Hi All, I just starting with Growthbook and I'm impressed so far. The feature flags are great and really easy to use. Now I'm experimenting with Experiments, using just a vanilla postgres database where we keep a lot of our metrics. My question is: If it's not hooked up to any canned analytics engine, are there some docs on creating a basic schema. For example if I try and create a new experiment it tries to import and throws a

relation "viewed_experiment" does not exist

error. Also I'm trying to figure out where the Logged-in users assignment table is.

Hi Team , I am trying to setup growthbook feature in golang. I end up getting the default value irrespective of the split set. Any leads on how to fix this?

b, _ := json.Marshal(jsonMap["features"])
features := growthbook.ParseFeatureMap(b)
contextGB := growthbook.NewContext().
	WithFeatures(features).
	WithAttributes(growthbook.Attributes{
                "Key": ID,
            })
gb := growthbook.New(contextGB)
if gb.Feature("backend_enable_digilocker").On {
            log.Println("FEATURE IS ON")
}

Is it a known issue that the Chrome Extension doesn't work with PWA's?

If our Web Servers are running in Kubernetes behind an Ingress + Service + Deployment but we want to use GrowthBook’s Webhooks to keep our Feature cache updated what are the best practices?

Is not there a JavaScript A/B test guideline in GrowthBook using only client side vainilla JavaScript in or I'm missing something?

Hi GrowthBook, I'm looking for a feature for Multivariate tests, and I haven't been able to find any. Are there any dedicated ways to do this? (e.g. experiment types, implementation options, ...)

GA4 + Growthbook Multiple exposures Using Growthbook for a while big fan. Does anyone use growthbook javascript sdk with GA4 and not have multiple exposures warning It seems that the default way to implement growthbook with GA4 as detailed by Graham always leads to the following 2 identifiers not being the same • id used to determine variant (gbuuid from the auto,min.js) • id used to determine the successful conversion (GA4 user pseudo id) Therefore all experiments involving GA4 + growthbook will have multiple exposure warnings after a couple of days of exposure. The only potential avenues i could comprehend to shut that down but have not tried are: • Wait for GA and use GA's id somehow as the id to hash (so they match) • Somehow set the ID for that GA user based on the gbuuid cookie (so they match) Has anyone had success with either route? I think we ought to take one of the following approaches: • document one of those 2 routes if those work/ there's a prefered route • add guidance to the GA4 implementation documentation detailing why it is, that it's okay/trustable + potentially for GA4 projects that it's worth bumping the setting

Warn when this percent of experiment users are in multiple variations

from 1% to much higher as it's a non issue in this context.

A somewhat stupid question: When adding Growthbook to our site (react) we need to send the information to GA4 (in our case) about which variation a user got. As I understand it that is done in a callback function called by GrowthBook when needed (

trackingCallback

). But What I don't find (or understand) is how the GA4 event needs to be structured so that it can be understood by GrowthBook once it ends up in BigQuery? Is there an example somewhere for GA4?

If there are two arguments to separate users, uid (is an anonymous cookie) and user_id (is a logged in user), but our users can have multiple accounts and one uid can have more than one user_id, right now we just delete those users, but maybe there is some solutions to leave them in the same group?

If I use the deviceId to split users in a feature, how would this work for events that are triggered in mixpanel with the distinct id for example?