REST Resource: accounts.mediationGroups.mediationAbExperiments

Resource: MediationAbExperiment

The mediation A/B experiment.

JSON representation
{
  "name": string,
  "displayName": string,
  "experimentId": string,
  "treatmentTrafficPercentage": string,
  "treatmentMediationLines": [
    {
      object (ExperimentMediationLine)
    }
  ],
  "controlMediationLines": [
    {
      object (ExperimentMediationLine)
    }
  ],
  "state": enum (ExperimentState),
  "startTime": string,
  "endTime": string,
  "mediationGroupId": string,
  "variantLeader": enum (VariantLeader)
}
Fields
name

string

Resource name for this experiment. The format is accounts/{publisherId}/ mediationGroups/{mediationGroupId}/mediationAbExperiment/ {mediation_group_experiment_id}. For example: accounts/pub-9876543210987654/mediationGroups/0123456789/ mediationAbExperiment/12345

displayName

string

The display name for the mediation A/B experiment.

experimentId

string

Output only. Unique identifier for the mediation A/B experiment. It is an output only property.

treatmentTrafficPercentage

string (int64 format)

The percentage of the mediation A/B experiment traffic that will be send to the treatment (variant B). The remainder is sent to the control (variant A). The percentage is expressed as an integer in the inclusive range of [1,99]. See https://support.google.com/admob/answer/9572326 for details.

treatmentMediationLines[]

object (ExperimentMediationLine)

The experiment mediation lines created for the treatment. They will be used for serving when the experiment status is RUNNING.

controlMediationLines[]

object (ExperimentMediationLine)

Output only. The experiment mediation lines for control. They are inherited from the parent mediation group. It is an output only field.

state

enum (ExperimentState)

Output only. The state of the experiment. It is an output only field.

startTime

string (Timestamp format)

Output only. The time at which the experiment was started (in UTC).

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

endTime

string (Timestamp format)

Output only. The time at which the experiment was ended or target to end (in UTC).

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

mediationGroupId

string

Output only. The mediation group id this experiment belongs to. This can be used for filtering the experiments in the list experiments API.

variantLeader

enum (VariantLeader)

Output only. The variant leader for the experiment according to some key metrics.

ExperimentMediationLine

The mediation group line for the experiment. It will be used for serving during the run of the experiment.

JSON representation
{
  "mediationGroupLine": {
    object (MediationGroupLine)
  }
}
Fields
mediationGroupLine

object (MediationGroupLine)

The mediation group line used by the experiment.

ExperimentState

Status of the experiment. This is a read-only field. The experiment can be in one of the three states and transitions as mentioned below: 1. Running: When the experiment is created, it is put into the running state. Both the control and treatment mediation group lines are served according to the treatmentTrafficPercentage specified.

  1. Ended: The experiment ends after 90 days if the winning variant isn't chosen.

  2. Expired: If the winning variant is chosen or it has been 30 days since the experiment ended, the experiment goes into the EXPIRED state and the variant B is deleted.

Enums
EXPERIMENT_STATE_UNSPECIFIED Default for an unset field. Do not use.
EXPIRED The experiment has expired.
RUNNING The experiment is running and thus used in serving.
ENDED The experiment is ended, but not yet expired. Serving continues in the original configurations while the experiment is ready to call an action.

VariantLeader

Variant leader for the experiment.

Enums
VARIANT_LEADER_UNSPECIFIED Default value of the enum. Do not use.
CONTROL The control leading the experiment.
TREATMENT The treatment leading the experiment.
INSUFFICIENT_DATA Collected data (impressions) are insufficient to determine a leader.
TOO_EARLY_TO_CALL Experiment hasn't run long enough to determine a leader.
NO_VARIANT_LEADER Neither variant is a decisive winner in the observed data.

Methods

create

Create an A/B testing experiment for a specified AdMob account and a mediation group.

stop

Stop the mediation A/B experiment and choose a variant.