This image was lost some time after publication.

To fend off Microsoft, Yahoo is conducting a test with Google to have the latter broker ads for Yahoo's search results. The notion: Google will be so much more effective at selling ads than Yahoo that it will raise Yahoo's revenues, and therefore boost its value well beyond Microsoft's $44.6 billion offer. The test is supposedly "limited," covering only 3 percent of Yahoo's search traffic. But Mike Stoppelman, a former Google engineer, thinks that the deal may be more far-reaching than Yahoo or Google would like you to think.

The journalist-math assumption is that 3 percent of search results is 3 percent of revenues. Of course not, Stoppelman points out. Google doesn't even display ads on many of its own pages, knowing it can't make money off them. His assumptions and conclusion:

  • Only 30 percent of search-results pages generate money.
  • 20 percent of those pages generate 80 percent of search-ad revenue.
  • 6 percent of Yahoo's pages, therefore, generate 80 percent of Yahoo's search revenues.
  • Google's 3 percent test could affect 40 percent of Yahoo's revenues.

If Stoppelman is correct, then Google and Yahoo have a problem. It's hard for Microsoft to argue against a 3-percent deal. But a deal which actually covers 40 percent of Yahoo's search business? That, combined with Google's existing market share in search advertising, could attract attention from antitrust cops. Google and Yahoo likely hoped to sneak this test past rivals and the government. Too bad Google doesn't have a monopoly on simple math.