TOYSRUS.COM, L.L.C. v. AMAZON.COM KIDS, INC.

 

NOT FOR PUBLICATION WITHOUT THE

APPROVAL OF THE APPELLATE DIVISION

SUPERIOR COURT OF NEW JERSEY

APPELLATE DIVISION

DOCKET NO. A-3391-06T23391-06T2

TOYSRUS.COM, L.L.C., TOYSRUS.COM,

INC., TOYS "R" US, INC., and

GEOFFREY, INC.,

Plaintiffs-Respondents/

Cross-Appellants,

v.

AMAZON.COM KIDS, INC. f/k/a

ROCK-BOUND, INC., and

AMAZON.COM, INC.,

Defendants-Appellants/

Cross-Respondents.

 
 

Argued December 17, 2008 - Decided

Before Judges Parrillo, Lihotz and Messano.

On appeal from the Superior Court of New Jersey, Chancery Division, Passaic County, Docket No.

C-96-04.

David R. Goodnight (Stoel Rives LLP) of the Washington bar, admitted pro hac vice, argued the cause for appellants/cross-respondents (Drinker Biddle & Reath LLP, and Mr. Goodnight, attorneys; Mr. Goodnight, Jill D. Bowman (Stoel Rives) of the Washington bar, admitted pro hac vice, Vanessa Soriano Power (Stoel Rives) of the Washington bar, admitted pro hac vice, Alfred Arthur Day of the Washington bar, admitted pro hac vice, Richard E. Brennan, and Walter J. Fleischer, Jr., on the brief).

Michael Dockterman (Wildman, Harrold, Allen & Dixon LLP) of the Illinois bar, admitted pro hac vice, argued the cause for respondents/ cross-appellants (Bray, Miller & Bray, and Mr. Dockterman, attorneys; Mr. Dockterman and Richard M. Hoffman (Wildman, Harrold, Allen & Dixon LLP) of the Illinois bar, admitted pro hac vice, of counsel and on the brief; Peter R. Bray, on the brief).

The opinion of the court was delivered by

PARRILLO, J.A.D.

This action involves a long-term contract between plaintiffs, Toysrus.com, LLC, Toysrus.com, Inc. and Geoffrey, Inc. (collectively, TRU), and defendants, Amazon.com Kids, Inc., f/k/a Rock-Bound, Inc. and Amazon.com, Inc. (collectively, ACT). TRU filed a complaint alleging breach and seeking damages and declaratory and injunctive relief. Following a bench trial, the General Equity Part judge terminated the parties' agreement, ordered them to cooperate during a ninety-day wind-down period, but denied their respective requests for damages and injunctive and declaratory relief. That opinion was memorialized in a March 31, 2006 final judgment and April 3, 2006 amended final judgment, followed by post-judgment orders of June 22, 2006, September 14, 2006 and October 20, 2006, all of which were incorporated in the court's February 27, 2007 final order, from which this appeal by defendants and cross-appeal by plaintiffs are taken.

The facts are rather complicated and somewhat in dispute. In August 2000, ACT and TRU executed a ten-year "Strategic Alliance Agreement" (SAA) providing that the parties "cooperate to launch certain co-branded portions of the [Amazon website] dedicated to the sale of certain toy, game and video game

[, including baby,] products selected and provided by [TRU]"; that ACT provide support for TRU's sales through services ranging from website development to product distribution; and that, in addition to a set up fee, TRU pay a significant annual fee, $50,000,000, additional fees based on its use of specified services, and commissions.

The term "co-branded stores" referred to a sub page of the Amazon website featuring products the parties offered for sale pursuant to the agreement. To reach the sub page, ACT was to create a co-branded toy and video game store navigation tab to be consistent with and "generally as prominent as the Major Tabs on the Amazon home page." Pursuant to the agreement, ACT also had control of the "functionality" of the co-branded stores, including "site navigation, site look and feel," subject to certain restrictions of reasonableness and non-discrimination respecting the way TRU's selected products were "position[ed] in search results." The parties would consult periodically regarding functionality issues.

Section 5 of the SAA addressed selection of the products to be offered on the website. Under Section 5.1.1, TRU was given broad discretion in the selection of its products for sale through the co-branded stores; however, Section 5.1.3 required TRU to provide a comprehensive selection of its products, including, subject to availability, "at least the then-current Top Toy and Video Game Product SKUs," at least 8000 "different Selected Product SKUs," and other specific numbers of baby products and children's sporting goods.

At least from TRU's perspective, the cornerstone of the agreement was its right to sell certain products exclusively, and Section 5 expressly addressed TRU's "exclusivity" rights. Thus, if TRU did not select the product for sale in the "co-branded" virtual stores under Section 5.1.1, then Section 5.1.4 permitted ACT and its affiliates, but not third parties, to sell the unselected product. TRU, however, was afforded the right of recapture, namely to resume or initiate sale of products ACT might then be selling pursuant to Section 5.1.4, provided TRU gave notice to ACT and agreed to purchase ACT's then existing inventory of the applicable product.

In other words, under the SAA, TRU's right to sell toys, games and baby products is "exclusive" only to the extent the agreement restricts sales by others. In addition to Section 5, and with specific reference to third parties, Section 12.1.1 limits ACT's right to sell or permit others to sell; and Section 12.1.2 exempts specified sales from those that ACT may not make or allow others to make.

Specifically, Section 12.1.1 prohibits ACT and its affiliates from selling or permitting a third party to sell any "Exclusive Products" and "Selected Exclusive Products" through the Amazon website. "Exclusive Products" are "toys and games" and "exclusive baby products" (both defined terms). "Selected Exclusive Products" are those "exclusive products" that TRU selects for sale in one of the two co-branded virtual stores. Section 12.1.1 also prohibits ACT and its affiliates from selling or permitting a third party to sell any "Selected Non-Exclusive Products," which are products, other than "Exclusive Products," that TRU selects to sell through either co-branded store.

The restrictions on sales included in Section 12.1.1 are "[s]ubject to Section 12.1.2," which permits certain sales and provides: "[TRU] acknowledges and agrees that nothing in this Agreement will prevent or otherwise restrict: (a) any sales of products . . . occurring in connection with Programmatic Selling Initiatives . . . ." "Programmatic Selling Initiatives" are defined as:

any area, feature or service of the ACT Site through which Third Parties may sell products or services on terms available to the general public (or, in the case of "sothebys.amazon.com," to a defined class of dealers (including without limitation, the existing "Auctions," "zShops," "sothebys.amazon.com," and "Amazon.com Advantage" areas and services of the ACT Site).

[(parentheses appear above as in the original)].

Thus, this so-called "garage sale" provision permits third parties to sell any products in connection with programs that qualify as "Programmatic Selling Initiatives," which, generally speaking, are available to all of ACT's users under terms available to the general public.

Section 12.1.2 also provides that the SAA does not restrict:

(c) ACT and his Affiliates from selling, and permitting Third Parties to sell, Exclusive Products through the ACT site (other than through the Co-Branded Stores), provided that such sales by ACT and its Affiliates, or any such Third Party . . . do not constitute more than three and one-half percent (3.5%) of the Exclusive Product Revenues for any Year (or, for the fourth Quarter of 2000, for such Quarter) . . . .

[(emphasis added).]

Thus, subsection (c) of Section 12.1.2 permits third-party sales of "Exclusive Products" up to a specified percentage of total sales, namely the three-and-one-half percent ceiling.

To recap, TRU was given the right to select products to be offered for sale in the "co-branded" virtual stores. Any exclusive product "selected" by TRU for sale pursuant to Section 5.1.1 became by definition a "selected exclusive product." While ACT or its affiliates were allowed to offer for sale exclusive products which TRU elected not to sell on the website, the SAA included no such right for third parties, other than under the limited exception of Section 12.1.2(c), which capped such sales at a 3.5% ceiling. In any event, third parties were not permitted to sell "selected exclusive products" other than through "programmatic" selling initiatives under Section 12.1.2(a).

On the other hand, TRU was required, under Section 8.2, to maintain a certain inventory of the items offered on the website to meet the anticipated demand. Specifically, TRU was obligated to use "commercially reasonable efforts" to ensure a 90% minimum level "in stock" availability on all selected products at all times "at one or more ACT distribution centers." In order to ensure compliance, TRU tracked the distribution center inventory through "feeds" with ACT, supplemented by its own "forecasts," its "catalog[,] and the web site." ACT assigned a so-called Amazon Standard Identification Number, or ASIN, to each item based on product information provided by TRU.

The SAA was to run until December 31, 2010, unless terminated earlier. Section 13 gave ACT the right to terminate in the event that it did not realize a minimum return under the parties' fee structure as provided in the agreement. Section 15.2 permitted "termination for cause" in the event of insolvency, or material breach and failure to cure after due notice. The parties also retained the right to terminate in the event one experienced a "competitive change of control" with respect to the other. In the event of termination, Section 15.4 identified provisions of the agreement that would survive. Under Section 15.6, the parties recognized a "post-termination transition period."

Pursuant to Sections 18.1 and 18.2, the parties agreed on a limitation of consequential damages and that neither would be liable purely as a result of termination. Section 22 addressed resolution of disputes, including mandatory but non-binding mediation. Delaware law was to govern the agreement, which included an integration clause.

By all accounts, the relationship forged between the two and memorialized in the SAA was unique and unlike any other arrangement either had entered into before. Clearly, the parties were venturing into pioneer territory, the frontiers of which were both dynamic and not fully realized at the outset. Yet the prospect of their complimentary contributions that would render the whole much more profitable than its individual parts made the merger inviting and attractive. To truly understand then why such a promising relationship collapsed less than midway through its expected lifespan, it is necessary to examine events both before and antecedent to execution of the contract.

TRU, a Delaware LLC, is a domestic and international retailer of toys, games and baby products. It considers itself "one of the fabulous brands of the . . . world," and prides itself on providing the toy consumer "the high probability of finding" a wanted item in its store stock over any of its competitors. "The real proof of the strength of the brand is in the last three or four weeks before Christmas," when TRU continued to carry popular items after competitors had sold out of product.

