LAUSR

Digital technologies lead consumers to engage with companies online after they see TV ads, and firms increasingly wish to coordinate TV advertising in real time with online marketing activities. As a result, firms are keen to measure how TV advertising affects consumers' online behavior, but a key question is over what window of time to measure this effect. The standard industry practice of using short attribution windows around an ad to measure a causal effect may miss the possibility that consumer behavior shifts over time due to, for example, intertemporal substitution. We collaborate with an online travel platform and evaluate the results of a field test where part of the country was exposed to TV ads while another part of the country formed a control group. Using the synthetic difference-in-differences approach, we find TV advertising leads to an instantaneous increase in online browsing and sales. However, we also document evidence for intertemporal substitution: consumers appear to move their online activities forward in time in response to TV advertising, leading to lower browsing and lower sales at times when no ad is airing. We further explore the effects of TV advertising on channel choices, device choices and promotion usage and discuss the implications for advertisers and the ad-measurement industry.