Incorporating long-term effects in determining the effectiveness of different types of online advertising

Although online advertising has become a full member of the marketing mix and is still growing in importance, studies of long-term, or lagged, advertising effects have generally either neglected online advertising channels or have treated online advertising as one homogeneous block. We analyze the short- and long-term effectiveness of different types of online advertising channels by incorporating separate time lags for each advertising channel. We look at the sales effect of email, banner, and price comparison advertising (PCA) using a sample of 2.8 million purchases and more than 1.1 million individual costumers aggregated to 365 days. Our analysis shows that email advertising has the longest effect, followed by banner advertising and PCA. We find that the length of the effect does not always go hand in hand with its intensity since, for example, banner advertising lasts longer than PCA but performs worse in terms of actual sales. This research yields important insights for theory and practice since it shows how to model long-term advertising effects and provides meaningful insights for improving the allocation of advertising budgets.