In the United States Court of Appeals For the Seventh Circuit
October 28, 2010
THE SMART MARKETING GROUP INC., PLAINTIFF-APPELLEE/CROSS-APPELLANT,
PUBLICATIONS INTERNATIONAL LTD., DEFENDANT-APPELLANT/CROSS-APPELLEE.
Appeals from the United States District Court for the Northern District of Illinois, Eastern Division. No. 04 C 146-Joan B. Gottschall, Judge.
The opinion of the court was delivered by: Wood, Circuit Judge
ARGUED APRIL 7, 2010
Before WOOD, EVANS, and SYKES, Circuit Judges.
Online commerce has ballooned in importance over recent years, and it is no surprise that automobile dealers are among those who are interested in exploiting its possibilities. This case, brought under the diversity jurisdiction, involves an effort by two companies to develop programs that would deliver location-specific, brand specific, internet sales leads to auto dealers. Publications International Ltd. ("Publications") ran a website that collected the raw sales leads, and Smart Marketing Group ("Smart") promised to develop programs that would market that information to auto dealers. Unfortunately, things did not go as planned. Publications terminated the agreement, and Smart sued for breach of contract. A jury saw things Smart's way and awarded it $5.6 million in damages. Given the deference we owe to the jury, we refrain from disturbing its verdict on liability. Its damages award, however, finds such slim support in the evidence that we conclude that there must be a new trial limited to damages.
In an effort to get a good buy on a new car, millions of Americans now turn to the web in search of free price quotes from local dealers. Publications for some time has run a popular website, called ConsumerGuide.com, that furnishes these quotes. Its website is based on a well-established automotive guide. Provision of these quotes has been profitable for Publications, because the quotes generate sales leads that can be sold to wholesalers, who in turn sell them to auto dealers. In 2003, Publications decided that it could earn more from its leads if it cut out the middlemen and sold directly to the dealers. In order to carry out this plan, it hired Smart to market its leads. Smart's two principals, Michael Welch and William Magarity, had extensive experience selling conventional leads and other promotions to car dealers, but neither had much familiarity with internet leads.
Initially, Publications and Smart developed two separate programs that Publications planned to sell to auto dealers nationwide. The first was called the Approved program; it involved selling dealers the right to be designated as a Consumer Guide Approved Dealership. This designation was supposed to entitle the dealer to use the Consumer Guide logo, to advertise in Consumer Guide's print publications and on its website, to display a plaque in the showroom, and to obtain 40 vehicle-history reports per month. The other program was called Leads & Listings; in principle, it was supposed to exploit the internet more effectively. The parties planned that each member dealership would receive an average of 16 location-specific, brand-specific internet leads per month from ConsumerGuide.com. A standard Leads & Listings contract was supposed to last for 12 months, although in practice it turned out that some dealers preferred shorter terms, and it appears that their wishes were respected.
The parties executed two temporary agreements in March and July 2003, under which Smart began actively selling both programs to dealers. To hear Publications tell the story, the Approved program was a dud. Smart began selling the program to dealers on a commission basis after the March agreement was executed, but the dealers did not like it. Publications says that it took too long for a dealership ad to appear in the bi-monthly Consumer Guide magazine, the dealers received paper certificates rather than wooden plaques, and worst, the dealers were not getting any leads. Many dealers demanded refunds, which they received, and Smart's sales of Approved contracts fell off sharply, down to five in July, eight in August, two in September, and none in October and November.
Still according to Publications, this put pressure on the parties to launch the Leads & Listings program quickly. In the July 2003 agreement, the parties decided to begin marketing the Leads & Listings program, which was set to become operational sometime in the fall. Smart was also to be paid on a commission basis under this program. Between late July and November 2003, Smart sold 428 Leads & Listings contracts; the contracts had varying terms and durations ranging from monthly to annual.
Smart's account of this early stage of the business reflects a different tone. The March 2003 agreement engaged Smart to sell the Approved program plus two others that were never launched. Publications was entitled to cancel the agreement for cause if Smart failed to enroll 240 dealers by March 31, 2004. Smart asserts that it had great success selling the Approved program. It had nearly 50 dealers enrolled by the end of April, and it added another 30 in May. By the end of June it had signed up 113 dealers and was on track to meet the contractual goal. Smart concedes that the dealers were grumbling about the program; in Smart's view, however, this was because Publications was not delivering what it had promised: dealers enrolled between March and May did not receive their "welcome" kit from Publications until early June; instead of plaques they got paper certificates; and their ads did not appear in the Consumer Guide magazine until October.