TRU was valued at $2.2 billion in 2000 and $6.6 billion in July 2005, its "brand" being worth about one-sixth of the company's overall value, according to its expert. As of 2000, TRU had approximately 1500 stores worldwide. Seeking to expand its ever-increasing market share, TRU introduced online shopping in 1999, but its "fledgling website" simply did not have the logistical capacity to respond to online consumer demand, and as a result, the company was left with "a big black eye."

Meanwhile, ACT was launched in 1995 as an online book seller, but consistent with its vision for an internet superstore, gradually sold other products, including online music, videos, DVDs, and toys. In fact, ACT had developed its own line of specialty toys, employing its own "buyers and planners who were responsible for identifying and purchasing toy products and creating relationships with toy manufacturers and vendors." By 2000, however, ACT had also launched three programs through which third parties could sell their products on Amazon.com: zShops, Auctions, and Advantage. The zShops program was successful by allowing merchants to list products for sale on the Amazon platform at a fixed price, while Advantage was a "consignment selling model" whereby "Amazon.com receives inventory . . . from the Advantage seller, and then when a customer comes to Amazon.com and buys that product, we pay the seller on . . . a consignment basis." Eventually, the Amazon website evolved into a virtual "marketplace," which allowed anyone to list items on the Amazon platform.

Considering its first foray into online shopping proved problematic, and concerned about fulfilling online orders for the upcoming holiday season, TRU approached ACT in January 2000 about a possible business alliance. From TRU's perspective, ACT could provide "web-hosting, order fulfillment, customer service and other functions." From ACT's standpoint, TRU would help "merchandise" and market its site and add to its assortment of selections. Thus, in initial draft agreements, it was envisioned that TRU would be responsible for inventory and marketing while ACT would address "the site, fulfillment, and customer service." As the idea took shape, TRU

would own the inventory. [TRU] would ship it. [TRU] would source it and ship it into the Amazon fulfillment facilities. And [ACT] would operate the site, connect customer demand to inventory through the ordering process, pick it, pack it, ship it, and at the time that it was shipped in and billed, they took ownership of the inventory . . . .

In giving up the right to sell its products on any "domain name" or website of its own or a third party's, TRU focused instead on "[m]aking sure the exclusivity of [its] ability to sell toys, games and baby products was maintained." TRU also wanted its shoppers to "be easily linked across" its product sites, as, for example, if one of its online toy customers wanted a Babies "R" Us product. There were a variety of reasons why exclusivity was important to TRU. According to John Eyler, TRU's Chief Executive Officer (CEO):

First of all, we had to ensure that we could get quality fulfillment [of an order], flawless fulfillment, as close as we could get. That's why it was our opinion that unless . . . [we] had a partnership arrangement to make sure it was successful, that the risk would increase that they simply would not focus enough time, attention, and effort to make sure that fulfillment was done correctly.

Secondly . . . [s]ince all this was all about brand, it was very important that this be done in a way that was perfectly clear to the consumer that they were dealing with Toys "R" Us. When we saw the way the site was presented as a co-branded site . . ., it was really the legitimizing of one source, a partner called Amazon, to be the best in the space of fulfilling our consumers' needs.

Thirdly, exclusivity was important to us because . . . we needed exclusivity to make sure that unconsciously our brand wasn't eroded by people who don't do this business as well as we do and make different choices than we do. We didn't want customer dissatisfaction from other sellers to taint our brand.

Finally, we needed a partner if we were going to go outside our own, where we could sell this assortment similarly to how we source. . . . So one set of eyes was buying it, pulling some to Amazon and some to [our] stores. If we had to do two assortment strategies, we couldn't do that.

Thus, Eyler envisioned a true partnership wherein Amazon and Toys "R" Us would, in effect, become "anchor" or dominant stores in the Amazon.com "virtual mall," with Babies "R" Us only slightly less prominent. According to Ray Arthur, TRU's Chief Financial Officer, Amazon was "primarily known for books, music and video," and TRU would "become attached to that space, the mall, and become another anchor tenant to sell toys and children's products," with TRU's "baby store" an additional "smaller store" at the online virtual mall. Eyler believed "[t]here certainly would be other stores [within the rest of the Amazon 'mall'], but there certainly would not be other toy stores or other baby stores."

TRU's vision was directly conveyed to Jeff Bezos, CEO and Chairman of ACT, who nonetheless "had a strong view that more assortment is better." From ACT's perspective, "the really founding basis of Amazon.com was to have selection[,]" and the "only way I know of to [have a huge inventory necessary for a large selection] . . . is with third parties, and that was understood at the time of these negotiations." In particular, Bezos wanted to be free to use other merchants or third parties to sell "the stuff Toys 'R' Us did not want to sell." ACT, already a "[l]eading online retailer," sought to be "the world's most customer-centric company, where customers can find and discover anything they might want to buy online."

Nevertheless, Mark Britto, ACT's senior vice-president of worldwide sales who participated in the negotiations, acknowledged that exclusivity was a critical feature of the transaction and that TRU's vision "was to have Amazon exit the toy business in [TRU's] favor and yet provide the fulfillment and customer service that is the hallmark of [ACT's] identification." Indeed, while the compromise ultimately reached provided only for limited "carve-outs" detailed in Section 12.1.2, every other section of the draft agreements that had made reference to third parties' ability to sell on the Amazon site had been removed from the final SAA. According to Harrison Miller, ACT's vice-president and general manager of toys who helped negotiate the SAA, ACT continued to convey the impression during the negotiations that TRU was the exclusive seller of toys on the Amazon website. Indeed, Miller further acknowledged that the SAA does not provide that third parties could sell products marketed by TRU on the Amazon website.

To be sure, the parties, as noted, agreed to certain limitations on TRU's otherwise broad exclusivity rights. Eyler recalled that the only exceptions discussed during negotiations were the inventory ACT already had on hand, ACT's right to "acquire and sell through our site any items that we did not choose to buy and to execute," and the so-called "zShops," or a "sort of flea market, people who had things to sell and sell in these environments they were creating." Such zShops, including "Auctions" and "Amazon.com Advantage," were acknowledged by Arthur, TRU's CFO, as the Programmatic Selling Initiatives (sometimes PSIs) carved out as an exception in the SAA. The second exception acknowledged by Arthur was what was referred to as "overlapping products," meaning that vendors who were not in direct competition with TRU could sell "certain products that could be on the fringes overlapping with other areas." Arthur cited the example of a pet store that sold dog gates that were essentially the same as TRU's baby gates. He also noted how TRU negotiated the cap on the sale of such "overlapping products" down from fifteen to three-and-a-half percent.

Arthur expressed his understanding of TRU's exclusivity rights thusly: "[I]f I said it once I said it a thousand times during negotiations, we would never see Wal-Mart, never see Target, never see eToys. We would never see our competitors on the market" with ACT. According to Arthur, therefore, "we had to get over that philosophical disagreement and we thought we did, because we never would have entered into an agreement that we weren't solely and exclusively in charge of."

At least for some time after execution of the SAA, ACT abided by TRU's understanding of its exclusivity rights. Thus, for example, agreements ACT signed with third-party merchants prior to September 2004 for sale of products on the Amazon website typically excluded the sale of toys, games and baby products. Most notably, on August 21, 2001, ACT and Target, a competitor of TRU, entered into their own SAA which, among other things, restricted Target from selling "toys, games, or baby products" through its "store" on the Amazon website. According to Target's president, Dale Ritschke, Target had a "desire to sell all of those types of products on the Amazon site, and we spent some energy around defining exactly what a toy product was or a baby product was so we . . . would not break [ACT's] previous agreement" with TRU.

Both parties agree that the first year of operation under the SAA went smoothly. Once the co-branded stores were launched in September 2000, an internet user typing in either "www.toysrus.com" or "www.babiesrus.com" was taken directly to the Amazon.com website, which featured a "Toys & Games" tab that, when clicked on, would direct a user to the co-branded toy store. Once "inside," if the visitor entered the name of the toy into the search bar and hit "go," search results were returned for products sold only by TRU. As a result of the "admirable performance" of both parties, the first year of operation was highly successful. Indeed, TRU's John Barbour boasted that "[TRU] became the #1 worldwide e-commerce site for kids in terms of traffic for the majority of the year."

Unfortunately, problems surfaced soon afterward. Most notably, ACT's August 21, 2001 deal with Target, one of TRU's biggest competitors, gave TRU pause for concern. Even though TRU was assured that Target was not going to be allowed to sell toys on Amazon.com, Target "had a second site off the platform where they did sell toys and where Amazon also provided the technology for the fulfillment." Eyler perceived ACT's deal with Target as a "thinly veiled way of getting around the exclusivity commitment that [ACT] had to [TRU]." For instance, while customers would not be able to "link from the Target store at Amazon.com to Target.com, and then buy a toy there," nevertheless on the so-called "subnav" portion of the Amazon website, Target "appears . . . underneath the toys and games tab." Arthur's understanding, however, was that because Target was not supposed to be offering toys and games on the Amazon site, "it would not be in the toys and games tab." Moreover, "[w]hen the Target deal was implemented on the home page of the Amazon platform in the sub-navigation area, there was a branded Target button that you could hit that would bring you to their boutique on Amazon." TRU, however, had "fought very hard to get branding put on the toys and games tab" during its negotiations, but ACT was unwilling to agree, giving TRU only a non-branded tab on product "categories." But to Arthur, "[w]hen that [Target] tab showed up on the home page, quite frankly, I saw it as much as branding on the tab as being right below it."

Furthermore, in an internal memo of September 2001, ACT's Miller pointed out that by the following year's holiday season, TRU could expect to see video games offered by Target on the Amazon site. Miller further asserted that "[i]t was a deal like this that we contemplated when we made flexibility in this category [of selling video games] a major deal point . . . ." Although the SAA between ACT and TRU had not made video games "Exclusive Products," ACT initially had allowed TRU to be the exclusive online seller of video games.

Whether or not meant as an accommodation for the Target deal, on April 17, 2002, the parties executed their first amendment to the SAA, whereby the fee escalator was discontinued and the base fee was capped at $50 million through 2004, with adjustment thereafter for the CPI through 2010. That same day, the parties also signed a consent to amendment and release by which TRU effectively relinquished its claims against ACT with respect to the Amazon/Target relationship.

