Hi everyone, I'm using Growthbook with Clickhouse for A/B testing analysis and my queries often time out or error out due to memory usage. Are there any tips/tricks to preventing this? I've tried creating Projections (read indexes) that sort by experiment_id and timestamp but it looks like all the JOINs that Growthbook performs causes the memory issues.

Hi everyone! I ran some test experiments - on user_id (logged in users) and user_pseudo_id (unlogged users). With logged in users the results somehow started to count, but with unlogged users the error comes out -

Missing identifier join table for 'user_pseudo_id' and 'user_id'

. I’m attaching a screenshot of the experiment assignment queries, the experiment settings and my experiment_viewed table. Previous question - https://growthbookusers.slack.com/archives/C01T6Q1SVFV/p1703079333186899 Could you please tell how and for what do I need to connect these identifiers?

Is there any way to track only users who were exposed to an experiment? Instead of tracking all the users. We want to run a test on the PDP and we want to track users who landed on the PDP and saw either the control or the treatment. We don’t want to track users who visited other parts of the site and never visited the PDP.

Hi can someone help me why I get this message failed to fetch when I try to edit something in GrowthBook dashboard

Hi! I am new to GrowthBook. I tried to implement a new feature that will split my traffic into two groups of users (50%/50%). I want to send some different events to GA based on which variant users get. I noticed that on localhost, it works, but when I deployed it to a real users, they only had one variant of the page. I am using Next.js and React in my app.

I have set the env variables and JWT SECRET. I can manually call to my_ip:3100 and get a response but the call to my_ip:3000 comes back with the above. Looked on the docker container and the env variables look good i.e. point to my_ip for APP_ORIGIN and APP_HOST

Hi. I'm new to growth book and AB test experiments in general. I'm currently trying to work with the visual editor. To enable this to work does the url attribute value when instantiating the growthbook object in javascript have to match the url targeted value of the experiment or is it possible to have the former use wildcards (to target the entire site) while using more specific experiment target urls? thanks for any help

Hey! Could anyone link me to some example unit tests using jest?

Hi! I want to count the CR metric from start of an event to success (let’s say event_start and event_success). Numerator - users who clicked on event_success after (!) event_start. Here I created usual binomial metric using SQL query Denominator - users who clicked on event_start. Here I have proportion metric from Fact table I tried to add denominator to my numerator, but I don’t have this option. How can I divide Numerator by Denominator in my situation? Maybe I need to change my metrics’ setup somehow?

hi growthbook. what’s the best way to create an experiment in two environments? This is our workflow right now: 1. create feature flag and only toggle active in dev environment 2. create AB Experiment in dev environment 3. develop our code 4. when we’re ready to go live, click on the 3 dots in the dev experiment and

copy to prod

5. toggle feature active in prod This was working fine, but now the environment thats copied to (so prod in this case), no longer ever evaluates to the variant. It always returns the control. I recreated the steps but swapping dev and prod (so create in prod first and then copy to dev) and i see the same result. Using Growthbook devtools, I see the variant value. But when I log the result from

evalFeature

it, I see the control value.