Smart recounts that the parties had delayed the Leads & Listings program while Publications worked on software designed to distribute the leads. By July, however, they decided to put the Approved program on hold and to focus their efforts on launching Leads & Listings. Smart was to contact Approved dealers first and offer them a discount if they agreed to participate in the Leads & Listings program. As of July 25, 2003, Smart was authorized to begin selling Leads & Listings contracts to dealers. A performance clause in the agreement required it to enroll an average of 30 dealers per month, for minimum monthly fees of $295 to $495 depending on the dealer's size. Smart enrolled over 100 dealers in the program by August 30. It continued to enroll dealers, while at the same time the parties attempted to negotiate a more permanent contract.
By the end of September, Publications had still not finalized the necessary software that was to deliver the promised leads. It asked Smart to limit new contracts to 125 per month; Smart complied with that request. It signed 114 new dealers in October and another 51 by mid-November. All was not well, however. Smart was having trouble getting paid for its efforts, and the parties could not agree on the compensation due for contracts that were converted from the Approved program to Leads & Listings. On October 21, Publications admitted that it owed Smart roughly $120,000. On October 24 the parties signed a two-year contract in which Publications agreed to have Smart sell both programs to dealers. That contract declared all prior agreements "null and void and superseded and replaced in full" by the October agreement. The October agreement included a commission schedule and provisions on termination. It allowed Publications to terminate the arrangement for cause if Smart misrepresented the Consumer Guide or its programs, if Smart engaged in business practices "that Consumer Guide in its sole judgment may negatively impact the Consumer Guide brand [sic]," or the attrition rate in any program exceeded 10% per month prior to the third month's bill.
Very soon after that agreement was signed, Publications decided to pull the plug on the entire relationship with Smart. Publications had not figured out how to deliver the promised leads, and the gap between its promises and its ability to fulfill them had become too wide. Its president, Richard Maddrell, had his lawyer write a letter to Smart on November 18 telling Smart that Publications was terminating the October agreement. The letter relied on each of the three reasons noted in the October agreement: it said that Smart had misrepresented Consumer Guide's programs; that Smart had engaged in practices that had a negative effect on the Guide; and that the attrition rate was in excess of 10% per month. The letter offered no explanation for these conclusions. Publications finally succeeded in bringing its software up to snuff in mid-December 2003. It contracted in mid-2004 with a different sales force to sell a similar leads-distribution program to auto dealers. Publications continued servicing auto dealers with leads until late 2005, when it sold the Consumer Guide automotive business.
Displeased with its termination, Smart filed this suit in the district court on January 9, 2004. Smart is incorporated in California and has its principal place of business in that state; third-party defendants Michael Welch, William Magarity, and Paul West, who are all affiliated with Smart, are also citizens of California. Publications is part of a chain of companies that begins with Consumer Guide LLC, a limited liability company. The sole member of Consumer Guide is PIL New Medial LLC; the latter's sole member is PIL E-Commerce LLC; and E-Commerce's sole member is Publications International Ltd., which is a corporation incorporated and with its principal place of business in Illinois. The amount in controversy exceeds $75,000, and so the district court had jurisdiction over the case under 28 U.S.C. § 1332.
Smart's amended complaint accused Publications (and Consumer Guide, which we do not need to address separately) of breaching its contract with Smart in three different ways; it also included one count of fraud and duress, and two quasi-contract counts. Publications filed counterclaims and cross-claims, but those have dropped out of the case by this time. The district court ruled on September 11, 2008, that the October agreement was valid and enforceable, and that this rendered the earlier agreements of no moment. It therefore dismissed Smart's claims alleging breach of the March and July agreements; it retained Smart's claims alleging a right to commissions on a quasi-contract basis for work done before the October agreement. The court rejected Publications's argument that it was entitled to cancel the October agreement, ruling instead that the contractual right to terminate had to be exercised in good faith, and whether Publications had done so was a jury issue.