In 2002 and 2003, continuing its policy of "fill[ing] out particular categories," ACT followed the Target deal with a series of "merchant@ agreements," or "specifically negotiated agreements where third parties could sell products on the site." According to ACT's Rudy Gadre, corporate counsel and vice president, these individualized agreements initially included the same types of prohibitions against selling selected toys, games and baby products as ACT included in its agreement with Target. At a certain point, however, the agreements were amended to permit the sale of "non-selected" items. When asked why ACT did not provide for the sale of non-selected items in its initial merchants@ contracts, Gadre explained that, originally, ACT lacked the "technical ability to screen out

. . . selected [Toys 'R' Us] products," that "there were discussions ongoing all the time with Toys 'R' Us [as to] how they could expand selection and . . . provide the SKUs that we thought we were missing," and that at that point ACT was concentrating on expanding other categories such as apparel, office products and sporting goods.

At around the same time, TRU began noticing "a lot of things that were disturbing to us as [other vendors'] products began showing up on the site, as discussions were being held of other initiatives that were going to be made that we felt would dilute our exclusive rights on the site." For example, running a check at one point Arthur typed in the word "toy" in the search box on Target's "boutique on the Amazon platform" and received 754 results, the very first of which was a toy that TRU did, in fact, carry. There were other instances as well, which according to Arthur, created "incredible confusion" for customers and was not what TRU negotiated for.

When the issue first surfaced, ACT "took the [offending] product down." In fact, ACT's Gadre agreed that, up to a point, when TRU challenged other products on the Amazon site, "we endeavored to take them down whenever possible." Eventually, however, "it became more of an issue." A number of times, ACT approached TRU for permission for third-party sales of "exclusive" toy items on the website in return for some concession on ACT's part. In 2003, for example, ACT's Van der Meulen offered TRU a revenue share for another merchant's sales of skateboards and accessories on the Amazon website. An amended agreement to this effect was drafted, but never signed. Nevertheless, ACT eventually began selling skateboard products that "overlapped" TRU products, with Gadre believing that such sales were permitted in accord with a "safe harbor," "handshake agreement" between Van der Meulen and Arthur. ACT never paid TRU any revenue from the skateboard sales. Moreover, if "exclusive" or "selective exclusive" products were out of stock, ACT would reclassify them as "non-exclusive", and allow third-parties to sell those items.

Simultaneously, paid advertisements began appearing on the Amazon site, including within the co-branded store, that consisted of one or more links to separate, third-party websites. When TRU's Arthur "hit on one of them," he found "strollers that were in [TRU's] exclusive category for sale." When Arthur complained about these so-called "sponsored links" or "hot links" that direct customers to third-party websites, in particular the fact that TRU's "primary competitors" such as eToys or Bye-Bye-Baby began "show[ing] up" as links, he was assured the practice would end.

Instead of being discontinued, however, the practice became a "permanent feature of the Amazon web site." That was significant for TRU, because not only was it "very difficult to get a customer, [but] . . . here we are sending a customer to someone else to buy product." TRU feared losing business as a result: "[O]nce someone goes off the site and transacts with someone else, there is a good chance that they will continue to transact with that person." Moreover, ACT did not have information as to which sponsored links were clicked on by customers, or of "what takes place once a user clicks on a sponsored link and goes to a third party web site." ACT got paid "based upon a click" on the sponsored link, although it was generally Google, and not ACT, who controlled "which third party web sites are advertised as Sponsored Links on the ACT site in response to a given set of keywords."

As a result of these growing concerns, Arthur and Eyler met with Van der Meulen and Bezos in New York City in the summer of 2003. ACT remained interested in increasing selection, and suggested compromising TRU's exclusivity rights in exchange for revenue sharing. TRU, however, was reluctant to agree to any exclusivity compromises because of concerns over quality control.

TRU also had concerns over the amount of compensation for its partial surrender of exclusivity rights as well as the workability of exercising its so-called online "recapture rights" when non-selected products sold by third parties became good market sellers. In this regard, Van der Meulen had wanted TRU to agree that it "would not recapture those products and select them and knock the third-party sellers off the site." TRU's position, however, was that "third parties can't sell on the web site."

According to Van der Meulen, as of the New York meeting, ACT did not plan to allow third parties to sell products with the same UPC as TRU's products, but acknowledged that it was a difficult problem to resolve. "Similar" but not identical products with UPC codes different from TRU's codes would be permitted, such as an item sold as a "set" but which might include some individual items offered separately by TRU. ACT wanted to add "selection," and Van der Meulen said it was simply more "scalable" to use third parties to add such selection than for ACT to add more products itself. Also as of the New York meeting, ACT did not have the technology to "weed out selected exclusive products," but was working on developing it.

It was at the New York City meeting that TRU was first alerted to what it perceived as perhaps the greatest threat to its exclusivity bargain, namely the impending implementation of ACT's 1X1 GUI technology, a dynamic step forward in the Amazon "marketplace" concept. 1X1 GUI was developed as a technology tool to provide "another means by which third party sellers could add items to the Amazon.com catalog" via detail pages with various product information that would appear in search results pages. A "detail page"

is ultimately a description of the product that a customer can choose to buy. It has many different features associated with it. The image of the product, the description of the product, editorial reviews about the product, customer reviews about the product, a list of items that are similar based on purchase history for that product, and other merchandising features associated with it.

The 1X1 technology merged the once separate conceptual identities of the retail stores and the PSIs, creating the one-stop, multiple-buying-options online superstore that Amazon is today, where merchants, retailers, and used items are all essentially "side-by-side" in the same online store.

ACT presented the 1X1 GUI technology as a programmatic selling initiative that was permitted under the SAA, but TRU disagreed, maintaining that the "one to one" technology was not in place when the parties executed the agreement, and that the parties drafted the definition "so that it would be limited to the type of programs in existence at the time the [SAA] was negotiated" and to permit ACT to develop similar programs. According to TRU, that was the parties' purpose for listing specific programs in the definition ("Auctions," "zShops," "sothebys.amazon.com" and "Amazon.com Advantage").

Quite obviously, no understanding was reached at the New York meeting. Bezos suggested the parties continue discussions but warned that eventually ACT "would introduce a different selection on the toy store whether we came to a conclusion or not." Discussions did continue, but to no avail:

We could not come to agreement. Amazon would not provide us with the functionality. They would rather run it themselves. Amazon was unwilling to let us enter into agreements with them and third parties to control the content, and they were unwilling to enter into agreements where we maintained our recapture rights.

Although Bezos was advised by his own people that TRU "would sue" if ACT implemented its proposal unilaterally, on October 27, 2004, ACT launched its 1X1 GUI technology for third-party sellers in the categories of toys, games and baby products. Then, on November 5, 2004, ACT signed a merchant agreement with eToys Direct, TRU's biggest competitor in e-space. What is more, in November 2004, ACT began amending its merchant@ agreements by allowing sale on the Amazon site of what were exclusive TRU products, subject to Amazon's right, with notice, to prohibit sale of particular items. Many of those vendors were toy companies. Once ACT "officially launched" its 1X1 GUI technology, "thousands" of what were otherwise exclusive products under the parties' agreement were added to the Amazon website by third-party vendors. In other words, this new technology allowed third parties to sell products in the same conceptual "marketplace" as TRU. Yet at the same time, ACT lacked the technology to prevent these same vendors from selling products "exclusive" to, or selected exclusively by, TRU.

By then, TRU considered its relationship with ACT "irreconcilably broken," and sought to have it terminated. In fact, even before ACT implemented its "one to one" technology, and as early as February 2004, ACT itself recognized that merchants@ vendors "could potentially create overlapping -- they could sell the same products as Toys 'R' Us products, and we wouldn't be able to control that." Indeed, on April 2, 2004,

faced with evidence of a continuing erosion of its "exclusivity," TRU sent ACT a list of more than one thousand items of "exclusive product being sold by other vendors." The list was not intended to be comprehensive, but represented TRU's best effort at identifying exclusive products being sold by other vendors. ACT itself calculated exposure when Michelle Rothman, ACT's senior manager of client services and responsible for the day-to-day operations supporting the TRU account, identified 119 selected exclusive products through May 10, 2004 that had been sold by third parties. But she acknowledged that as of that point, ACT still did not "have all the technologies or all the processes . . . in place" to stop "the addition of exclusive products using the 1X1 GUI tool."

On May 21, 2004, TRU filed a complaint against ACT in the General Equity Part seeking injunctive and declaratory relief, and damages. TRU asserted specific contractual breaches in ACT's use of the 1X1 GUI technology, which it claimed was not a Programmatic Selling Initiative; in its permission to third parties to sell exclusive products in excess of the 3.5% safe harbor provision; and in its alteration of the site's navigation and "search functionality" to lessen TRU's prominence. More generally, however, TRU also alleged breach of the covenant of good faith and fair dealing with respect to its exclusivity rights. ACT answered and counterclaimed. Upon application by TRU for a preliminary injunction, on June 3, 2004, the court issued an order temporarily restraining ACT from implementing its "one to one" technology in connection with ACT's sale of products identified in its toy, game or baby product categories. ACT sought leave to appeal, which we granted on September 16, 2004. While that matter was pending, on November 30, 2004, TRU sought a second temporary restraining order to address interim actions ACT had allegedly undertaken to market "exclusive" products during the 2004 holiday season in direct competition with TRU. On December 3, 2004, the trial court signed a second order to show cause with temporary restraints.

Following a hearing, on December 23, 2004, the court issued an order enjoining ACT from selling items known as "Selected Exclusive Products" under the SAA, and otherwise setting forth that ACT's failure to comply would result in the imposition of damages and sanctions. On December 30, 2004, TRU filed a motion for sanctions, asserting that ACT was continuing to violate the court's December 23 order by permitting third parties to market "Selected Exclusive Products," and on February 28, 2005, the court issued its opinion sanctioning ACT. By order of March 21, 2005, the court awarded TRU $18,000.

On April 12, 2005, we vacated the preliminary injunction and restraints as directed in the trial court's June 3, July 26, and August 12, 2004 orders. The matter proceeded to trial between September and November 2005. At trial, both parties traced their relationship from inception to present, offering somewhat different accounts of the problems that developed along the way. From ACT's perspective, the company performed many beneficial services for TRU, including creating special features and promotions, designing on-line TRU store value or "gift" card protocol, and setting up "drop shipping." ACT complained about TRU's inventory shortcomings and failure to provide the fulfillment information that ACT needed to ensure TRU "was planning on stocking the products that they had just selected."

