I should like to see published evidence of the effectiveness of Internet ads. I'm of the firm belief that Internet advertising is virtually useless. I think it

I should like to see published evidence of the effectiveness of Internet ads. I'm of the firm belief that Internet advertising is virtually useless. I think it almost exclusively benefits the ad networks and a minority of high-traffic ad-carrying websites. One paper (DOI 10.3386/w20171), relating to search ads, concluded that they saw a strong correlation between people who click on ads and people who already regularly purchase from the advertising vendor. i.e. When a vendor pays for clicks on search ads, most of the time they're paying for people who would have bought from them anyway. In that study, for campaigns where naive analysis implies there was a 4000% ROI, an analysis which takes the correlation into account indicated a negative ROI. I know from experience that another confounding factor is that marketing departments and agencies often have perverse incentives. Campaigns are often judged by the net revenue from customers who interacted with that campaign, which incentivizes campaigning towards the people already most likely to buy whatever is being advertised. Additionally, even if Internet ads do influence their viewers, the most likely influence it has on people who see it despite having installed an ad blocker is to make them dislike the host website and the advertiser.