I think the article's ideas on advertising, particularly the idea that brand advertising HAS to come to podcasting because direct advertising won't scale is a b

I think the article's ideas on advertising, particularly the idea that brand advertising HAS to come to podcasting because direct advertising won't scale is a bit flakey. Why won't direct advertising scale? Direct advertising opens up the long tail of companies. Small Business X can't spend money on branding but can spend money on acquiring paying customers. In fact, direct advertising has become so popular thanks to things like AdWords, Facebook Ads, etc. that companies need to find new places to address their target audiences and podcasts might just be a great place to do that. The stuff with coupon codes, special URLs, etc. are pretty trivial to set up and a company like Midroll is going to coordinate between you and the podcast so you can line up that stuff well ahead of time. There was also this: "The not-so-secret reality about podcast ads, though, are that advertisers are quite concentrated: a FiveThirtyEight intern heroically listened to the top 100 shows on the iTunes chart and counted 186 ads; 35 percent of them were from five companies. More tellingly, nearly all of the ads were of the direct marketing variety." I did not read the 538 article on this, but it makes a lot of sense for a company to carpet bomb their advertising. So if you're listening to all the top 100 podcasts in a one-week period, you might have an advertiser hitting a lot of those podcasts in a one-week period. However, if you listened to the same 100 podcasts three weeks later, you might hear another set. Presumably, the companies finding routine success are the ones who you hear all the time. On the carpet-bombing strategy, if you deploy in relative isolation, you should be able to measure the real effect of the advertising and not just the people who came through a URL or used a coupon code. If I normally sell $100 per week and the week I advertise I sell $200, I don't have to rely on campaign tracking to assume attribution. Wait a few weeks and advertise again and see if the effect holds up.