Most notably, for present purposes, ACT detailed its "very diligent and best efforts" to catch "selected exclusive products" being offered by third parties. In this regard, ACT sent regular daily notices to TRU about third-party offerings from 1X1 GUI and merchants@, to which TRU could respond by selecting any of the listed offerings to sell itself, in which case, according to Allen, normally ACT would remove the third-party offering within twenty-four to forty-eight hours. At that time, ACT was "implement[ing] electronic UPC matching and blocking technologies" to screen new listings on the Amazon website to protect TRU.

Despite these efforts, ACT acknowledged that even as of the time of trial, third-party vendors were still selling products on the site that were exclusive to TRU. In fact, in a search in November 2004, Rothman identified twenty-nine selected exclusive products being sold by third parties on Amazon's site. In December 2004, she was notified of forty-six selected exclusive products being sold by other vendors. Even so, ACT calculated that since July 2004, sales of "allegedly violative products by third parties . . . are minimal," less than $2000, compared to TRU's estimated sales of $215 million in the same period. According to Rothman, TRU's 2005 sales to date of toys and games were $177 million, an increase of about twenty-five percent over 2004.

TRU disputed both ACT's calculations and its ability to monitor compliance. At trial, Arthur demonstrated the method he used to confirm there were other sellers of exclusive items on the Amazon website, citing as specific examples Target and eToys, whose products were accessed by the "toy and games tab." Arthur offered a list, as of April 2004, of 4000 exclusive products that ACT noted were offered by other vendors on the site. Sullivan, too, described "screen shots" from the Amazon website of exclusive products being offered by other vendors, and referred to a list of forty-six exclusive items being offered by other vendors.

TRU's expert, Mark Hosfield, a CPA and a CMA (Certified Management Accountant) with a Masters degree in finance, accounting and operations management, opined that ACT violated the 3.5% "safe harbor" provision of the SAA and that, in any event, did not "have the means and mechanism in place to monitor" compliance with the contractual requirement. Citing as an example the 2003 sales by third parties of "skateboard products and accessories" of approximately $233,000 based on ACT's own records, Hosfield was ultimately unable to form an analysis for 2003 or 2004 because of incomplete data provided by ACT. More specifically, an ACT memo from November 2004 purported to address only 145 (38 sold by ACT, 107 sold by third parties) of the "over 4,000 products that Toys was complaining about." Moreover, according to Hosfield, data supplied by ACT "changed" between its November 2004 submission and April 2005 supplement.

Hosfield also faulted ACT for the "process that [it] went through to gather information" because it "didn't produce complete data." For instance, in tracking sales information, ACT used "GL" or "general ledger" product classifications, so that a search only of "toy" or "baby" product sales in Amazon's database would not disclose sales of toy or baby products that had, for whatever reason, been given a different GL designation, such as "pets," "sporting goods" or "home and garden." Another shortcoming was that "[a]nything they hadn't . . . classified as exclusive products wouldn't be found by the [computer] searches." Hosfield cited as examples "school supplies," "diaper bags" and "back packs." Consequently, ACT's analysis of products falling within the 3.5% window "missed" approximately 1860 records of sales of the 4000 TRU alleged, comprising "several hundred thousand [dollars] in sales."

Yet another problem was ACT's inability to totally "separate out" TRU sales of TRU products from third-parties' sales of exclusive products. In fact, when TRU ran its own "spider" program through the Amazon website, it located 5900 ASIN product numbers of exclusive products sold on Amazon's site by third parties historically, and yet TRU could locate only 900 of those numbers in the data ACT had provided it directly.

Thus, Hosfield concluded that the "processes that Amazon was going through in an attempt to find product identifying numbers or ASINs for exclusive products being sold by third parties was only catching a small portion of the data." The expert opined that by comparing the UPC prefix information to the "number of products being offered on the web site" as "toys," it was possible that ACT had uncovered perhaps only ten percent of the actual figure.

Based on the information Amazon did provide, however, Hosfield was able to undertake an analysis for the period January 1, 2005 through May 7, 2005, though that, too, was incomplete. As a result, Hosfield was "able to find" about $2 million of exclusive products sold by third parties on the Amazon website during that period. Hosfield also considered sales of exclusive products from "zShops" and "Auctions," and the potential relationship between sponsored link fees and actual revenue from sales. As to the latter, he noted that ACT received about $8.2 million in sponsored link monies in 2004, and calculated sponsored link income between January and May 2005 at $3.37 million. Sponsored link income, however, was not generated by product sales but from customer clicking "the link on [the Amazon] site where [customers] can buy products elsewhere." Although Hosfield could not say whether ACT-sponsored link revenue was related to "toy and baby product sponsored links," based on the insufficient data supplied, he nevertheless opined that "the sponsored links do have an effect . . . on the sales on the web site of the various [TRU] products that they have, and it's a negative effect for the most part." Hosfield thus summed up his conclusion:

The final conclusion is that the data that was produced--and there was data that wasn't produced. We're aware of that [from an email in Amazon's expert's file]. The data that was produced, and knowing what we know about what wasn't produced and the success of the searches that they performed, it's likely[,] in my opinion, that the 3.5 percent limit has been violated. The data to me indicates that.

And I've also determined that they don't have a reliable mechanism in place to monitor or control third-party sales of exclusive products and to monitor their compliance with the 3.5 percent.

To be sure, Robert Brunner, ACT's expert in accounting and financial data analysis, disagreed with Hosfield's analysis, concluding that "sales by Amazon and third parties of exclusive products are significantly below the three-and-a-half percent level." Brunner performed alternative calculations including "Auctions" and "zShop" sales, "ran an alternative scenario with programmatic selling initiatives not included," and did analyses both with and without the 1X 1 GUI factor. For 2003, Brunner found Amazon and third-party sales of selected exclusive products to be .079 percent and sales of exclusive products to be .32 percent of TRU's sales. For 2004, Brunner calculated those respective amounts at .07 percent and 1.18 percent, though he also cited the 2004 sales of exclusive products "without the inclusion of any programmatic selling initiatives" at .47%. For 2005, the numbers were .14 percent for sales of selected exclusive products and 1.68 percent for exclusive products. Without 1X1 GUI, the 2005 sales by Amazon and third parties of exclusive products was .53%.

In all, Brunner ran thirty-two different scenarios. He characterized his calculations as "very conservative." Given the 3.5% safe harbor cap, Brunner estimated that in order for the "dollar value" of exclusive products to have been exceeded, extra sales would have had to amount to $9 million in 2003, $9.4 million in 2004, and $2.3 million as of trial in 2005. Brunner thus concluded that Amazon was "well within the safe harbor" for each of those years regarding selected exclusive products, regardless of whether one analyzed the data using TRU's approach or even a "combined approach" of his and Hosfield's analyses.

Brunner further disagreed with Hosfield's assessment of ACT's capacity to monitor its compliance with the terms of the SAA. He believed ACT "has the information controls in place to ensure its compliance with a three-and-a-half percent provision." As to tracking future compliance, Brunner claimed that ACT "has got an extremely sophisticated and large and complex information system," so that "[a]ll the data to do this analysis is there."

Regardless of whether the 3.5% safe harbor provision had actually been breached, TRU described how it had been harmed as a result of ACT's cumulative actions:

I think there's a variety of ways we've been harmed. One is, there's been a lot of publicity about the fact that we and Amazon are fighting. So any time you have a fight in public, it damages the credibility of both organizations.

Secondly, our [online] store [at Amazon] essentially over the last two years, has disappeared. Contractually it was to be one of the five premiere tabs on the Amazon site, and today it is a category within the Amazon site of which there are 30 or more categories . . . . Our branded designation does not even remotely resemble what you saw on the screen earlier, which is what we were promised as the co-branded impact on the site. And today it's hard to even define what our store is. It certainly isn't a tab any longer, and you can be in our store and look for a product and end up on somebody else's site. So we don't even have inclusiveness of the walls on the site so that . . . a customer . . . may wind up someplace else, when they think they're at Toys "R" Us.

So all the things we were concerned with at the beginning of our relationship, to make sure the brand was protected, to make sure the editing helped the customer find what they wanted, and to buy products that weren't appropriate [sic] or dangerous or from an unscrupulous seller or had a bad experience with, all the things are possible on the site today.

Along these same lines, Peter Sealey, TRU's expert in the assessment, value and use of brands, valued TRU's "brand" at about $1 billion. He explained consumer expectation in finding the Toys "R" Us brand on the Amazon site, and how when "other products . . . show up on the site," "it destroys the integrity of the brand." Among the problems he identified were issues over product safety standards, the ability to make returns, and "consumer confusion." He opined that TRU was clearly harmed by the Amazon website's representation and display of other merchants' goods "under the Toys 'R' Us umbrella." Although TRU was not seeking remedial damages, Sealey did not believe that damage to the Toys "R" Us brand could be fairly compensated, in any event, because the harm was "irreparable and I don't know how it could be ameliorated." He concluded that the relationship between the parties needed to be terminated immediately.

On the other hand, Jim Sheppard, ACT's expert in retail and branding, found no objective measure of any harm to the TRU brand as a result of third-party sales of any exclusive product, and noted that in fact TRU's sales increased about twenty percent "in 2005 after the third parties came on." Nor did he find any objective evidence of consumer confusion or safety concerns over products. Instead, Sheppard believed that both sides had benefited from their business arrangement.

Regardless of whether TRU had suffered actual damage to its brand, TRU was seeking to recover a portion of its base fee attributed to its exclusivity rights, which ACT allegedly violated. Although the SAA did not specify a portion of the base fee attributable to exclusivity, TRU nevertheless maintained that exclusivity "was related to the fixed cost associated with running a toy site." On this score, by comparing ACT's agreement with Target, which did not provide for exclusivity, Hosfield calculated that for the period of February 1, 2003 through September 2005, TRU paid ACT $65,321,429 for its exclusivity rights. To lend further "context [to] the exclusivity value," Hosfield cited the 31.5% drop in video games sales in 2004 after TRU had ceased enjoying exclusive vendor status. In Hosfield's opinion, not even the $65 million would make TRU whole because "what it bargained for was exclusivity," but in failing to receive that benefit TRU also lost "the increased sales, increased profits and those kinds of things" that it had hoped to get in bargaining for exclusivity.

