Hi, searching for suggestion here. The case is when we want to use url parameter to control the feature flag. like

/signup?feature-a=true

is there a best practice for growthbook to support it? I know we can define a force rule and inject url parameter to setUser. but it seems not the best way to do it.

Hi! I’m trying to setup email and this error appears in logs when I try to send one:

[Error: 140435061991296:error:1408F10B:SSL routines:ssl3_get_record:wrong version number:../deps/openssl/openssl/ssl/record/ssl3_record.c:332

Does anyone have an idea how to fix it?

If I'm using the JS React Library, and a backend library like Kotlin or Python, is there a way to sync the browser plugin's variant selection tool to the backend growthbook sdk as well

Hey, I want to followup on this https://growthbookusers.slack.com/archives/C01T6Q1SVFV/p1644555632324309 we use GA4, ill explain my current understanding we setup our experiments in growthbook these experiments are evaluated on the client these events should be sent to GA in a way for example (please suggest a better way), i'm not really a fan of dimensions

ReactGA.event({
  category: 'Experiment',
  action: 'Experiment Viewed',
  dimension_1: experiment.key,
  dimension_2: result.variationId,
});

GA4 is then linked with BQ (i think we didn't setup this part correctly, because I can only see tables beginning with

ga_sessions_

) lets just assume we will fix that and we will see tables called

ga_events_20220416

is it okay to have daily tables of events, or should we have one big table of events?

hey i also have a separate question about the JS SDK any time feature is evaluated, trackingCallback is fired, and our analytics system is called this is great for some A/B tests but for example, imagine you want to experiment when a modal should show variant A: show modal after user has been in the app for x amount of time variant B: show modal after user has done something in the app in order to setup the logic in the app, we have to evaluate the feature on app start this automatically fires the trackingCallback this is not optimal, because the user might not have seen the modal, but the trackingCallback has been fired from what I understand, to counter this, we would have skip trackingCallback and call our analytics manually, when the modal has been shown to the user, right?

Hello @User and team👋🏼 - i'm assessing GrowthBook amongst other CRO/Testing platforms. Can anyone clarify if GrowthBook supports these features: 1. Experiments that cross multiple domains or subdomains 2. Personalization (down to the person/ID) beyond segment targeting 3. Existing Tableau or Amplitude integrations 4. Existing Segment integrations 5. Data portability / export 6. Multi-armed bandit options 7. Multiple objectives/goals for an experiment 8. Dedicated SLA/support options 9. Number of active users 10. Notable brands/sites using Growthbook Thanks for your help... couldn't gauge these questions from the website. 🙏🏼

hello @future-teacher-7046 @fresh-football-47124 what should i provide for API_HOST env variable inside docker.compose-yml, if i am having DNS mapped to the APP_Origin?

Hey GrowthBook team! Thanks for this amazing product 👍 Question: For self-hosting, what happens if the MongoDB container or the GrowthBook API goes down (chaos engineering)? The MongoDB container will disappear, since all containers are ephemeral. Context: We are currently testing out self-hosting of GrowthBook through K8s on GCP. The plan is to run GrowthBook and MongoDB in separate containers (without MongoDB Atlas).

I have another question regarding authentication/authorization when using GCP (BigQuery) with GrowthBook: • Is it possible use something else than GCP Service Account key (user-generated key) for authentication/authorization. Implies: Workload Identity for GCP (https://cloud.google.com/kubernetes-engine/docs/how-to/workload-identity) • If it’s not currently possible, is this something you would consider making an option in the future? This is for additional security and control over which services have access to the GCP project. Service Account keys don’t expire and it’s harder to reason about and know who has access to this key.

Hi team. I'm looking to start trialling GrowthBook, but I'm having trouble setting up my first experiment. I'm getting a "Please fill in this field" message on the New Experiment Analysis form, despite having selected my only assignment table. I feel like there's something I've left out, but not sure what.

Hey folks. I have a question about experiment phases - why are we able to control traffic split and percentage traffic indepdent from the experiment override traffic split in the features page?

Hey guys, if I used JS SDK in my web app, is there any way that I can easily override the status of feature flags in current page? This is mainly for developer testing purpose

Hey team, I think I may have found a bug introduced in the latest update. So I’m using an anonymous_id for the experiment configuration, and then I have a user_id only metric that make use of the identifier join table defined in the source. This stitching works perfectly. However, when I add an activation metric (also user_id only) it no longer joins the

__identities0

CTE to the user_id only metric I mentioned above. I’m pretty sure this was not the case before I upgraded to the new version. What do you guys think?

