Tuesday, July 01, 2008

Is it false to make your competitor's service worse and then say you're better?

NetQuote, Inc. v. Byrd, 2008 WL 2552871 (D. Colo.)

Previous coverage. Prior summary:

NetQuote operates a web site that allows individuals to submit information about themselves and their insurance needs. NetQuote sells that information to insurance brokers and agents, who then contact the individuals with an insurance quote.

NetQuote sued MostChoice, a competitor, and Brandon Byrd, its employee. NetQuote alleged that MostChoice employed Byrd to pretend to be individuals interested in insurance quotes. He thus submitted hundreds of false inquiries to NetQuote’s web site, knowing that NetQuote’s clients would receive bad information that could not lead to a sale. NetQuote’s clients complained about the bad information, and some ended their relationships with NetQuote. To add false advertising to injury, MostChoice advertised itself as having superior accuracy and reliability in insurance referrals compared to NetQuote.

Defendants admitted that Byrd made at least 394 fake submissions. MostChoice also counterclaimed for “click fraud” under Georgia law.

NetQuote’s Colorado fraud claim required a knowingly false representation of material fact, made to a person who didn’t know the falsity with the intention that the victim act on it, and resulting damage. Defendants moved for summary judgment on the ground that NetQuote couldn’t show reliance or proximate cause. The court disagreed.

On reliance, defendants argued that NetQuote often receives bad leads, and issued $150,000 in credits to customers for bad leads before Byrd started his attacks. Thus, NetQuote can’t rely on the content of the leads it receives. NetQuote responded that its reputation depends on selling high-quality leads, and the court agreed that summary judgment was inappropriate on this ground.

Defendants then argued that reliance was impossible because no human at NetQuote read Byrd’s submissions; they were just distributed electronically. The court continued to disagree—reliance by a computer is possible. But there was a question whether the reliance was justifiable. Colorado law was not clear on whether this was a requirement of a fraud claim. NetQuote’s director of technology development testified that the system filtered out false names like Mickey Mouse, dubious key sequences like asdfg, and fictitious area codes. Defendants argued that this was an insufficient filter because it only avoided obvious errors.

The court found a question of fact (which I think is generous—given NetQuote’s product, the burden should have been on defendants, conceded bad actors, to identify a more intensive identity verification that would still be cost-justified and that would have screened out Byrd’s fakes). Justified reliance doesn’t require a perfect filtering system. NetQuote took affirmative steps to limit false leads. Normally, the justifiability of reliance is a question of fact, and it’s so here.

Likewise, on proximate cause, NetQuote submitted evidence that a large number of bad leads played a substantial role in its loss of two major accounts, as well as some local accounts, and that it spent $128,000 in employee time to suss out and stop Byrd’s conduct. Under Colorado law, proximate cause is generally for the jury. Here, NetQuote’s evidence was sufficient to defeat summary judgment.

Here’s a tidbit: one of the major clients reported that it received 476 leads from NetQuote; 421 were called, and 95 were unworkable/fake. The 325 workable leads had resulted in no sales. Defendants argued that the 325 unproductive leads showed that the fake leads had negligible impact on the client’s rate of sales, destroying the causal connection between Byrd’s acts and NetQuote’s loss of the account. But NetQuote presented testimony that, when agents tried to use the leads, they found a very high number of wrong leads, and this started in the first week of the operation. Given that agents were compensated partly on a commission basis, they were reluctant to pursue weak leads, and so they only worked on NetQuote leads at the end of the day, further diminishing their chances of success even when the information was valid. So the poor results on true leads might also be traced to the fake leads.

Separately, general notions of restitution allowed NetQuote to seek compensation for employee time spent dealing with Byrd’s fakery. Showing the chutzpah of a parricidal orphan, defendants argued that NetQuote shouldn’t be able to recover for that time because it received a benefit—it upgraded and improved its lead filtering system, hardening it against future assaults. The court modestly held that any benefits NetQuote received could be subtracted from its damages. I suspect law-and-economics theorists would have more to say about this; my own intuition is that, if NetQuote only invested in an upgraded filtering system because of the high rate of fake leads, the fact that its system is better now does nothing to decrease its damages. If Byrd had tortiously crashed into NetQuote’s delivery van, leading NetQuote to buy a new van that, because it was newer, had more and better features than the old van, that wouldn’t mean Byrd did NetQuote a favor.

