But doesn't this ignore the selection effect discussed in the article? IE to use your numbers say you get a $100 sale for every $10 spent on ads. And say your profit margin (outside of ads) on that sale is 20%, so you earn $20-$10 = $10 in profit.

However, what if 3/4 of the people who clicked on your ads would have found you anyway, through organic search results or other means? Then that $10 spend only actually generated $5 in net income before ads, or a $5 loss after the cost of the ads.

It's tough to say what that number is without doing some testing, but the point is that you can't simply treat all the business coming in via ads as attributable to the ad spend.

True, you could very well cannibalize your own advertising efforts. I think that may be a problem for brands who already advertise themselves. For example, I agree it makes no sense for Ebay to place an ad for the search term "ebay" if they are already the first search result. If you have a well known brand, you will have a lot of people who will search you out directly. In that case it's a matter of running campaigns and see what sticks - if you spend 10% of your budget on an agency's ad campaign, and very few new customers turn up, you may be wasting money.

But if you have a small to medium online shop, I think it makes sense to have someone run ad campaigns for you on as many domains as possible - and pay only for high quality traffic. If you re-sell shoes from Nike, you will have a very hard time getting a click from someone who would have bought at your shop anyway, because your competitors will pay for Google PLA and other ad campaigns and anyone who is looking for a good offer will run into their ads.

Since you're taking the example of Ebay. When Ebay doesn't pay for ads, the top paid result Amazon most of the time. When Amazon also runs out of ad money, it's a less reputable site taking the top site, sometimes a plain scam.

I think the selection effect is much larger for big, well-known companies. Everyone knows eBay, so advertising is obviously going to give decreasing returns.

For smaller, lesser-known companies, I think advertising is much more effective. The chance is much lower that people will know about your product and buy it than if you spent more on advertising.

Almost all tech companies run lift studies / incrementality tests on all of the largest channels. For net new users, the incrementality is usually quite good. (for SEM-Brand and other high intent traffic, not so much... )