search

facebookexperimental / Robyn

Robyn is an experimental, AI/ML-powered and open sourced Marketing Mix Modeling (MMM) package from Meta Marketing Science. Our mission is to democratise modeling knowledge, inspire the industry through innovation, reduce human bias in the modeling process & build a strong open source marketing science community.
https://facebookexperimental.github.io/Robyn/
MIT License
1.13k stars 334 forks source link

Channel by channel scenario modeling & response curve questions #1016

Open NataliaVConnolly opened 2 months ago

NataliaVConnolly commented 2 months ago

Hello,

I have two questions that I would really appreciate your help with:

  1. For the media response curves generated by robyn_allocator, are those incremental or absolute values? Meaning, a point on a curve with, say, x = $2k and y = 1, does this mean if I spend $2k in addition to the baseline number spent during the date_range I will get an additional 1 conversion, or does it mean if I spend $2k I will get 1 conversion?

  2. Is there a way to do channel-by-channel scenario modeling? Meaning, suppose I'd like to know what the response will be if I spend $X on channel Y while keeping the spend on all the other channels the same. Is there a way to do this?

    Thank you! Natalia

AdimDrewnik commented 2 months ago

Spend for selected channels can be set to be one fixed value.

jameslei23 commented 1 month ago

I guess the answer for 1st question is the latter one;
As for the second question, you can do budget allocation by only changing the upper bound of a specified channel while fixing the upper and lower bound of other channels to 1. Then you can get the uplift result.