Unfortunately it seems that conventional wisdom is correct. In the first week of my paid advertising experiment I have seen no increase in sales. Of course any possible benefit in increased exposure is difficult to quantify.
It's also unknown whether it has increased awareness of my book. And also there's still the question of whether advertising can help when a book is released from a more well known author. So I will repeat the experiment at a later date with new releases to see if there's any impact.
My experiment has also used smaller sites with a limited budget, so maybe a larger campaign would have better results, but whether it would be enough to justify the extra expense is debatable.
However it is the first week in a month long campaign, so I'll report in as the weeks progress to see if there has been any change. But the conclusion so far is that advertising isn't an effective way to get known as a relatively unknown author.