On that basis, the case proceeded to trial. The central question, given the earlier rulings, had to do with Smart's damages. Smart was asking for lost profits of approximately $8.8 million. It relied on an accounting expert, Martin Birnbaum, to support that number. Birnbaum reviewed each of the 555 dealer contracts that Smart had sold and summarized their terms in a spreadsheet. Using that information, Birnbaum then projected the profits that Smart would have earned over the remainder of the two-year term of the October agreement. He assumed that Smart would have enrolled 37 Approved and 110 Leads & Listings dealers each month, less attrition of 2% and 1% respectively. He also assumed that there would be a 25% annual increase in the number of leads available.
Publications contested Birnbaum's qualifications as an expert. The court agreed that Birnbaum had no firsthand knowledge of the automotive industry, and so he knew nothing special about attrition and renewal rates for contracts like the ones Smart was selling. Birnbaum conceded that he had relied for that information on Smart's founders, Welch and Magarity, and that he did not know whether their estimates were typical in the industry. The court ruled that he was not qualified to render an expert opinion on the final damages figure, but it permitted him to explain his spreadsheet so that the jury would under-stand how Smart arrived at its lost profits figure.
Publications called two experts at trial. The first, Dillon MacDonald, was presented as an expert in the business of marketing to car dealers. His past experience was thin, but the court allowed him to testify that Birnbaum's assumptions were unreasonable and that it was not possible for the Consumer Guide website to produce enough leads to provide a dealer with an average of 16 targeted leads per month. Smart criticized MacDonald for badly underestimating the number of visitors to Consumer Guide's website and for failing to take into account the possibility that Publications might purchase additional leads on the wholesale market. Publications's second expert was Jeffrey Katz, an accountant. Katz adjusted Birnbaum's calculations to reflect what he regarded as more realistic assumptions: no sales of the Approved program, 35 sales per month of Leads & Listings, and attrition of 10% enrolled dealers per month. Katz also assumed that dealers would enroll in only one of the two programs.
The jury found that Publications breached the October agreement when it terminated Smart in November 2003, and that Smart was entitled to recover lost profits of $5,612,500. After the verdict was entered, Smart moved for prejudgment interest under the Illinois Interest Act, 815 ILCS 205/2. The district court, citing First National Bank of LaGrange v. Lowrey, 872 N.E.2d 447, 479 (Ill. App. Ct. 2007), denied the motion on the ground that the amount owed was not a fixed or easily calculated amount due. Publications then moved for judgment as a matter of law pursuant to Federal Rule of Civil Procedure 50(b), or in the alternative for a new trial or remittitur of damages under Rule 59, arguing that the evidence did not support the verdict of liability or the damages awarded for lost profits. The court denied those motions. Publications has appealed on both points, and Smart has cross-appealed from the denial of its motion for prejudgment interest.
Logically, the first question is whether there is any reason to upset the jury's finding that Publications breached the October agreement when it informed Smart on November 18, 2003, that it was canceling the deal. We do not need to concern ourselves with that, however, because Publications has limited its arguments on appeal to the damages issue. On the latter point, Publications offers two general arguments: first, that Smart's case was so deficient that the district court erred by denying its motion under Rule 50(b); second, that the damages award rested on such shaky ground that a new trial on damages is necessary. We address these points in turn.
Under Illinois law, which applies to this case, Smart had the burden of presenting to the jury sufficient evidence on which to determine the amount of its lost profits to a reasonable degree of certainty. Chicago's Pizza, Inc. v. Chicago's Pizza Franchise Ltd. USA, 893 N.E.2d 981, 994 (Ill. App. Ct. 2008). While courts do not ask for mathematical precision, they demand more than conjecture or speculation to support the jury's award. In re Estate of Talty,877 N.E.2d 1195, 1207 (Ill. App. Ct. 2007). Normally, an established business is able to satisfy its evidentiary burden by providing data about its past profits. Tri-G, Inc. v. Burke, Bosselman & Weaver, 856 N.E.2d 389, 407 (Ill. 2006). New businesses have more trouble coming up with hard evidence, though they are entitled to try. As the Supreme Court of Illinois has observed, "Generally speaking, . . . courts consider evidence of lost profits in a new business too speculative to sustain the burden of proof." Id. Even with these favorable legal rules, however, Publications faces a difficult task. Only if we are convinced, after looking at all of the evidence in the record, that there was not enough to support the jury's verdict may we find that Publications was entitled as a matter of law to a ruling that damages were not established. Tate v. Exec. Mgmt. Servs., Inc., 546 F.3d 528, 532 (7th Cir. 2008).