Hello there! I am working on getting Growthbook in my react app but for some reason the

setFeatures

and

setAttributes

aren't working, it says they aren't part of the

Growthbook

type, has anyone else run into this? I've tried re-downloading the package and restarting the code editor but no luck 🤔

Hello, I'm setting up growthbook for the first time. I have it running on ec2 right now. It all appears to be working, except I can't upload. I'm getting the error "myserver.mydomain.com:3000 says Failed to fetch" in the UI. The server logs see a bunch of 204 and 304 status codes.I tried with both using the local storage and with an s3 bucket - no difference.

Hi All. I recently started using Growthbook and I’m in the process of setting up some of my first features. Right now I’m just testing in our staging environment and setting up some force rules. When I use

id is equal to

it works as expected but when I use

id is in the list

, the users in the list are not being served the correct feature. Has anyone else experienced this and/or does anyone have tips to get this working? Thanks!

@future-teacher-7046 Another question for you. Is it possible to track if a user is in multiple experiments? We are using the trackingCallback implementation with Segment shown in the React SDK docs. But when I look at the Segment events for a user who is in multiple experiments, I only see the experiment key for the first experiment they’re in.

Hello team, I am trying out Growthbook for the first time and not able to find examples on how to do split testing. I user, action (impression vs click) in 2 different table for 2 different models. As I see we can only use one bucket I merged both the tables and added new column as varient 0/1. I am trying to sample the data between 2 variant and see clicks/impression similar to "sample conversion" example but I dont see any reference data for it. I see the sample conversion result data is entered manually (user count, conversion for control and variation). How can built these metrics dynamically? Is there any walkthrough example?

Hi again folks, I've setup a k8s deployment with gb & mongo containers and a service which with a public IP that will forward requests to ports withihn those containers. Attaching the yaml --

however I'm running into the same issue as before

Hello, I’m trying to add Growthbook on my react app for the first time. Is it possible to control the feature flags for a non-production environment? I noticed that on the GrowthBook portal, there is only a toggle for the

production

env. Is pointing my local env to a production build the only way I can test out the feature flags locally? Or could I add something to the portal to allow GrowthBook to manipulate the flags for any non-prod environment? Any insights will be appreciated. Thank you.

Hi Graham, I've added the JS for visual editor as per instructions to this page -- https://www.lbfastly.com/Project%20Cache-Me!.html But I keep getting the message in visual editor that it can't find it

any idea what may be going on?

btw the container on my laptop is able to edit the above page since I have this on the page <script async src="http://localhost:3100/js/key_79da8451ee2599e7.js"></script>

This message was deleted.

Hey there! We’re having issues using the auto-discovery feature for GCP that was recently introduced on our self-hosted GrowthBook instance. We have created a Service Account in our GCP project ‘wanted-weasel’ but want to read from BigQuery tables in a GCP project called ‘learning-honeybee’. Afair, service accounts can be used across GCP projects so I’m wondering why we’re getting this error. Why does it still ask for ‘wanted-weasel’ when I’ve configured ‘learning-honeybee’ as default project?

Hey, am I correct in understanding that the id set as attribute in the growth book application does not have (and does not need to have) any connection to the user_id used in analysis/data source. And that the minimum implementation of user attributes is just an attribute for unique id, like

new GrowthBook({
  attributes: {
    id: "iowsho98291nx"
  },
});

?

Hello, can someone guide me on how the traffic and split is applied? I have my control and variant segments in the same source table in ~50-50 split (total users=4128841, control=2064184, variant1=2064657). I have traffic set to 100 and split 50-50 however in my result the traffic allocation looks off . Also when I change the split I get sample Ratio mismatch warning. User count wrt to dimension

varient=0 and value=1
2062596
varient=0 and value=2
2061
varient=1 and value=1
2062117
varient=0 and value=2
2067

👋 We currently use Google Optimize, but keen to move to Growthbook Using something like the visual editor, we'd like to be able to set up redirect tests - so when a user arrives on a page, if they're in the test group, they'd be redirected to the alternative page How can that be achieved?