Bruce Budge, a CPA, and ACT's expert in damage quantification, disputed Hosfield's conclusion. Among his cited reasons were the inappropriateness of "comparing" two contracts to "try to isolate" the various "exogenous factors" between the ACT-Target and ACT-TRU deals that could have impacted the amounts paid ACT, and Hosfield's failure to consider the "benefits of exclusivity" that TRU did receive while the parties were performing together.

Following trial, the court issued a lengthy written opinion, finding that ACT had breached the implied covenant of good faith and fair dealing concerning the exclusivity aspects of the SAA and, consequently, that the agreement should be terminated in accordance with its terms. But the court awarded no damages to either party, concluding that TRU had failed to prove the SAA contemplated any specific value for plaintiff's exclusivity rights, and that ACT, on its counterclaim, had failed to prove TRU was in breach.

Thereafter, the court supervised the transitional period of "winding down." Both parties brought enforcement motions during that time regarding ACT's right to withhold sums from online sales in payment of interim service fees, and before the motions were decided, ACT began unilaterally withholding those sums. The court rejected ACT's right to have withheld those monies, and awarded interest on the withheld sums and counsel fees to TRU after finding that ACT had willfully failed to follow the court's earlier order of June 22, 2006.

On appeal, ACT challenges the trial court's decision terminating the SAA, denying its fee requests during the winding down, and sanctioning ACT for violating court orders. ACT also wants money damages on remand. TRU cross-appeals from the order denying its claim for damages.

(I)

In attacking the termination decision, ACT contends that the court "ignored the plain language of the agreement"; erred in allowing extrinsic evidence concerning the scope of exclusivity; failed to support its finding that the implied covenant of good faith "supplanted" the express contract terms; and erred in applying the "inequitable" remedy of termination. Moreover, because the grant of termination was improper, ACT argues that it should be allowed to seek damages. We reject each of these claims.

Admittedly, the court found no technical breach of any specific provision of the SAA. Yet, clearly, the court viewed ACT's conduct in a negative light:

[ACT's] conduct has not been consistent with the drafters' intent in reaching the Agreement. Pursuing third party sellers to expand selection in the face of [TRU's] objection is a breach of the parties' agreement envisioned in the language of the 3.5% Safe Harbor. It is clear from the parties [sic] conduct and the testimony of all of the witnesses that [ACT] would not be in an Agreement that limited the potential for un-restricted assortment and selection of products offered. [ACT] says their intent and understanding of Section 12.1.2 was to have an Agreement that allowed third party sales of exclusive products. The language as drafted whether intentional or inartful gave [ACT] the words to play the game their way.

By contrast, the court found that for TRU, "exclusivity was at the heart of" the SAA, and that TRU consistently dealt with ACT as a "'toy partner.'" But ACT's "build up" away from that goal was "subtle," and involved manipulation of such things as "varying the definition of" Programmatic Selling Initiatives, failing to "track or maintain sales," creating new technology, and "alter[ing] the appearance" of the site. As a result, the court concluded that even if "[e]ach action" by ACT alone was "arguably within the agreement, looked at together [they] demonstrate[] a material shift in the configuration of the partnership." Such a "shift creates a breach at the heart of the Agreement," and neither an injunction nor revision could reconcile the "ambiguity and conduct."

In granting termination of the SAA and directing a "winding up" of the arrangement, the court chose a remedy it believed to be equitable and workable. Otherwise, to force the parties into an ongoing relationship for the remainder of the contract term "would be inviting five years of continued litigation." Moreover, the evidence had demonstrated that ACT lacked the ability to police third-party sales as effectively as needed.

We view the present contentions against this background. As a threshold matter, no one disputes that Delaware law governs the SAA "in all respects." In New Jersey, "[g]enerally, where parties have agreed in a commercial agreement to be governed by the laws of a particular state, our courts will uphold the contractual choice if it is not violative of the public policy of New Jersey." Kalman Floor Co. v. Joseph L. Muscarelle, Inc., 196 N.J. Super. 16, 21 (App. Div. 1984), aff'd, 98 N.J. 266 (1985). Here, no one asserts any public policy violation in the application of Delaware law.

A contract claim under Delaware law requires proof of a contractual obligation, a breach by the defendant, and resulting damage to the plaintiff. H-M Wexford LLC. v. Encorp, Inc., 832 A.2d 129, 140 (Del. Ch. 2003). The court's duty is to enforce the parties' reasonable expectations at the time of the contract. Comrie v. Enterasys Networks, Inc., 837 A.2d 1, 13 (Del. Ch. 2003), aff'd, No. 271, 2004 Del. LEXIS 562 (Del. December 6, 2004).

If the agreement is clear on its face, the court applies the language in the same manner as would a reasonable third party. Ibid. But if terms are ambiguous, the court considers extrinsic evidence such as the parties' statements and actions, their prior dealings, the commercial context, and any custom or usage in the trade. Ibid. In other words, "statements made during the course of the negotiation, courses of prior dealings between the parties, and practices in the relevant trade or industry" are deemed admissible. U.S. West, Inc. v. Time Warner, Inc., No. 14555, 1996 Del. Ch. LEXIS 55, at *12 (Del. Ch. June 6, 1996). The whole contract, rather than isolated provisions, is considered. O'Brien v. Progressive N. Ins. Co., 785 A.2d 281, 287 (Del. 2001).

Moreover, where the court finds a contract ambiguous and the extrinsic evidence does not force a clear conclusion, the court can construe a contract in accord with an objectively reasonable meaning that is in fact held by the first party while the second party knew or had reason to know that the first party held such an understanding. Comerie, supra, 837 A.2d at 13. That rule of construction is known as "'the forthright negotiator principle.'" Ibid. (citation omitted).

In Delaware, as in New Jersey, the construction of a contract is a question of law, and reviewed de novo. Rhone-Poulenc Basic Chems. Co. v. Am. Motorists Ins. Co., 616 A.2d 1192, 1195 (Del. 1992). But Delaware appellate review of a Chancery Court's fact findings is done "with a high level of deference. So long as the Court of Chancery has committed no legal error, its factual findings will not be set aside on appeal unless they are clearly wrong and the doing of justice requires their overturn." Montgomery Cellular Holding Co. v. Dobler, 880 A.2d 206, 219 (Del. 2005).

ACT first faults the court for relying on extrinsic evidence to undercut the plain meaning of the SAA, the "technical" terms of which the court found ACT did not violate. We disagree.

In the first place, in our earlier decision vacating the preliminary injunction, we had already indicated our view that the SAA was arguably ambiguous at least with respect to whether the 3.5% "safe harbor" provision applied to Programmatic Selling Initiatives:

We do not suggest that the parties' intentions with respect to the exemptions established in Section 12.1.2 are sufficiently clear to preclude consideration of extrinsic evidence or to require an ultimate decision in ACT's favor. We simply hold that, viewed in light of the general structure and related provisions of the contract, the provisions permitting and defining "programmatic sales initiatives" are sufficiently favorable to ACT's position to preclude a finding that [TRU's] ultimate success is reasonably probable.

[Toysrus.com, LLC v. Amazon.com Kids, Inc., No. A-0292-04T5 (App. Div. April 12, 2005) (order reversing and vacating preliminary injunction).]

And although we had also suggested that the language of the SAA did not appear to limit the definition of PSIs to existing programs, we made that preliminary observation in connection with the factors necessary to support issuance of injunctive relief, and not with respect to the merits of the ultimate "exclusivity" issue. In any event, the SAA never addressed the 1X1 GUI technology in its definition of "PSI," and even more significantly, never defined "exclusivity" per se.

But even in the absence of any ambiguity, resort to extrinsic evidence was proper here as relevant to the claim of breach of the implied covenant of good faith and fair dealing. Under Delaware law, as in New Jersey, an implied covenant of good faith exists in every contract. Dunlap v. State Farm Fire and Cas. Co., 878 A.2d 434, 441-42 (Del. 2005); Sons of Thunder v. Borden, Inc., 148 N.J. 396, 420 (1997). While courts have said the term "good faith" has no exact definition, "[t]he covenant is 'best understood as a way of implying terms in the agreement,' whether employed to analyze unanticipated developments or to fill gaps in the contract's provisions." Id. at 441 (citations and footnotes omitted). The principle cannot be invoked to countermand the parties' intent or to create wholesale a new legal duty, and it does not apply if the conduct complained of is authorized by the contract terms. Ibid. Nevertheless,

[s]tated in its most general terms, the implied covenant requires "a party in a contractual relationship to refrain from arbitrary or unreasonable conduct which has the effect of preventing the other party to the contract from receiving the fruits" of the bargain. Thus, parties are liable for breaching the covenant when their conduct frustrates the "overarching purpose" of the contract by taking advantage of their position to control implementation of the agreement's terms. This Court has recognized "the occasional necessity" of implying contract terms to ensure the parties' "reasonable expectations" are fulfilled. This quasi-reformation, however, "should be [a] rare and fact-intensive" exercise, governed solely by "issues of compelling fairness." Only when it is clear from the writing that the contracting parties "would have agreed to proscribe the act later complained of . . . had they thought to negotiate with respect to that matter" may a party invoke the covenant's protections.

[Id. at 442 (footnotes and citations omitted).]

Because the goal is always to enforce the parties' reasonable expectations and mutual purpose, the presence of "plain language" in the contract itself, otherwise controlling, should not be viewed in total isolation so as to defeat those expectations, including the implied good faith covenant. See 5 Corbin on Contracts 24.7 (Perillo and Bender rev. ed. 1998) ("Before the meaning of words in a contract can be plain and clear, at least some of the surrounding circumstances must be known; and proof of the circumstances may make plain . . . a meaning that was not apparent . . . in the absence of such proof . . . ."). Thus, Delaware follows the rule that in a given case, "[a] preliminary consideration of extrinsic evidence may be necessary to determine whether this sort of hidden or latent ambiguity exists." U.S. West, supra, 1996 Del. Ch. LEXIS 55, at *31 n.10.