One preliminary question is whether the venture between Publications and Smart falls within the category of "new businesses" at all. Publications had been supplying internet leads to car dealers through middlemen, and Smart's principals had experience with conventional (i.e., non-web-based) promotions for car dealers. Neither side had ever engaged in an enterprise that wedded these two elements. The record leaves no doubt that the web-based element of the program was critical. Publications points out that the Leads & Listings program depended on the ability of the Consumer Guide website to generate enough leads to satisfy the average dealer's entitlement to 16 leads per month. The hard part was to develop a software program that would eliminate false leads, such as those with incorrect contact information, and then sort the remaining leads by geography and car-make. Publications initially looked into purchasing software from a middleman it had been using, Dealix, but it ultimately chose to develop its own product (and in the course of doing so failed to meet several deadlines). We are satisfied that the evidence readily supports the application of the "new-business" rules to this arrangement.
Much of the evidence tended to show how difficult it is to predict how successful the parties' venture would have been. Publications's problems with the software affected not only the Leads & Listings program, but also the Approved program. Some evidence indicates that the dealers did not want the Approved program without internet leads. Apparently there were no dealers in the end who chose to participate only in the Approved program, once Leads & Listings was available. There must have been some independent value to the Approved program, however, because 77 dealers remained in it after they signed up for Leads & Listings.
The efforts that Smart had made before the October agreement was signed are inconclusive, because all of this work was done before the software was ready in December 2003. Smart had to prove what its profits would have been from the date of the breach, November 18, 2003, until the contract expired in October 2005. The record is sorely lacking in concrete proof of that number. Neither party appears to have been able to track down the number of leads actually generated during the contract period (recall that Publications wound up selling leads through a third party). Their estimates are not based on anything obvious and they vary widely: Publications thought it was about 5,000 leads per month, and Smart thought it was somewhere between 10,000 and 40,000 per month. These numbers, however, are just guesses. And they seem to be guesses about gross numbers of leads, not the number that Smart in the end could have sold to dealers.
Without an established inventory of leads in hand, ready to be sold, both parties' forecasts are at best predictions-more or less optimistic-comparable to those that courts have seen in other start-up cases. See, e.g., Drs. Sellke & Conlon, Ltd. v. Twin Oaks Realty, Inc., 491 N.E.2d 912, 916-17 (Ill. App. Ct. 1986) (holding that a doctor could not recover lost profits related to the delayed opening of his new office); see also Mindgames, Inc. v. Western Publishing Co., 218 F.3d 652, 658-59 (7th Cir. 2000) (rejecting a claim for lost profits in connection with sales of a new board game). Given the fact that both parties' projections of leads necessarily depended on Publications's successful development and deployment of the sorting software, Publications argues that this case is no different from Computrol, Inc. v. Newtrend, L.P., which also happened to be governed by Illinois law. 203 F.3d 1064, 1070 (8th Cir. 2000). In Computrol, the Eighth Circuit found that the plaintiff's estimates of likely profits from a new software program were too speculative to satisfy Illinois law. It was influenced in this holding by the fact that the party claiming breach of contract had assumed that the program would be ready to use on schedule, even though there already had been delays in its development. Id. at 1071. Publications points out that other courts similarly have been resistant to awarding lost profits for new businesses that rely on cutting-edge technology. See, e.g., Trademark Research Corp. v. Maxwell Online, Inc., 995 F.2d 326, 333 (2d Cir. 1993) (holding that lost profit estimates for CDROM sales were too speculative when the CD-ROM database technology had never been implemented in the trademark field).
Smart counters that the general concept for the business that the parties were trying to establish was far from new in the market-many other wholesalers had already perfected the process of sorting internet leads. Moreover, it says, neither the Approved program nor the Leads & Listings program were the kind of new business venture that has concerned Illinois courts. Smart had been promoting both programs for many months by the time the October agreement was signed, and it had already built up a track record with the 428 Leads & Listings contracts it had sold. Publications notes that Smart's records reveal that it actually lost $475,805 during its relationship with Publications; there is no evidence that the deal would have become profitable after mid-November 2003. (Many new business ventures fail, after all; evidence is necessary before a court can conclude that any particular one would have been successful.) But Smart responds that these figures represented start-up costs, not losses, and that even those expenses would have been smaller had Publications paid the commissions that it owed.
