Adstock Decay | Definition
Adstock decay quantifies the rate at which advertising effects diminish over time in an adstock transformation, specifying how quickly the impact of marketing exposure fades from consumer memory and purchase consideration. This single parameter profoundly influences how marketing mix modeling (MMM) attributes value across time periods, determining whether yesterday’s campaign receives substantial credit for today’s sales vs. effects evaporating almost immediately after exposure ends.
Mathematically, the decay parameter typically ranges from 0 to 1, where values closer to 0 indicate rapid decay (effects disappear quickly) while values approaching 1 represent slow decay (effects persist for extended periods). Example: A decay parameter of 0.3 for digital display advertising means that only 30% of this week’s impression impact carries forward to next week (0.3¹), just 9% remains into week 2 (0.3²), and effects become negligible after 3 weeks. Conversely, a decay parameter of 0.8 for television brand advertising indicates that 80% of effects persist week to week, 64% into week 2 (0.8²), and 51% into week 3 (0.8³)—with cumulative influence building over months. Getting these parameters right determines whether models correctly capture long-term brand building vs. short-term performance marketing dynamics.
The estimation challenge lies in separating true decay from other temporal patterns in the data. Statistical models must distinguish between genuine carryover effects and spurious correlations that might emerge from seasonal patterns, competitive dynamics, or random variation. Advanced MMM implementations employ grid search procedures that test numerous potential decay rates for each channel, identifying values that maximize model fit while maintaining theoretical plausibility. Bayesian modeling approaches incorporate prior knowledge about expected decay rates based on channel characteristics and academic research, constraining estimates within reasonable ranges while allowing data to refine these priors as evidence accumulates.
Strategic implications extend far beyond technical model accuracy to fundamental marketing strategy decisions. Channels with high decay parameters justify sustained investment even when immediate returns appear modest, as effects accumulate over time to generate long-term value that short-term optimization would miss entirely. A brand pausing TV advertising with 0.8 decay would still receive 80% of the prior week’s impact for one week, but by week 4, only 41% remains—illustrating why sustained presence matters for high-decay channels. Low-decay channels demand more tactical, campaign-focused approaches where spend concentrates during high-intent periods rather than maintaining continuous presence. Understanding these dynamics enables marketers to balance portfolio strategies between channels optimized for immediate response and those building durable brand equity through adstock effects. Kochava MMM automatically estimates optimal decay parameters; these values understand not just what effects exist but how long they persist—critical intelligence for determining optimal campaign duration, budget pacing, and the ideal balance between sustained brand investment and tactical performance campaigns.
