

id

attribute, which in my context would be overwritten as soon as someone logs in and is associated with their logged in

id

. So my question: how do I get both of these pieces to work at the same time?

I've been seeing this a lot so I was curious if there are any best practices to avoid this.