In the end, one cannot escape the conclusion that Smart's sale of 428 Leads & Listings contracts represented only pre-orders for future, hoped-for leads. Those projections rested on the available evidence of the predictable number of monthly leads. MacDonald, one of the experts called by Publications, conceded that the website produced about 5,000 leads per month in 2005. Publications had been selling the leads on the wholesale market to Dealix and Info4Cars for some time. Smart reasons that this number represents proven market demand for the leads, and that this in turn shows that this was an established market, not a new one. See Milex Prods., Inc. v. Alra Labs., Inc., 603 N.E.2d 1226, 1237 (Ill. App. Ct. 1992) (concluding that lost profits were reasonably certain where evidence showed that a new product had an established market).
Even if we thought that the price of the internet leads was established, Smart would still need to show how successful it would have been in selling those leads to car dealers. Its evidence on this point fell short. Welch based his estimate of the number of leads Smart could sell to dealers in part on his experience selling Consumer Guide in 2001, but he did not provide any hard data from comparable firms to corroborate his calculation. We cannot find any place where Smart even estimated the ratio of whole-sale leads to retail sales. The most we see is the conservative estimate advanced by Publications, which is based on MacDonald's testimony that Smart might be able to sell dealers 350 leads out of an inventory of 5,000. But even this estimate is problematic, largely because it was not contemporaneous. It was based instead on MacDonald's experience with Consumer Guide 18 months after Publications canceled the contract.
There are other reasons, too, why MacDonald's testimony does not provide the necessary precision. When his firm sold leads to dealers, it appears to have bundled the Consumer Guide leads with leads drawn from 180 other websites; only in that way could a dealer be sure of receiving 33-34 leads per month. There is no evidence that dealers would want to receive only the two to three leads that MacDonald estimated Smart would have been able to deliver. Compare TAS Distributing Co., Inc. v. Cummins Engine Co., Inc., 491 F.3d 625, 635-36 (7th Cir. 2007) (explaining that it was speculative under Illinois law for a plaintiff to compare its sales to those of a third party selling a slightly different product). Thus, while Smart might have shown that dealers were eager to sign up for a Leads & Listings contract promising 16 leads per month, it has not demonstrated that the same dealers would have wanted a contract that delivered only two or three leads per month.
Publications insists that any assessment of Smart's damages requires heaping one speculative inference on another, and another, and another. To calculate damages, the jury had to assume first that Publications's software would work properly, next that it would produce the promised number of leads for enough dealers, and finally that Smart would be able to sell these leads to the dealers. The evidence of Smart's past profits was predicated on its projections about the likely inventory of leads, but these predictions were unreliable. The jury's task would have been easier if either party had provided solid evidence of other closely analogous businesses, but they did not. That said, we are reluctant to treat this as a case in which the court erred by sending the matter to the jury at all. It is conceivable that the evidence taken as a whole might have supported some level of damages for Smart. Rather than engaging in conjecture ourselves about what that number might have been, however, we prefer to say only that we find no error in the court's denial of Publications's Rule 50(b) motion for judgment as a matter of law, and to move on to its motion for a new trial on damages under Rule 59.
In our view, Publications has succeeded in showing that the jury's verdict was excessive, and thus that it is entitled to a new trial limited to damages. The question whether the jury's damages award was excessive is controlled by Illinois law, and we review the district court's ruling on Publications's Rule 59 motion for an abuse of discretion. Gasperini v. Center for Humanities, Inc., 518 U.S. 415, 418-19 (1996); Shick v. Illinois Dept. of Human Servs., 307 F.3d 605, 611 (7th Cir. 2002); Briggs v. Marshall, 93 F.3d 355, 360 (7th Cir. 1996). A district court is entitled to find that an award is unsupported by the evidence and thus that a motion for a new trial under Rule 59 should be granted, even if there was enough evidence in the record to justify sending the issue to the jury in the first instance (a finding that would require denial of a motion for judgment as a matter of law under Rule 50). See 9B CHARLES ALAN WRIGHT & ARTHUR R. MILLER, FEDERAL PRACTICE AND PROCEDURE § 2531, at 65-67 (2d ed. 1995).
