I found the issue, it breaks when i have spaces in the key 😉

Hello guys! I'm trying to setup growthbook and, on first look, it works OK. But during manual tests, I've figured out that the chrome extension and experiment view event has different variations than those applied in the application. Has anyone faced such an issue? Any idea how to resolve?

Seem to be getting this when logging in with Google or manually

changing to incognito, password change, etc. did all not work!

hi, is there a JS SDK event or similar i can listen to for when an experiment override is selected in the devtools?

URGENT: I am getting

CDN usage exceeded" + cannot upgrade (UI says highest plan)

— need to ensure experiments stay unaffected We have live experiments—can you confirm no impact to config delivery/assignment and temporarily lift/reset/extend our enterprise trial through the weekend if there is any impact. Screenshots attached

Hi all, I had two questions I'm hoping someone can help with: 1. I want to set up a few of my metrics so they are not reported on until a couple of days after the user has been in the experiment. I added a metric delay of 72 hours to one metric to test it and it showed "no data" for every single experiment, even completed ones. Any advice? 2. Updating data automatically: I thought I set this to refresh the data when it's over an hour old by going to General➡️ Experiment auto-update frequency. BUT I loaded in new data yesterday and this morning there were still experiments that said "Last Updated 7 days ago." Any help on this?

Is there a way to get the

<https://api.growthbook.io/api/v1/experiments/{id}/snapshot>

endpoint to accept

dimension

and

phase

parameters? I noticed the api used by the cloud interface accepts them but I get a 400 "Request body: [] Unrecognized key(s) in object: 'phase', 'dimension'" when I try.

Hey all, I have a question, when using the Python SDK (

growthbook==1.3.1

) with

GrowthBookClient

and

client.initialize()

, I get an "Unknown feature" warning when evaluating a feature that’s defined in the dashboard but disabled in all environments. • Is this the expected behavior? • Or does it suggest the SDK isn’t updating feature definitions properly? Notably, value changes (e.g.

true

→

false

) are reflected fine, but when a feature is disabled, it seems to stop updating in the SDK’s cache. Would appreciate any clarification! Thanks!

Hello, quick question on the Activation metric that can be set for experiment analysis. In our case, we decided to have 2 dedicated events when: • a user is assigned to one or multiple experiments (as there can be multiple concurrent experiments): it's

experimentAssigned

• a user should be "activated" for the experiment meaning only from this point on they should be considered for it, and not from the start (this is because some of our experiments are assigned Server side and for other reasons). This is

experimentViewed

Now, our Experiment assignment table exposes all the fields as required and it's working fine. We've then added a very simple metric for the

experimentViewed

which will expose the

user_id

and

timestamp

. But the

experimentViewed

metric does not seem to be precise enough for the analysis to work. Take the following example: • User is assigned to Experiment A - variant 1 and Experiment B - control • This creates 2 distinct experiment assignment entries which share the same timestamp • Let's say that they navigate to a part of the UI where the

experimentViewed

is triggered for Experiment A (experiment B is triggered in another place of the UI) • If i use the

experimentViewed

metric in my analysis of Experiment B, the user with that given user_id will be included because the event was triggered, but not for the right experiment_id Is there a way to specify the right experiment_id in the Activation metric exposed fields for this to work? Or am i doing this the wrong way? Thanks!

Hi team, we are running an experiment & we are using experiment_viewed as an event for growthbook. But for some reason that specific event is not being triggered from our end properly. We do have another event named profile_created where we are storing the experiment_id & variation_id. Can we use experiment results data from profile_created event in growthbook? How can we change that? Also the experiment is showing some errors like ratio mismatch & we think that it was because of the event issue itself. Thanks!

Hello, we're running back-to-back A/A tests. Yesterday we launched the 4th test with no gap between tests and started a new phase with re-randomization. After one day, it's showing significant negative results. I hope it flattens out in 3-4 days. If it stays significant, that will be a big red flag that something is wrong with the setup. As 3rd test was significant. We found a major issue reviewing Tests 1 and 2: 98.1% of returning users got identical assignments to their previous test. That's sticky assignments, not fresh randomization. We stopped Test 1 and relaunched it with a new phase. Tests 2 and 3 had proper randomization with fresh 50/50 splits for returning users. Maybe we did the relaunch wrong? Should we create a new experiment instead of just changing the status back to running after test was stopped? As starting new phase looks like works, after 4th test we will be able to confirm this. Additional questions: • Are there known issues with Cloudflare Worker + GrowthBook randomization? • Could re-randomization cause systematic bias? • Anything else we should know about re-launching test? That is not mentioned on Growthbook website?

