hi there, when the

Date

dimension is selected as in the attached screenshot, do the daily results include all behavior (regardless of the day it took place) for the users that viewed the experiment on that day?

Hi, we just tried using the presentations feature and the output looks nice but seems not customizable at all. For example, we can’t seem to include a dimensional analysis or segmented results for an experiment. Is that possible? (cc @wide-jordan-14758). Related question (our goal is to pull experiment results into a larger external report). How (if at all) would you recommending downloading custom reports we’ve created for an experiment? The jupyter notebook export feature is helpful, but for this we’re hoping to export the already-visualized data.

Hey there, I had a question around how Growthbook knows that a user has been assigned during a % based feature flag, as in how does Growthbook store the state so as to cap at 10% of users? Love using Growthbook so far, and I am self hosting if that’s useful info. Thanks!

Hi there, I am still wondering about the purpose of the new SDK connections feature / page. It's just creating a regular API features endpoint like on the old page, right? Are there plans to (automatically) remove existing (old) endpoints in the future releases? CC @adventurous-dream-15065 @mysterious-iron-16289

Hi, a few questions on tracking experiments: 1. the tracking data needs to be sent to Analytics platform like Segment or Amplitude, is that the only way? would it be possible to dump the data to S3 and pipe to snowflake ? 2. from the doc, it doesn't seem it's possible to track Feature Flag, does that mean if we want to track that, we need to convert the feature flag into an experiment ?

Another question regarding feature flag. Inside A/B Experiment, is there a way to do targeted assignment? Basically using Growthbook's Targeted inside A/B Experiment

Hey @future-teacher-7046 We are currently doing an a/b test that the numbers look pretty weird, the total number of signups showing in the experiment does not reflect what we can see in mixpanel. it is basically a fraction of what it should be. In the image as you can see in the last days (experiment started on january 16) We should have at around 1500 signups but in the experiment it is a bit over 600 only. Could you please help us understand why the numbers do not match?

Hi team, We are currently self-hosting Growthbook and it has been very helpful for our team and company. We have a use case where we need to tag the owner/creator of long-running feature flags on slack or an issue on our project tracker. I was looking at the

/features

API and found that the

owner

information is available but it is a free text field and

revision.publishedBy

contains the email address of the person who did the last change. We are unable to co-related the free text owner field with the email address as multiple people can have the same name. Is there a way to get the email address of the creator of the feature flag?

Hi Team, Currently we are doing a POC for GrowthBook We are able to deploy it in our local env. When we try accessing it via localhost, it works. But when we try using IP instead of localhost, it doesn't. Can someone guide? Thanks Swapnil Sanu

Hi there, I am interested in audit logs of user activity Growthbook UI (the way how A/B experiments are configured, what exactly users are doing and what they change). Is there any way to get that data? In Growthbook docs I little information about audit trail for feature flags values changes. But that is it. So, please point me onto proper documentation if there are audit logs capabilities. Thanks in advance.

Hi everyone I have a test running but with errors in the report query (Failed to update). Does anyone have any idea what it could be?

Hi everyone. Have a question that is not directly related to Growthbook but rather to A/B testing in general. Precisely, sample size estimation (now that you have Frequentiest framework in your product, it’s even more relevant). So, when estimating the required sample size for the experiment, one usually needs the base conversion rate so they can estimate the required sample size. The problem we have is that our conversion rate fluctuates wildly, so even if the historical data shows the base conversion rate of 9%, it can significantly drop or get higher during the experiment due to some external factors (like, we started getting less quality traffic from our PPC ads). Are there any ways to take this into account so that we still have the statistical power of at least 80% even if the “real” conversion rate drops during the experiment for both variants? My first guess would be to “underestimate” base conversion rate, but i’m not sure I’m on the right track here. Say, we looked at the data for the last 3 months and discovered that our conversion rate was 9%, would it be statistically and mathematically valid to specify the base conversion rate to be 8% instead of 9%, so that we “safeguard” ourselves against conversion drop during the experiment? Of course, it would require larger sample size, but that’s a sacrifice we are willing to take.

Hi @future-teacher-7046 @fresh-football-47124 If I want to get GB feature based on the project id, like features only for specific project then how I can achieve this my use latest version of js sdk?

@fresh-football-47124 @future-teacher-7046 we (idealo) discovered two bugs in our self-hosted instance: https://github.com/growthbook/growthbook/issues/971 and https://github.com/growthbook/growthbook/issues/926 Please reach out to me if you have questions about. CC @mysterious-iron-16289 @adventurous-dream-15065

Experiment tracking in Segment

We are currently tracking the experiment via Segment event but want to seek advices on what to pass down to Segment. Right now we only pass

exerpimentId

and

variationId

. However, we are likely to expand beyond A/B testing on those who have

user_id

. It might be

device_id

for example. Would it be better to pass all the targeting attributes inside Segment as event properties?