In Illinois, "[a]n award of damages will be deemed excessive if it falls outside the range of fair and reasonable compensation or results from passion or prejudice, or if it is so large that it shocks the judicial conscience." Best v. Taylor Mach. Works, 689 N.E.2d 1057, 1079 (Ill. 1997) (internal quotation marks omitted). Publications first attempts to satisfy this standard with the argument that the district court should have disregarded some of the evidence on which the jury was permitted to rely. We are not persuaded, however, that the district court erred when it allowed the jury to hear evidence about the parties' course of dealing before the October 2003 agreement was concluded. Smart wanted to use this evidence to estimate what Publications's commission payments to it would have been if Publications had not breached. Publications is on somewhat firmer ground when it objects that Welch and Magarity (Smart's principals) improperly offered lay testimony about their projections for the attrition and renewal rates for the two programs. Neither one was basing his estimates on his own experience in the day-to-day affairs of the (new) business. See Von der Ruhr v. Immtech Int'l, Inc., 570 F.3d 858, 862-65 (7th Cir. 2009) (excluding a business owner's testimony about marketing a new drug because he had little personal experience with the process). It is difficult to say, however, how much this testimony harmed Publications. Welch's and Magarity's estimates were the only foundation for Smart's damages model, but the jury did not accept that model in its entirety. Instead, the jury awarded 64% of Smart's requested $8.8 million. How it came up with that discount rate is unclear.
Indeed, the record is sorely lacking in evidence that would have supported such a high estimate of Smart's damages. There were numerous weaknesses in Smart's case: Welch and Magarity were starting up a new internet business for which they had little to no relevant prior experience; Smart had no solid evidence about the number of leads that the Consumer Guide website was producing; and Smart did not provide reliable evidence about the ratio of wholesale leads to retail sales. Smart has no answer to these problems apart from its insistence that it had a proven track record of sales for the two programs during the period before the October agreement and before the breach. But, as Smart itself admits while discussing a different point, almost all of those sales were made in contemplation of a future program, and the dealers were becoming frustrated by Publications's failure to deliver what it had promised. Smart was not going to earn much if Publications continued to struggle with its software or if the website proved incapable of delivering approximately 16 leads per month. Publications itself might have fallen into the trap of taking too rosy a look at its prospects; there is no evidence suggesting that the Consumer Guide website would produce anything more than 10,000 leads a month, which is nowhere near the 40,000 leads that Smart projected. And, it is worth repeating, these leads had to be converted into information tailored to each dealer's location and brand of auto.
We conclude that this damage award fell so far outside anything the evidence might have supported that the district court abused its discretion in refusing to order a new trial on damages.
The only question that remains is Smart's cross-appeal on the issue of prejudgment interest. Although this is no longer directly relevant, given our decision to set aside the damages verdict, we will say a word about it since the same question is likely to arise again.
Publications complains that Smart filed its motion too late and thus forfeited this issue, but we do not read the rules that rigidly. Rule 59(e) motions for prejudgment interest must be filed before judgment is entered, First State Bank of Monticello v. Ohio Cas. Ins. Co., 555 F.3d 564, 572 (7th Cir. 2009), and Smart complied with this rule (even though it did not submit the motion before the jury rendered its verdict). See Osterneck v. Ernst & Whinney, 489 U.S. 169, 175-76 (1989). The district court thus correctly reached the merits of Smart's motion.
In Illinois, plaintiffs may be entitled to prejudgment interest for contractual damages if the damages are "subject to easy computation." Oldenburg v. Hagemann, 565 N.E.2d 1021, 1029 (Ill. App. Ct. 1991). Everything we have said thus far demonstrates why this case fails to meet that standard. See also Cushman & Wakefield of Illinois, Inc. v. Northbrook 500 Ltd. Partnership, 445 N.E.2d 1313, 1321 (Ill. App. Ct. 1983) (denying prejudgment interest on damages award for lost future commissions). The district court committed no error, clear or otherwise, when it rejected Smart's request.
The judgment of the district court on liability was not challenged on appeal, and so we do not disturb it. The judgment on damages, however, is VACATED and the case is REMANDED for further proceedings consistent with this opinion.
© 1992-2010 VersusLaw Inc.