For example, as the Delaware courts themselves have noted, the concept of good faith requires consideration of the "spirit" of the agreement, as well as its expression: "In sum, the implied covenant of good faith '"is the obligation to preserve the spirit of the bargain rather than the letter, the adherence to substance rather than form . . . ."'" Dunlap, supra, 878 A.2d at 444 (citations omitted) (in context of insurance contract, the implied covenant of good faith and fair dealing "requires that the insurer act in a way that honors the insured's reasonable expectations"). The Restatement takes a similar view:

Subterfuges and evasions violate the obligation of good faith in performance even though the actor believes his conduct to be justified. But the obligation goes further: bad faith may be overt or may consist of inaction, and fair dealing may require more than honesty. A complete catalogue of types of bad faith is impossible, but the following types are among those which have been recognized in judicial decisions: evasion of the spirit of the bargain, lack of diligence and slacking off, willful rendering of imperfect performance, abuse of a power to specify terms, and interference with or failure to cooperate in the other party's performance.

[Restatement (Second) of Contracts 205 cmt. d (1981).]

Moreover, from a policy standpoint, whether or not parties must comply with the implied covenant of good faith should not turn on the extent of plain language in an agreement alone, for otherwise it would be possible to circumvent such good faith requirement by carefully crafting the contract around it. In fact, under Delaware law, parties are prohibited from disclaiming the implied covenant of good faith. Del. Code Ann. tit. 6, 1-302(b) (2008).

Nevertheless, ACT argues that the implied covenant of good faith "cannot supplant" the express terms of the SAA, and cites Dave Greytak Enters., Inc. v. Mazda Motors of Am, Inc., 622 A.2d 14, 23 (Del Ch.), aff'd, No. 64, 1992 Del. LEXIS 121 (Del. March 16, 1992) for the proposition that "where the subject at issue is expressly covered by the contract, or where the contract is intentionally silent as to that subject, the implied duty to perform in good faith does not come into play." But in that case, a car dealer sought to change the location of operations when the contract it had with the manufacturer expressly provided that the dealer would only operate at the one location. Therefore, the implied covenant of good faith could not be used to contradict the express agreement of the parties, which would have resulted in a better deal for the dealer than it had struck for itself as to location. Id. at 22-23. Even the court in Dave Greytak acknowledged that the good faith covenant comes into play when it is clear from what the parties otherwise "expressly agreed" that "they would have proscribed the challenged conduct as a breach of the implied covenant of good faith had they thought to negotiate with respect to the matter." Ibid. Such is the case here, where the covenant does not conflict with any "express" terms of the SAA, and permeates the entire contract.

Defendants also assert that because the SAA "expressly addressed the subject of exclusivity," extrinsic evidence was not needed to interpret its meaning, citing Allied Capital Corp. v. GC-Sun Holdings, L.P., 910 A.2d 1020, 1032-36 (Del. Ch. 2006). In that case, the court noted that the implied covenant of good faith analysis was relevant "only . . . when the contract is truly silent with respect to the matter at hand," and only when "the expectations of the parties were so fundamental that it is clear that [the parties] did not feel a need to negotiate about them." Id. at 1032-33 (footnote omitted).

But as the quote from Allied Capital makes clear, ACT's attempt to bar extrinsic evidence on the issue of exclusivity as if it were a single or straight-forward term in the SAA is misplaced. Here, "matter at hand" involves the extent to which TRU's rights of exclusivity can be exercised under different scenarios, so the fact that exclusivity was dealt with in part in the writing should not necessarily preclude consideration of extrinsic evidence that illuminates what the parties intended in other contexts where the meaning is not quite so clear. Ibid. Moreover, the fact remains that it would have been difficult for the parties to have drafted the agreement in such a way as to comprehensively dispose of this issue given all the post-agreement market and technology developments. Cf. id. at 1035 (footnote omitted) ("[C]ourts should be most chary about implying a contractual protection when the contract could easily have been drafted to expressly provide for it").

Here, the court applied extrinsic evidence not to contradict the written agreement, but to better understand it and give force to the parties' expressed purposes. See Eagle Indus. v. DeVilbiss Health Care, Inc., 702 A.2d 1228, 1232-33 (Del. 1997) (in appropriate circumstances, court may consider extrinsic evidence even of an integrated contract); see also 5 Corbin, supra, 24.11 ("[i]nterpretation of a contract," by which the court searches for the parties' intended meaning, "is a process that is distinct from, but is sometimes confused with, application of the 'parol evidence rule'"). Thus, having found that extrinsic evidence was properly admitted, we are further satisfied that the record evidence supports the finding of breach of the implied covenant of good faith.

On this score, the court, applying "covenant-of-good-faith" principles, found that ACT's conduct, although not in breach of any particular provision of the SAA, nevertheless cumulatively violated its "overarching purpose," namely to protect TRU's exclusivity rights. In this regard, the court expressly found that "exclusivity was at the heart of" the parties' contract, and that their mutual intent as to exclusivity "could be seen in [their] establishment of the co-branded store, advertising, and in the accessing of toys, games and baby product catalogs." This very purpose, however, was compromised by a series of actions and omissions subsequently taken by ACT, including changes in web page navigation making TRU's products difficult to access; employing 1X1 GUI technology and Programming Selling Initiatives to increase, to TRU's detriment, third-party sales of toys, games and baby products; failing to adequately track and calculate the extent to which third parties were selling products even ACT agreed were exclusive; and the agreement with Target whereby ACT maintained, again to TRU's detriment, the Target URL for Target to sell the toys, games and baby products Target could not sell on Amazon's site.

These findings are well supported by the evidence. Arthur made it clear that TRU believed it was getting exclusivity for selected toys, games and baby products, and that the safe harbor was a very narrow exception. Not only did Eyler confirm that understanding, but TRU's actions continued to be consistent with the premium it maintained on exclusivity. For example, when proposed by ACT, TRU refused to permit a competitor skateboard vendor on the site, or to otherwise surrender its exclusivity rights, out of concern over issues such as quality, safety and customer satisfaction, matters that were paramount to TRU entering into the deal in the first instance. Even ACT's Britto agreed that the safe harbor language was intended as a "fudge factor" for "creep between categories," rather than a way to circumvent exclusivity. In the same vein, ACT's Miller classified the safe harbor as addressing "incidental sales."

What is more, ACT's Van der Meulen knew early on that Arthur believed TRU had "categorical exclusivity in toys, games and baby products," yet he did not take issue with that understanding until some point after the agreement was signed. Notwithstanding its promises of exclusivity, ACT lacked the technological ability to track third-party sales of exclusive products at the time it entered into the agreement, and even after implementing 1X1 GUI technology in the toy, game and baby categories, it still lacked that ability. In other words, from the day it signed the SAA, ACT knew that it could not scrupulously guard and enforce TRU's exclusivity from third-party sales of exclusive products on the site to the extent that TRU expected. In light of what ACT understood were TRU's goals, its actions amounted to a clear breach of the forthright negotiator principle. Comerie, supra, 837 A.2d at 13.

The fact that the court found insufficient proof that ACT exceeded the 3.5% safe harbor provision is not dispositive of TRU's claim of breach of the implied covenant of good faith. In the first place, we have already explained why the two concepts are not necessarily mutually exclusive. Secondly, the court properly acknowledged TRU's claim that any failure on its part to prove the 3.5% limit had been breached was attributable to the "insufficiency" of the data provided by ACT. Indeed, the court referenced the special master's earlier finding that ACT, though not "actually contemptuous" of discovery orders, had been "less than forthcoming with documentation."

This finding, of course, underscores the larger issue of monitoring third-party sales compliance under the SAA. ACT implemented 1X1 GUI technology in the toys, games and baby categories regardless of knowing it did not have an analogue deterrence system to block third parties from selling exclusive TRU products, and to monitor compliance with the 3.5% limit. Until TRU had requested information on compliance, ACT "did nothing in late 2003, early 2004" to track sales that would qualify within the safe harbor. And even after ACT had "put into place various tracking methods to . . . calculate and retrieve data with regard to" third-party sales, the volume of data, as noted by the court below, made any searches difficult and unreliable. Indeed, Rothman acknowledged that ACT addressed sales of exclusive products by third parties when brought to its attention, but that ACT did not have the technology to stop third parties from selling exclusive products through the 1X1 GUI tool. As inferred by the differing conclusions of experts Hosfield and Brunner, at time of trial ACT still had not implemented a reliable system through which the 3.5% cap could be accurately monitored.

Yet, even with all the system's shortcomings, at the very least the data "showed a steadily escalating volume of sales by third parties through the Amazon site." In fact, given ACT's past inability to track and monitor such data, with a growing base of selection, ACT would be increasingly unable to track and monitor third-party sales affecting exclusive products, and indeed, still lacked that ability at time of trial. Thus, the court reasonably concluded that while "[t]he preponderance of the evidence does not technically establish a breach of the 3.5% safe harbor[,]" "it will not be long before that safe harbor not only is reached but breached by those third party sellers."

Having therefore found sufficient credible evidence of breach of the implied covenant of good faith, we next address the question of remedy. Given the fundamental philosophical and business differences between the parties, as well as ACT's continuing inability to successfully track the sale of exclusive items by third parties on its site, the court concluded that any remedy short of termination "would be inviting five years of continued litigation." We agree.

Section 15.2 of the SAA permits termination "for cause" in the event of a "material breach." Moreover, Delaware law allows for termination of a contract where the breach affects the "substance" of the contract. DeMarie v. Neff, C.A. No. 2077-S, 2005 Del. Ch. LEXIS 5, at *15 (Del. Ch. January 12, 2005). Delaware subscribes to the principle that a material breach gives rise to excused performance, whereas a breach that is only "slight" entitles the non-breaching party to damages. Biolife Solutions, Inc. v. Endocare, Inc., 838 A.2d 268, 278 (Del. Ch. 2003). The factors to consider in determining whether a failure to perform is material include

(a) the extent to which the injured party will be deprived of the benefit which he reasonably expected; (b) the extent to which the injured party can be adequately compensated for the part of that benefit of which he will be deprived; (c) the extent to which the party failing to perform or to offer to perform will suffer forfeiture; (d) the likelihood that the party failing to perform or to offer to perform will cure his failure, taking account of all the circumstances including any reasonable assurances; and (e) the extent to which the behavior of the party failing to perform or to offer to perform comports with standards of good faith and fair dealing.

