Adstock
A transformation that models marketing's lagged effect — exposure today drives conversions over many days, not just immediately.
Last updated
Definition
Adstock applies a decay function (Geometric or Weibull) to today's spend so the model sees the residual effect of past weeks. Geometric adstock has one parameter (decay rate) and an exponential decline; Weibull has two and can model an S-shaped delay (peak effect 2-5 days after exposure, then decay). Brand campaigns typically have longer half-lives (14-30 days) than performance campaigns (1-3 days). Without adstock the model attributes all of a campaign's lift to the day spend hit, which under-credits brand work.
How it applies in India
For a short, heavy festive campaign, such as one around Diwali, adstock decides how much of the burst is still credited with sales in the weeks after the festival.
Frequently asked questions
Geometric vs Weibull adstock — which should I use?
Geometric is simpler + faster to fit. Use Weibull only when you have evidence of a delayed peak (typically TV / OOH / longer-consideration B2B). Performance channels (Meta / Google) almost always fit fine with Geometric.
Related terms
- Saturation CurveA transformation that models diminishing returns — doubling spend doesn't double conversions; the curve flattens.
- Marketing Mix Modeling (MMM)A statistical method that quantifies how each marketing channel contributes to a sales outcome over time, using historical spend + revenue + exogenous variables.
- Adstock Half-LifeThe number of days for an ad's effect to decay to 50% of its initial impact.
Try WatEase
Run your WhatsApp commerce on the platform built for India — Cloud API, GST invoices, UPI checkout, opt-in tracking, and 0% markup on WhatsApp messages (competitors add 20–25%). Pricing is on request — quoted by the WatEase team.
Get Started Today