On tortious interference with contract, defendants argued that they were protected by the competitor’s privilege, and that they didn’t intend to cause any customer to terminate a contract with NetQuote. Under Colorado law, a competitor may induce a third party to end a contract terminable at will if the act doesn’t employ wrongful means and is otherwise lawful. But there was a genuine issue of material fact over whether defendants’ conduct was fraudulent, which would be wrongful means. On intent, defendant MostChoice’s chair testified that he hired Byrd to reverse engineer NetQuote’s customer list by submitting bogus leads, apparently to get the names of agents who then followed up on the leads, and he did in fact have his employees contact NetQuote’s customers. This was, unsurprisingly, enough to avoid summary judgment.

NetQuote’s Lanham Act claim concerned MostChoice’s website ad copy claiming that its leads were “Better Than NetQuote Leads.” The court earlier rejected a puffery defense. Defendants moved for summary judgment on the ground that NetQuote had no evidence of consumer confusion. But the court adopted the rule from other circuits that, if MostChoice’s conduct was intentionally deceptive, NetQuote would be relieved of the burden of showing direct evidence of consumer confusion. MostChoice could, however, rebut any presumption of confusion caused by intentional deception.

Here, NetQuote argued that intentional deception could be inferred from MostChoice’s efforts to submit over 3,500 false leads to NetQuote. Byrd was hired to submit false leads for 20-30 hours a week over 9 months. The court agreed that a reasonable juror could find that MostChoice intended to harm NetQuote’s lead quality, making its acts intentionally deceptive.

There’s an understandable conceptual step skipped here: Were MostChoice’s leads better quality than NetQuote’s leads, either in their Byrd-degraded state or in their “natural” state? If MostChoice’s leads were better, for whatever reason, then the claim is not false, which may be why NetQuote wasn’t arguing literal falsity. But failure to disclose MostChoice’s interference with NetQuote’s leads isn’t the usual kind of nondisclosure that creates misleadingness: No matter the cause of the quality of NetQuote’s leads, it’s still the case that they have that quality, and clients could be disappointed by it. It may well be true that disclosing MostChoice’s interference would leave most clients unwilling to contract with an entity that invests in degrading its competitor’s product. But then we have a nondisclosure claim—and a difficult one to prove, since nondisclosure is rarely actionable.

The real issue is that MostChoice took action to make its claim true, but if true, it was true only because MostChoice deliberately harmed the quality of NetQuote’s service. Ordinarily, a competitor may take action to make its claims true—by, for example, designing its product specifically to achieve better results on some metric than the competing version. We would never say that the resulting claim “X delivers superior battery life to Y” was false just because X didn’t disclose that the superiority was the result of design choices, even though those design choices may well mean that Y performs better on some other, unadvertised metric. On the other hand, this is the unusual case where the challenged acts degraded the competitor’s service rather than improving the advertiser’s, and I can hardly fault the court for wanting to grant a remedy. I do think tortious interference might be a better fit, though.

NetQuote did have a separate argument for literal falsity, based on a different claim that MostChoice advertised that it obtained all its leads “from people who visit the site,” but had purchased leads from LeadCo, a third-party aggregator. MostChoice’s CEO testified that MostChoice had only done so for a short period, and NetQuote had no evidence that MostChoice purchased leads during the time it made the challenged claim on its site, so its claim based on that statement didn’t survive.

MostChoice counterclaimed under Georgia law for “click fraud,” based on NetQuote employees’ acts in clicking on MostChoice ads for which MostChoice paid per click, without any intention of using MostChoice’s service. NetQuote employees allegedly did this 25 times from October 2004 through August 2005, and 27 times from January 2007 through September 2007.

The court held that there was no reason to think Georgia recognized a tort of “click fraud.” Even if Georgia did so, there was no evidence NetQuote engaged in misrepresentation. Its employees occasionally clicked on MostChoice’s ads to see what services it was offering as part of their practice of checking out what the competition was doing on Google, Yahoo! and other search engines. MostChoice’s CEO agreed that it was not fraudulent to click through from a search engine to look for information on the site without ultimately filling out an application. There was no evidence that NetQuote’s clicks “were prompted by some illicit purpose,” so NetQuote secured summary judgment in its favor on this claim.

No comments: