Describe your experience with multi-touch attribution models. When would you recommend a first-touch versus a W-shaped attribution model?
