search

facebookincubator / GeoLift

GeoLift is an end-to-end geo-experimental methodology based on Synthetic Control Methods used to measure the true incremental effect (Lift) of ad campaign.
https://facebookincubator.github.io/GeoLift/
MIT License
182 stars 55 forks source link

Other features in GeoLift MarketSelection and ROAS #90

Closed ilaria-giunti closed 2 years ago

ilaria-giunti commented 2 years ago

## Bug description Please enter a clear and concise description of what the bug is.

Session information

Please paste the output after running sessionInfo() in your R session.

Reproduction steps

Enter steps to reproduce the behavior.

Expected behavior

1- I've seen normalize, dtw, model and some other features in GeoLiftMarket Selection, how do they work?

2- Can you explain to me even why should I use CPIC to say "that's the avg ROAS I need in this experiment" if I work with sales? I ask so because the formulas are so different: The formula for ROAS is Revenue /Adv Spent The formula for CPIC is Adv SPent/ Incremental Conversions

Schermata 2022-06-30 alle 00 55 33 
Output goes here

Additional context

Add any other context about the problem here. (like proxy settings, network setup, overall goals, etc.)

ArturoEsquerra commented 2 years ago

Hi @ilaria-giunti!

  1. The normalize parameter scales the outcome model to speed-up inference which could be useful for extremely large data-sets. dtw specifies the method which we will use to measure the similarities between locations. Dynamic Time Warping (DTW) takes into account different "speeds" or "accelerations" between the KPI on test units. In any case, you can get a more detailed explanation of these parameters and more in the package documentation (you can access it, for example, with ?GeoLiftMarketSelection).
  2. If your KPI is revenue you should use 1/iROAS as the input for the cpic calculation. However, if your KPI (Y) is in units, you should use the cost per incremental conversion.