Effectiveness in Advertising - A Question

This post is based on a general topic I have been thinking quite a lot about recently - the social construction of reality. Given I'm studying to be a social psychologist, that is probably not too surprising. Still, it is strange stuff. Follow me, dear reader, down this dark rabbit hole...

Advertising is ubiquituis. Every day, we see hundreds, if not thousands of advertisements, from places we expect them, like tv and radio, and from less expected places like toilet stalls and roof tops beneath flight paths. But this kind of saturation comes at a price - $144 billion a year is spent on advertising in the US and $385 billion a year is spent worldwide. For the superbowl, advertising costs as much as $5+ million per minute. If I were spending all that money, I'd like to know how effective it is. So it is not very surprising that there have been hundreds of research studies that have sought to discover how effective advertising really is.

I've been reading a meta-analysis (a study of many studies) of 389 real-world studies on the effectiveness of TV advertising. Most of these 389 studies are generally based on advertising a product on tv and then looking to see how much sales change. I found the results rather intriguing. Anyone want to take a guess at how much the typical increase was?

I'll give the study results with explanation on the weekend. Until then, please ponder away. Accurate answers will be rewarded with a pony, possibly a unicorn if the answer is presented in iambic pentameter.

The original paper is:

Lodish, L.M., M. Abraham, S. Kalmenson, J. Livelsberger, B. Lubetkin, B. Richardson, and M.E. Stevens (1995), “How T.V. Advertising Works: A Meta-Analysis of 389 Real World Split Cable T.V. Advertising Experiments,” Journal of Marketing Research, 32 (May), 125–39.