Just joined and am trying to gain an understanding of the self-hosted option so that I can advise my boss on whether this is a good solution for us (only been on the job for two-and-a-half weeks, so don't want to screw up). In particular I am trying to get my mind around feature flags and what is the environment to get them to work. Running it locally seems pretty straightforward, as everything runs in a docker container and presumably I just need to have one of these in my local docker environment. Things get a little murkier when I think about how this would work in our qa environment and then again in our production environment. Ours is a web app running in the cloud. So would we then need to have a docker container also running in our qa cloud environment and our production environment? And would the feature-flag-enabled application interact with these at run-time? Or is it somehow possible to put all feature-flag information in a configuration file of some kind that the application could consult to know how to manage the features? Please excuse my ignorance, but I really just started looking at this this morning and am trying to get an overview as quickly as possible. Thanks much.

Hi By any chance is it possible to integrate growthbook with moengage ?

I already asked this in the webapp support chat but figured slack would be easier: Is it possible to create a metric that is only counted once if a user performed a specific event x times? If yes, how can I do that?

Hey, I am trying to create an experiment with google analytics v4 datasource. I setup the query to get the data from bigquery in the metrics and it is working fine. But inside the experiment, in the query generated it is showing

event_*

instead of the source dataset path like in the screenshot. This is causing an error saying

Table "events_*" must be qualified with a dataset (e.g. dataset.table).

Any idea whats causing this ?

Hey! We’re using Growthbook in our React project for feature toggling. When we have enabled a feature it first logs

true

to indicate that it’s enabled, but when navigating to another page it suddenly logs

false

. Is this because of a cache issue? Is there a way to get around it?

Hey team! We love GrowthBook and are ready to start our very first A/B test (🎉), but we are struggling a bit with Google Analytics as a data source. We would love to: • identify a visitor's country based on the GA dimension "country" • split the traffic 50/50 for only certain countries as part of an experiment Could you help out with how we can configure this in the UI? (we are not on GA4 yet)

Hi! Could you please explain, whether it is possible to run AB test only on certain segments, based on user events? • Those who have seen some banner/or just topped up their account • Or those who haven’t participated in other AB tests I’ve seen “targeting conditions”, but this is based on user property which you should add manually for each case. I’ve also seen ‘segments’ feature, but couldn’t find a description of how it works. And looks like it is applicable only for post-analysis. Also, is it possible, for example, upload a large sample of users from database (or just from .csv), and run experiment only on these users? That is create manual segments and run experiments only on them.

Hi team! One quick question about namespace feature. Is namespace isolated for each environment? Or they are being considered as a whole regardless of environment.

Hello, Guys! In Growthbook have a way to run two or more A/B test at the same time and at the same place in platform without interfering each other? We can configure for a public (of a test) not be able to be sort in other test?

Hi everyone, hope you can help clarify how the experiment platform works. 1. I configured an experiment from the feature menu setting the experiment to 100% of traffic with a 50/50 split. Then from the Experiment menu I created a new phase of 100% of traffic with 80/20 split. If I go back to the feature I still see the 50/50 split. What phase of my experiment is currently running, the 50/50 or the 80/20? 2. When I mark an experiment as won, what happens to the split. Does the split more to 0/100 or do I need to do that manually? What about lost, or inconclusive?

Hi all, looking for some advice on how to implement a metric. I’ve got a custom denominator on my metric, and the timestamp of the events in the denominator occur after the numerator timestamps. What I’m seeing is that the count in the numerator is lower because the timestamp comparisons that Growthbook adds as part of the query it generates look for numerator timestamp >= denominator timestamp (+/- conversion delay) and numerator timestamp <= denominator timestamp + conversion window. I’ve set a large conversion window for both numerator and denominator. I’ve also explored adding a negative value for conversion delay, which does include more of the numerator events, but unless I set it to align with the experiment start time it’s including events from before the experiment too, which is not what I want. Do you have any guidance on how I should implement the metric such that I have a count of all events for the numerator and denominator after experiment start for this setup? Or am I doing something incorrectly which is making this trickier?

Is it possible to create a metric for mixpanel that will rely on two events? The metric would trigger if any of the events happened for example

Heyo - we at idealo successfully built an integration in next.js with pure SSR 👍 and are wondering how we can now do the tracking callback to track the "experiment viewed" event? According to https://docs.growthbook.io/lib/react#pure-ssr we don't really have access to the GrowthBook instance in the browser, and there doesn't seem to be a tracking option on the server instance. How would we solve this? cc: @adventurous-dream-15065

Hi ! Running into a weird problem that we created an experiment based on a feature, but it didn't show up in the UI as running test. So it's invisible. and we can't delete it or stop running it. But when we try to recreate the experiment with the feature, it says errror due to "duplicated experiment id " which we can't see hence delete..

Hi guys, I found your project on the GitHub and noticed that you provide ClickHouse support Would you be interested if I created a pull request that migrates that to the official ClickHouse Node.js client?