Hi! Our team is using GrowthBook together with Amplitude. I’m a client-side developer and I’d like to clarify one point.

export const gbInstance = new GrowthBook({
  apiHost: import.meta.env.VUE_APP_GROWTHBOOK_API_HOST,
  clientKey: import.meta.env.VUE_APP_GROWTHBOOK_CLIENT_KEY,
  enableDevMode: !isProduction(),
  plugins: [autoAttributesPlugin()],
  trackingCallback: (experiment, result) => {
    $analytics.track({
      event_type: 'Experiment Viewed',
      event_properties: {
        experimentId: experiment.key,
        variationId: result.key
      }
    });
    console.log(`key-${experiment.key} result-${result.key}`);
  }
});
const initializeGrowthBook = async () => {
  try {
    if ($analytics) {
      gbInstance.updateAttributes({
        id: $analytics.amplitude.getUserId(),
        deviceId: $analytics.amplitude.getDeviceId()
      });
    }
    await gbInstance.init({ streaming: true });
    gbFlags.initialize(gbInstance);
    return gbInstance;
  } catch (e) {
    return null;
  }
};

I’m implementing it like this, and in the plugins I specify

autoAttributesPlugin()

. That generates attributes for targeting (one of them is

id

). When setting up an experiment in the admin panel, you can assign it based on

anonymous_id

or

user_id

. How is this

id

connected with

user_id

or

anonymous_id

? Or do I need to explicitly set

user_id

and

anonymous_id

when initializing? If so, how should this be connected with Amplitude, since it has both

user_id

and

device_id

?

Hi, I’ve set up sticky bucketing using the Swift SDK and have some questions about how it actually works on the client. It seems that the sticky bucketing assignments are fetched when the GB SDK is initialized. However, I’m curious about handling logout/login and changing user attributes. Does GB go out and refetch the assignments when the user attributes change? Or.. would you recommend reinitializing the SDK when the user attributes change which would then trigger a refetch of the assignments?

Hey folks, anyone using this alongside Solid Gate?

Hi! Please clarify who can assist me with this domain migration issue?

Hi, what is the recommended way to manage experiments between different environments, e.g. development, staging and production?

Hello everyone, I’d like to ask: are stream-event requests counted under the “Million CDN Requests” item in the billing? Also, is there any way to track stream-event separately?

Is there auto fact tables for GA4? https://github.com/growthbook/growthbook/pull/3062 It didn't show up for me

Hey guys, we are sending 1 million requests per day despite only having 5k DAU and our feature flags never change. How can we reduce the number of requests to something more reasonable like 100k per day? This unexpectedly costed us $400+ bill last month

Hi GrowthBook Team, We’re currently using GrowthBook in our Flutter application and have recently migrated to using remote evaluation. As part of this, we’ve set up the growthbook-proxy service. We’ve also implemented a custom GrowthBook SDK in Flutter tailored to our internal needs. However, we’ve encountered an issue we’d appreciate your help with: • When we define a feature of type String and apply a force rule for a specific user (we pass a user ID via attributes during SDK initialization), everything works correctly — the forced value is returned as expected. • When we define a feature of type bool and force it to true for a specific user — it also works fine. • But when we force the same boolean feature to false for a specific user, the feature does not appear at all in the proxy response (we expect it to appear with a value of false). This makes it impossible for us to differentiate between a flag that was explicitly forced to false and one that simply wasn’t returned at all. Could you please help us understand if this is a known issue or expected behavior? Should boolean false values still be returned from the proxy when forced via the admin panel? Thanks in advance for your support!

Hey GB team, I am trying to understand what is the criteria behind the message: Chance to Win (not enough data) As you can see in the screenshot below, I have two metrics with a 7D conversions window. One shows significance, the other doesn't. Could someone please help me understand what could have gone wrong here? More on thread 🧵

Hi GrowthBook Team, Can you confirm whether GrowthBook can consume kafka events for metrics and dashboarding?

I can see experiment data in BigQuery, but in Growthbook I get "No data yet. Make sure your Experiment is tracking properly." I'm using the Script Tag SDK with GA4 (hardcoded). Any ideas on how to fix this? Thanks!!

