question

Name avatar image
Name asked

Setup experiment example

I find it difficult to understand how %(% traffic allocation, % of target audience) works in experiments.

Let's say I need to implement the following experiments:

1. 50% of the audience in a certain segment is involved, 50% of this audience receive default data, and the rest of the received data override title data.

2. 75% of the audience in a certain segment is involved, 25% of this audience receive default data, and the rest of the received data override title data.

How setup these experiments?

Is this right?

1. 50% traffic allocation, 50% of target audience of controll variant, 50% of target audience of treatment variant with override title data - image1

2. 75% traffic allocation, 25% of target audience of controll variant, 75% of target audience of treatment variant with override title data - image2

And another question, only accounts that were created after the start of the experiment participate in it?

10 |1200
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

1 Answer

·
Gosen Gao avatar image
Gosen Gao answered

First, the sum of each Exclusion Group traffic allocation should not exceed 1. The Exclusion Group is designed to make sure the players in the same group will not join two Experiments at the same time.

You could set 50% players in a specific segment join the Experiment A and 50% players(the rest) join the Experiment B. In the Experiment A, 50% will get the default Title Data, and the others(50%) will get the Title Data Override A. In the Experiment B, 25% will get the default Title Data and the others(75%) will get the Title Data Override B.

In this case, for players in a certain segment, 37.5 percent of them will receive default Title Data, 25 percent of them will receive Title Data Override A and 37.5 percent of them will receive Title Data Override B.

And the settings should be:

1. 50% traffic allocation, 50% of target audience of control variant, 50% of target audience of treatment variant with override title data A

2. 50% traffic allocation, 25% of target audience of control variant, 75% of target audience of treatment variant with override title data B

If I misunderstood your request, please feel free to let me know.

10 |1200
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Write an Answer

Hint: Notify or tag a user in this post by typing @username.

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.