How to use the AI A/B variant parameters for automatic testing:

There are 3 options to work with when doing A/B variation.

variant_distribution_type = MANUAL_EQUAL
In this case, you don't need to add any thing else, it will equally distribute the variants across all the leads
variant_distribution_type = AI_EQUAL
In this case, you need to add the winning_metric_property and lead_distribution_percentage
- winning_metric_property: Is the property you want the AI system to focus on when deciding what variant is the best perfomer, you can choose from OPEN_RATE, CLICK_RATE, REPLY_RATE,POSITIVE_REPLY_RATE
- lead_distribution_percentage: Is the % of the leads you want to be used in your lead pool to help find the winning variant based on the winning metric property. Make sure it's atleast 20% for a fair calculation
variant_distribution_type = MANUAL_PERCENTAGE
In this case, you need to allocate the % distribution on a sequence variant level (seq_variants array objects), it will select the first 10 that add up to 100% in total distribution for variance allocation

NOTES:

If you pass the id it updates the sequence
If you don't pass the id, it creates that sequence step
If you pass an empty array to an existing campaign, it assumes you're deleting the sequences and as a result the sequences and the analytics associated to it will be removed*