Stockholder Derivative Action Against Yahoo Based on Click Fraud Rebuffed–Brodsky v. Yahoo

By Eric Goldman

Brodsky v. Yahoo! Inc., 2008 WL 4531815 (N.D. Cal. Oct. 7, 2008). The Justia page. Previous blog coverage.

You may recall this stockholder derivative lawsuit against Yahoo alleging that Yahoo hyped its stock prices by overstating its ad business’ progress and by inflating revenues through artifices like relaxed anti-click fraud standards. The court has dismissed the complaint with leave to amend, principally on the basis that the plaintiffs have not been specific enough with their allegations. Most of the opinion is rather technical legalese, but I thought the discussion about the click fraud allegations were interesting enough that I’ve quoted them in their entirety:

Plaintiffs also assert that Defendants made false statements about Yahoo!’s revenues over the Class Period. Plaintiffs allege that Defendants manipulated their click fraud filters and delayed refunds fraudulently to boost revenues. As a result, Defendants’ financial statements were overstated by $387 million over the Class Period.

Plaintiffs arrive at the $387 million figure by citing three magazine articles and two press releases. CAC ¶¶ 204-208. Some of these sources estimate that click fraud accounted for ten percent of all pay-per-click revenue in the search industry while other sources estimate the fraud rate as high as thirty-five percent. Plaintiffs adopt the ten percent rate and allege that $387 million of Yahoo!’s $3.878 billion in sponsored search revenue was attributable to click fraud.

Plaintiffs point to the statements by CW 3, CW 6, CW 8, CW 9, CW 10, CW 11 and CW 12 to support the click fraud allegations. For the complaint to survive the pleadings stage, Plaintiffs must describe these CWs’ roles in Yahoo!’s revenue recognition process, or whether these CWs had any first-hand knowledge of Defendants’ accounting decisions. See In re U.S. Aggregates, Inc. Sec. Litig., 235 F.Supp.2d 1063, 1074 (N.D.Cal.2002) (accounting fraud claim not corroborated by CW statements where “none of the confidential witnesses have any first-hand knowledge of [defendant’s] accounting decisions”).

CW 3 worked as an Engineering Manager for Overture until Yahoo! acquired Overture in 2003. After the acquisition, CW 3 worked in the Business Information Systems group at the Overture facility until October, 2004. CAC ¶ 22. CW 3 claims that “Yahoo! decided in late 2004 to ‘relax’ the business rules and filters in the click-fraud detection system.” CAC ¶ 22(d). “CW 3 estimates revenues generated from the relaxation in rules represented approximately 25% of Overture’s operating revenue.” CAC ¶ 22(f). CW 3 learned of this rule relaxation from Yahoo!’s Loss Prevention manager. CAC ¶ 22(d). CW 6 was a sales representative for Yahoo! and had regular communication with customers who complained about click fraud. Id. CW 6 noted that “15% of the revenues generated in his/her group was created via click fraud and irrelevant clicks from poor content match.” CAC ¶ 25.

The Court has no basis to determine whether CW 5’s or 6’s estimates of Yahoo!’ s revenues satisfy the pleading requirement under the PSLRA. For CW 5’s or 6’s statements to carry any weight at the pleadings stage in this action, Plaintiffs must describe their roles in Yahoo!’s revenue recognition process, or whether they had any first-hand knowledge of Defendants’ accounting decisions. Also, because CW 3 was not a Yahoo! employee for most of the Class Period, the Court cannot rely on his statements to support claims of false revenue reporting for the entire Class Period.

CW 8 was a Manager of the Overture Loss Prevention organization until February, 2006. CW 8 noted that “there was an effort inside Yahoo! to relax the click-fraud detection standards.” CAC ¶ 27. CW 8 met with Defendant Decker some time after Yahoo! was sued in 2005 for click fraud, and the two discussed click fraud. Id. Through CW 8’s statement, Plaintiffs successfully allege that Defendant Decker had general knowledge of the click fraud problem, but Plaintiffs have not shown how CW 8 knows about an effort to relax Yahoo!’s click fraud detection standards, or how CW 8 knows that this effort translated into misstated revenues. Similarly, Plaintiffs have not shown whether CW 8 had any firsthand knowledge of Defendants’ accounting decisions. Therefore, CW 8’s statements do not support Plaintiffs’ allegations of revenue fraud.

*8 CW 9 worked for Yahoo! in the Customer Solutions group from December, 2003 to February, 2007. CAC ¶ 28. Defendant Decker fired CW 9 in 2007, after the Class Period, for mishandling a customer complaint that might have been related to click fraud. Id. CW 10, Engineering Director for Yahoo! until January, 2005, gave Defendant Decker access to the revenue reporting system at the Overture Pasadena facility. CAC ¶ 29. CW 10 observed that Yahoo!’s ability to filter out non-billable clicks was impacted by not having adequate resources, such as enough computer servers. Id . CW 11 worked for Overture and then Yahoo! as an advertising account manager until December, 2004. CAC ¶ 30. CW 11 described “click tsunamis” at Yahoo!, when a search brought up results that led to thousands of unwanted clicks. Id. Advertisers were charged for these clicks, but rarely realized sales from them. Id . Plaintiffs have not shown whether CW 9, CW 10 or CW 11 had a role in Yahoo!’s revenue recognition process, or whether they had any first-hand knowledge of Defendants’ accounting decisions. Therefore, their statements do not support revenue fraud allegations either.

CW 12 worked for Yahoo! as an Operations Sales Manager until October, 2006. CAC ¶ 31. At weekly customer service meetings, CW 12 learned that “Yahoo!’s revenues began to decline ‘month by month’ beginning in 4Q 05.” CAC ¶ 31(g). CW 12 attended weekly Customer Service meetings where she learned that “because Yahoo! was not meeting its traffic forecasts, the Company was not attaining its revenue forecasts associated with those clicks in 4Q 05.” Id . CW 12 also recounted that the “running joke at Yahoo! Search Marketing was that there was a ‘dial’ on the click-fraud detection system which Yahoo! turned down at the end of the quarter to allow more billable click activity to be passed on to customers.” CAC ¶ 31(l). Hearing at a meeting that revenue forecasts will not be reached is not equivalent to knowing that Yahoo! misstated its revenues. Similarly, recounting jokes about altering the click fraud dial does not satisfy PSLRA’s pleading requirements. See Limantour v. Cray, Inc., 432 F.Supp.3d 1129, 1141 (W.D.Wash.2006) (rejecting confidential witness statements based on “gossip and innuendo”). Therefore, CW 12’s statements do not meet the PSLRA’s heightened standards to prove revenue fraud either. In sum, Plaintiffs fail to plead with particularity their allegations that Yahoo! issued false financial statements.

It’s nice to see the judge recognized there’s a difference between click fraud rates in the abstract (whatever those mean) and the rate of actual overcharging experienced by advertisers, which is almost certainly lower. It’s also good to see that the judge isn’t blindly accepting the scuttlebutt from former employees, many of whom probably have worthless options or a down stock portfolio. In any case, it will be interesting to see if the plaintiffs can produce any witnesses who can testify about the rate of Yahoo’s click fraud overcharging sufficient to satisfy legal standards.