[Ibid. (footnote omitted; citing Restatement (Second) of Contracts, supra, 241).]

Here, the test of materiality has been met. It is clear that the parties' disagreement over the extent of TRU's exclusivity rights concerned the very core or essence of the SAA. ACT's goal of "increas[ing] the availability of assortment and selection daily" and TRU's goal of maintaining exclusivity ran on a continual collision course and served to undermine the mutual trust and common purpose essential to maintaining the business arrangement struck in the SAA. Moreover, the admitted difficulty of policing the extent of third-party sales of exclusive products made substantial future compliance unlikely. To compound the problem, TRU is unable to calculate with any degree of definiteness the extent of its past losses, or the likely amount of future ones. Contrastingly, from ACT's perspective, it appears from both the number and diversity of its existing and potential vendors that it will be able to continue offering toys, games and baby products on its site without undue hardship. And lastly, as already noted, ACT's behavior has failed to conform to the standards of good faith and fair dealing. Thus, the balance of relevant factors weighs heavily in favor of the remedy of termination.

(II)

ACT also claims the court erred in rejecting its counterclaim for damages based on TRU's alleged failure to keep adequate stock. Specifically, ACT's counterclaim alleged that TRU failed to provide the "minimum selection of top-selling products" and to keep them in stock as provided in the SAA. ACT approximated such damages at $750 million, although at trial, asserted losses of only $4.716 million.

The General Equity judge rejected this claim, finding ACT did not establish that TRU had "failed to exercise efforts to ensure in a commercially reasonable practice they are in compliance with the number of products required by" the SAA. The judge also found that ACT did not give TRU the chance to "cure a defect," and never advised TRU "what percentage of inventory was necessary to be maintained at a location at any given point in time." In short, the judge concluded that, despite the fact that the parties communicated regularly about inventory, ACT never expressed to TRU its expectations with respect to replenishing inventory according to any particular timeframes, and never discussed with or provided TRU with a "concrete plan to change the pattern of behavior." We are satisfied these findings are supported by sufficient credible evidence in the record.

Section 5.1.3 of the SAA set forth the obligations regarding product selection, but in general, it required TRU to provide at least 8000 SKUs for toys and games and 2500 SKUs for baby products. And while, under Section 8.2, TRU had to exercise "commercially reasonable efforts" to maintain a 90% inventory, that term is not defined in the SAA. However, a section of the Delaware Code dealing with secured transactions lists several indicia of "commercially reasonable." Del. Code Ann. tit. 6, 9-627 (2008). For example, a disposition of collateral is commercially reasonable if made "in the usual manner on any recognized market," at the current price in any recognized market, or "otherwise in conformity with reasonable commercial practices among dealers in the type of property that was the subject of the disposition." Id. at (b)(1),(2) and (3). Moreover, "[c]ommercial reasonableness is normally a question of fact," and a court's finding of commercial reasonableness is affirmed if supported by substantial evidence. Addessi v. Wilmington Trust Co., No. 42, 1987 Del. LEXIS 1209, at *5 (Del. August 18, 1987).

While ACT complains of alleged losses from insufficient inventory stocks, it fails to address the "commercially reasonable" predicate, however defined. In fact, the (a) and (b) sub-provisions in Section 8.2 requiring the 90% inventory are expressly drafted "[w]ithout limiting the generality of the foregoing," which explicitly incorporates the "commercially reasonable" standard. Thus, theoretically, any level of inventory shortfall may be excusable so long as "commercially reasonable" efforts are utilized to minimize such shortfall. Therefore, simply demonstrating the amount or extent of the shortfall alone is insufficient to prove a breach. This is particularly so in light of Eyler's explanation of how difficult it was in the toy business to profitably anticipate sales, and the court's penultimate finding "that the volume of e-commerce requires a constantly shifting inventory that must be re-evaluated and re-adjusted based on the pattern of product movement."

There is yet another reason why ACT's counterclaim must fail. At no time did ACT ever notify TRU that it was in breach of the SAA for being out of stock based on product availability. After all, the spirit of the arrangement anticipated that if ACT had such ongoing concerns, they should have been voiced. As it was, the parties were in daily and weekly contact about inventory information, so that notice to TRU, with an opportunity to correct any shortfall, would not have been an obstacle.

Moreover, the SAA provided a detailed array of dispute resolution mechanisms, none of which ACT has demonstrated were satisfied. For example, sections addressed a method for resolving general disputes (Section 22.2), "questions" (Section 22.3), unresolved disputes (Section 22.4), and mandatory mediation (Section 22.6). The parties also were to cooperate with one another throughout the term of the contract (Section 21.2). In fact, with specific reference to inventory, Section 5.4 of the SAA set forth the manner in which ACT was to notify TRU if it objected to including any inventory on the site. Failure to abide by the dispute resolution provisions generally barred a claim, unless the claim could expire by dint of the statute of limitations, and even in that event, the dispute resolution mechanisms had to be pursued simultaneously with any lawsuit (Section 22.2). Finally, Section 15.2 permitted "termination for cause" in the event of material breach, but only after failure to cure and due notice. ACT has failed to show compliance with any of these requirements.

(III)

ACT next contends the court erred in rejecting its claim for payment for services during the "winding down" period, and requiring defendants to pay counsel fees and interest on "disputed" sums. Specifically, the court ruled that TRU had no obligation to continue paying ACT the proportional base fee during the so-called transition period even though ACT continued to provide services throughout this time.

Some background is in order as to the procedural developments in the wake of the court's March 1, 2006 termination decision. As part of that decision, the court ordered the parties to proceed to wind down their joint business affairs in accordance with Section 15.6 of the SAA, by cooperating in good faith for three months of transition, during which ACT was to maintain the co-branded stores and TRU was to deliver "Select Product units" to ACT. Disputes, however, quickly arose over the extent to which ACT would provide certain information and services during wind down. Consequently, on March 31, 2006, following a hearing, the court issued an order

specifying the parties' obligations during wind down, setting an end-date of June 30, 2006, and submitting interim disputes to mediation, while reserving for itself any issues having to do with the actual terms of the order. Significantly, the order did not specify any fees to be paid by TRU during this period.

After filing its appeal and unsuccessfully pursuing stay applications, on April 6, 2006, ACT filed a motion in aid of litigant's rights seeking service payments from TRU during wind down. At an initial hearing, the judge raised jurisdictional concerns in light of ACT's interim appeal, and asked counsel to get direction from this court. On June 1, ACT informed the trial court that it was withdrawing its motion in favor of a lawsuit recently filed in Washington State. On June 8, the trial judge informed the parties she did not consider ACT's motion withdrawn, but had reserved decision pending resolution of its jurisdictional question. Because, in the meantime, ACT began withholding sums otherwise due TRU, on June 14, TRU filed its own motion in aid of litigants' rights, challenging ACT's withholding of monies allegedly due. By order of June 22, 2006, the court determined that it retained jurisdiction to address the base fee issue and then denied ACT's motion "to compel [TRU] to pay fees during the wind down" because the SAA did not provide for payment of any such fees during wind down, and therefore it could not re-write for ACT a better deal than it had made for itself.

Despite this order, disputes continued over customer information and ACT's continued retention of TRU's sales monies because of what ACT claimed was TRU's unjust enrichment. Following another hearing, on August 4, 2006, the court concluded ACT's "self help retention of [TRU] funds is improper, legally and equitably." Referring to its previous ruling that ACT was "not entitled contractually to the base fee or fulfillment fee," the court concluded that ACT's "retention

. . . is in direct disregard of the Court's order and the remedies available pursuant to case law and Court Rule." Consequently, by order of September 14, 2006, the court directed ACT to return to TRU the sum of $13,083,655, which ACT had retained to offset amounts it asserted TRU owed for ACT's interim services. The court also awarded TRU $45,000 in counsel fees, and $104,368 in interest for the sums ACT had wrongfully withheld.

Section 15.4 of the SAA, "Effect of Termination," provided as follows:

Upon termination or expiration of the Term for any reason, all rights and obligations of the Parties under this Agreement will be extinguished, except that (a) all accrued payment obligations hereunder will survive such termination or expiration, and (b) the rights and obligations of the Parties under Sections

. . . . [thirty-five sections are listed by section number including Section 13.1] will survive such termination or expiration.

Section 13.1 was entitled "Amounts Payable to" Amazon, and among the Section 13.1 sub-sections expressly identified in Section 15.4, and therefore having survived termination, were 13.1.4 ("Selected Product Sales Commissions") and 13.1.5 ("CPI Adjustments"). But subsection 13.1.2 ("Co-Branded Stores Base Fees") was not among those sections saved by Section 15.4. Because ACT does not assert that the base fee is an "accrued payment obligation" under Section 15.4(a), and because subsection 13.1.4, the base-fee provision, is not among the sections excluded by Section 15.4(b), we conclude, as did the trial judge, that the base fee charges did not survive termination of the SAA, and therefore, are not due ACT during wind down.

ACT takes issue with this interpretation, however, and urges us to construe Section 15.4 as applying after the wind down period, not during the wind down period. We decline such a reading for several reasons. In the first place, the language of the SAA on this point is clear and explicit. Section 15.4 applies "[u]pon termination or expiration of the term for any reason." Undeniably, the trial court's order "terminated" the SAA. The fact that another provision, Section 15.6, entitled "Post-Termination Transition Period," provides for a three-month transition hardly alters the result. In other words, the "wind down" period envisioned in the SAA begins at termination and runs three months; but, regardless, the sections saved by Section 15.4 begin at termination. Nor does it matter that certain provisions of the SAA - i.e., Section 11.4.4, "Restrictions on Usage" of the co-branded stores customer information - survive even beyond the three-month "wind down." The fact remains that the failure to include Section 13.1.2 in Section 15.4's "survival" clause clearly indicates its exclusion therefrom regardless of the longevity of the shelf life of other provisions expressly incorporated therein.

