Did your campaign cause the sales?
Test incrementality on your own numbers
Sales went up after the campaign. Would they have gone up anyway? Drop in a weekly file with a change date and see the jump at the change, how big it is, and how sure we are, with the example data one click away.
Three questions this answers
- Did the number move after the launch, the campaign or the price change?
- How big was the change, and what is its likely range?
- Is the change more than the ordinary ups and downs this series always has?
What your file needs
- One row per week (or day).
- A date column and the number you care about, such as signups or sales.
- A column or date that marks before and after the change.
- Optional: a group column, such as region.
What you get back
- The jump at the change and the change in direction afterwards.
- What the series was heading for without the change, and the gap from it.
- The size of the change with its likely range.
- How much of the movement is ordinary noise.
- The method named and the R code.
Have a campaign to check?
Try the example file first, then drop in your own weekly numbers.