hello, what is the best way to configure a/b/c test in gb? boolean feature ofc allows only for two versions. Should I create feature type number and in code check for example if feature value equals

1

,

2

or

3

?

Hi Weronika, are you wanting to include all users in the experiment and split the traffic equally among them, like 34/33/33? If so, you would first create a new feature with the 3 variations (or 2 variations and 1 control). Then, set up the experiment in the GrowthBook UI, specifying the percentage of users you want to see each variation (like 34/33/33). Once everything is implemented and configured, you can start the experiment and GrowthBook will handle the random assignment of users to each variation.

ok, perfect. Is there any way to force to see experiment on production? I mean can I force it by using some query? This is my feature flag:

{
  "defaultValue": "map-on-listings",
  "rules": [
    {
      "condition": {
        "device": "desktop"
      },
      "coverage": 1,
      "hashAttribute": "id",
      "seed": "wpyw-mapy-na-leady",
      "hashVersion": 2,
      "variations": [
        "control",
        "map-on-listings",
        "new-keywords"
      ],
      "weights": [
        0.3334,
        0.3333,
        0.3333
      ],
      "key": "mapa-na-listingu",
      "meta": [
        {
          "key": "0",
          "name": "control"
        },
        {
          "key": "1",
          "name": "map-on-listings"
        },
        {
          "key": "2",
          "name": "new-keywords"
        }
      ],
      "phase": "0",
      "name": "Wpływ mapy na leady"
    }
  ]
}

Is there any way to force to see experiment on production?

You can used a Forced Value Rule in the production environment. Make sure it's the first rule defined for the experiment, since rules are applied in the order they appear on the screen (top to bottom).