This is the query we are using>

SELECT
  user_pseudo_id as anonymous_id,
  TIMESTAMP_MICROS(event_timestamp) as timestamp,
  experiment_id_param.value.string_value AS experiment_id,
  variation_id_param.value.int_value AS variation_id,
  geo.country as country,
  traffic_source.source as source,
  traffic_source.medium as medium,
  device.category as device,
  device.web_info.browser as browser,
  device.operating_system as os
FROM
  `learning-heroes-bq`.`analytics_372696867`.`events_*`,
  UNNEST(event_params) AS experiment_id_param,
  UNNEST(event_params) AS variation_id_param
WHERE
  ((_TABLE_SUFFIX BETWEEN '{{date startDateISO "yyyyMMdd"}}' AND '{{date endDateISO "yyyyMMdd"}}') OR
   (_TABLE_SUFFIX BETWEEN 'intraday_{{date startDateISO "yyyyMMdd"}}' AND 'intraday_{{date endDateISO "yyyyMMdd"}}'))
  AND event_name = 'experiment_viewed'
  AND experiment_id_param.key = 'experiment_id'
  AND variation_id_param.key = 'variation_id'

Hey i'm on the Enterprise trial plan, and I think I'm running into this bug: https://github.com/growthbook/growthbook/issues/3252?utm Anyone else experiencing that?

Hey all, I'm back with yet more integration and SRM problems. We're trying to do split URL testing but it's proving to be not as simple as "just look at the repo". Running on Next.js, I'm trying to handle all of this Growthbook stuff in middleware so we can handling the bucketing and the redirection before the client renders to avoid flashing. I'm using

@growthbook/growthbook

Node SDK alongside Segment to make a node call when the middleware loads. It fetches or creates an ID (at this point matching with Segment's anonymous ID), then sets up Growthbook, gets the features/experiments, and then performs the redirect. However I'm only getting about half of the traffic redirecting to the other URL. Code in 🧵

Hey! 👋 We're testing out Growthbook and I'm trying to wrap my head a little around the different SDK's and how they would work in our current setup. We have a Node application serving both static and React pages with Fastly CDN in front. We rely heavily on caching pages as they're created by news editors. We've previously used Fastly to split traffic between users in A/B test experiments but would like an easier to use setup. Looking into the SDK for Fastly it seems interesting with the cache KV store to prevent reaching out to Growthbook with the large amount of requests we get. However, I don't understand how the Fastly Growthbook Edge App sends feature flags to the origin? How are they passed along and how would I retrieve them when developing different variants in the origin application? The important thing is to have the Fastly Application help us do traffic splitting while maintaining a high cache hit ratio. I would think that the Edge App should send all the feature flag information to the origin and the React SDK would pick that up instead of reaching out to the growthbook server, but I'm not convinced how that would work. If somebody could fill me in on these details or knows of a knowledge article it would be much appreciated.

Hey hey, I'm new here so first of all welcome everyone! I'm just starting my Growthbook adventure and I've already stumbled upon a couple of things I'd like to clarify. I have an Inngest application running on Next.js API routes (with app router) and I'd like to leverage feature flags and experiments within that app. I've set up Growthbook integration based on the quick start guid popping up in the dashboard when you select new

SDK Connection

and my code currently looks like that:

import { growthbookAdapter } from "@flags-sdk/growthbook";
import type { Identify } from "flags";
import { flag } from "flags/next";

growthbookAdapter.setTrackingCallback(async (experiment, result) => {
  console.log("Viewed Experiment", {
    experimentId: experiment.key,
    variationId: result.key,
  });
});

type UserAttributes = {
  id: string;
};

export const identify = (() => {
  return { id: "abc" };
}) satisfies Identify<UserAttributes>;

export const testFlag = flag<boolean>({
  key: "test_feature",
  adapter: growthbookAdapter.feature<boolean>(),
  identify,
});

export const stringTestFeature = flag<string>({
  key: "string_test_feature",
  adapter: growthbookAdapter.feature<string>(),
  identify,
});

My issue is that whenever I change the value for any of my flags, the SDK returns old value until I restart my local server. Is it only because locally it's a long running server and the issue is not gonna be present when deployed to Vercel, or do I have to perform some additional steps in order to enforce refetching the data from Growthbook? Thanks upfront!

I have self hosted growthbook. Today in the afternoon, I added few support groups with 100,000 entries and attached it to an experiment. After enabling the experiment, I am unable to open the dashboard. The /auth/refresh API is giving timeout. I restarted the docker compose but still same issue...Please help