Second, the fact that Section 15.4 included numerous sections addressing payment and apportionment of sums by the parties further supports the conclusion that Section 13.1.2 was excluded on purpose. For example, among such sections referenced in Section 15.4 were Sections 6.5 ("Vendor Payments"), 8.3 ("Storage Costs and Risk of Loss"), 9.2 ("Selected Product Unit Proceeds"), 9.6 ("Sales Tax Collection"), 13.1.4 ("Selected Product Sales Commissions"), and 13.1.5 ("CPI Adjustments"). Thus, the critical factor was not that the parties might owe each other duties and obligations beyond three months after termination; rather, the key was which duties and obligations they agreed would survive.

Lastly, ACT argues that Section 15.4 is not the exclusive source of duties and obligations during the wind down period since it honored TRU's exclusivity rights during this time, despite the fact that Section 12, governing exclusivity, was not included in Section 15.4. As a result, therefore, of providing a benefit for which it would ordinarily have been compensated through the base fee, ACT contends it is now entitled to a portion of the base fee regardless of the SAA's termination. We reject this argument. ACT's duty to honor TRU's exclusivity is clearly implied in the provision requiring good faith cooperation during the wind down period, as are the duties specified in Sections 9.3 and 9.4, not mentioned in Section 15.4, but having to do with the necessary functions of processing orders and shipping product.

As to the award of counsel fees, ACT contends that the court abused its discretion because ACT had not acted "contumaciously" in continuing its earlier practice of withholding sums due TRU as fees for "services rendered." We discern no abuse of discretion.

Under Rule 1:10-3, the court in its discretion may award counsel fees to a successful litigant in enforcing his or her rights, regardless of whether or not the "act or omission" also amounts to a contempt. On appeal, an order under Rule 1:10-3 is reversed only if an abuse of discretion occurred. See Saltzman v. Saltzman, 290 N.J. Super. 117, 125 (App. Div. 1996).

ACT seems to be arguing here that because the court's June 22, 2006 order denying its motion "to require payment of the base fee by [TRU] for the period encompassing the wind down" did not expressly direct ACT to repay TRU the sums withheld as a pro rata share of the base fee during this time, then it was not in violation of that order. Of course, as soon as the court explicitly so provided in its subsequent order of August 14, 2006, ACT complied. This argument is meritless. The latter order simply reflects the substance of what the court had earlier decided. There is thus no valid distinction between an order denying ACT's motion to withhold sums and an order directing ACT to pay TRU the sums being wrongfully withheld. It goes without saying that sums ACT withheld in the interim were withheld improperly and were required to be returned. Yet, it was not until TRU's subsequent enforcement motion and the court's August 14, 2006 order that ACT tendered the money. That was more than six weeks following the court's denial of ACT's motion. Consequently, ACT's failure to repay such sums in the interim justified the counsel fees ultimately awarded TRU.

On the question of interest, ACT stipulated below that it was paying the $104,368.63 as ordered while preserving its right to appeal. However, ACT now offers no opposition other than in connection with its counsel fee claim. Nor does ACT challenge the court's authority to award interest generally. Regardless, ACT's improper unilateral act in withholding sums otherwise due TRU not only deprived TRU of the sums themselves, but of their use, which was unfair. See Zuhlcke v. Zuhlcke, 136 N.J. Super. 266, 269-70 (Ch. Div. 1975) (citation omitted) (in absence of controlling precedent, "courts of equity are free to decide all questions pertaining to the allowance of interest according to the 'plainest and simplest considerations of justice and fair dealing' in the given case"). The award of interest was therefore proper.

(IV)

ACT also complains of sanctions imposed against it for violating a series of court orders beginning July 26, 2004, which, among other things, preliminarily enjoined ACT from allowing eToys, Inc., or any third party, to sell, on ACT's site, products selected by TRU for sale. As reasons for overturning the sanctions, ACT argues that it had the right under the SAA to allow third-party sales up to at least the 3.5% ceiling and that, in any event, it had not "willfully" violated the court order.

In support of its motion to enforce litigant's rights, TRU proffered evidence that ACT was notified of forty-six "selected" products being sold on ACT's site by third parties as of December 2004. Indeed, ACT acknowledges that at least thirty of the forty-six products so identified by TRU were improperly listed online by third-party vendors. In mitigation, however, ACT pointed to certain steps it had taken to comply in good faith with the court's interim injunctions, including:

technological review of third-party sales; reporting procedures on third-party sales of new products; working "closely and cooperatively" with TRU to settle disputes; and giving TRU advance notice of products third parties intended to list for sale online. Yet, TRU had to take still more steps to force ACT to remove numerous items, including the three particular toys for which the court ultimately sanctioned ACT: toys known as "Stock Market Tycoon," "Blender Set," and "Coffee Maker," all sold by Banana Junction. ACT's Rothman even conceded that Stock Market Tycoon had been "listed after" the court's July 26, 2004 order on November 29, 2004. According to Rothman, that toy and others indicated in her chart of January 19, 2005, were removed from the site on December 21, 2004. Nevertheless, the twenty-nine other products ACT removed on December 21 had all been put online before the court's July 26, 2004 injunction.

But the fact remains that between July and December 2004, ACT had at least twenty-nine--and possibly as many as forty-six or more--of TRU's selected products online being offered by third parties, and that it was only through TRU's vigilance and resort to judicial relief that ACT responded. In fact, notwithstanding all of the reviews and technological checks ACT had in place, it still took defendants at least five days to respond to TRU's December 16, 2004 request to remove products, most of which ACT subsequently agreed needed to be removed. Thus, the mere fact that many of the offending products predated entry of the court's preliminary injunction does not render ACT any more compliant when thereafter ACT refused to remove them from the site. Rather, the court's order required ACT to be vigilant, both with respect to existing products as well as future listings. Tellingly, while reference is made to several protective measures undertaken by ACT, there is no mention of any item that ACT itself removed without TRU's solicitation or demand.

After a hearing on February 28, 2005, the court, crediting TRU's proofs, found that ACT had been notified of its continued breach of the injunction and that TRU had to make additional application to the court to obtain ACT's compliance. The court also found that, under the circumstances, ACT had "either directly violated this Court's Order, or to have [sic] been careless and/or disingenuous in following" the order. The court then imposed sanctions, calculated at three products for six days, or $18,000, consistent with its earlier order warning of a potential $1000 per violation fine. Although the evidence indicated additional violations, the court's sanction was to "make a point to compel compliance, not to punish." We perceive no abuse of discretion in this determination.

The purpose of sanctions is "to afford relief to a litigant who has not received what a Court Order or Judgment entitles that litigant to receive." D'Atria v. D'Atria, 242 N.J. Super. 392, 407 (Ch. Div. 1990) (addressing former Rule 1:10-5, now part of Rule 1:10-3). Such relief is "coercive in nature," designed to effectuate enforcement of a court order. Ibid. Thus, any monetary sanctions should be "related to the litigant's damages," and not be purely punitive. Id. at 408. Clearly, monetary sanctions under Rule 1:10-3 are proper to compel compliance with court orders. Ridley v. Dennison, 298 N.J. Super. 373, 381 (App. Div. 1997).

No one disputes that a monetary sanction imposed pursuant to Rule 1:10-3 is a proper tool to compel compliance with a court order. Neither does ACT contest the actual amount of the sanction nor seek to excuse its non-compliance because we ultimately reversed the underlying preliminary injunction order. Indeed, it is evident from the tenor of the trial court's ruling and the order imposing sanctions that the sanctions did not constitute specific and overt punishment for ACT's violation. On the contrary, sanctions here were designed to coerce compliance with and to facilitate effectuation of the court's temporary injunction. Moreover, there was a proper factual basis for the court's sanction.

By virtue of the temporary injunction, ACT had an affirmative duty to scrupulously police its site and vendors for third-party sales of TRU's products. While it is true ACT did take certain steps in that regard, the number of violations in the period involved, coupled with the time it took to correct such violations once notified, reasonably supports the imposition of sanctions. On this score, good intentions do not permit one to violate a court order. See Warren County Cmty. Coll. v. Warren County Bd. Of Chosen Freeholders, 350 N.J. Super. 489, 513 n.16 (App. Div. 2002) ("A good motive does not excuse a willful violation of an unreversed court order."), aff'd in part and modified in part on other grounds, 176 N.J. 432 (2003). Nor does the fact that the SAA arguably allowed third-party sales up to a 3.5% ceiling excuse ACT's violation of a court order clearly barring any sales of "selected" products, an interim restriction reasonably imposed under the circumstances.

Furthermore, the amount of the sanction was reasonable. Indeed, ACT was not penalized for the full twenty-two days that Stock Market Tycoon was on the site after implementation of the court's order, as even ACT acknowledges. Looked at in that way, moreover, the court's award did not even penalize ACT for the majority of the products improperly left on its site. See Bd. of Educ. of Middletown v. Middletown Twp. Educ. Ass'n, 352 N.J. Super. 501, 509-10 (Ch. Div. 2001) (under Rule 1:10-3, in obtaining compliance with orders, court was not required to "utilize a particular method" of sanctions which should be "carefully tailored and sometimes issued in a particular sequence"). Under the circumstances here, involving potentially thousands of products and hundreds of vendors, and ACT either unable or unwilling to adequately police third-party sales as required, the trial court's sanction did not constitute an abuse of discretion.

(V)

TRU cross appeals from the denial of damages for ACT's breach, claiming that the court erred in requiring evidence of a specific value attributable to the "exclusivity" aspect of the SAA. We agree and reverse and remand on this point.

Specifically, the court found that although the parties had a "shared vision" for "future growth" in what they each perceived as a "unique relationship," "what was not shared in the drafting of this Agreement was a specific or discernible value attributable to that unique relationship." In other words, despite its finding that "exclusivity" was at the core of the SAA, that ACT breached the agreement's implied covenant of good faith with respect to exclusivity, and that TRU was, in fact, damaged generally in terms of the "alteration" in its "unique position, its inability to plan, and inability to strategize" in good faith with ACT, the court, in rejecting TRU's claim for damages, found no evidence that the parties intended to assign any particular value to the contract's exclusivity feature. We disagree.

In Delaware, "recovery may not be had for damages which are speculative or conjectural." Coleman v. Garrison, 349 A.2d 8, 12 (Del. 1975), overruled on other grounds, Garrison v. Med. Ctr. of Delaware, Inc., 571 A.2d 786 (Del. 1989).

There must be some reasonable basis upon which a jury may estimate with a fair degree of certainty the probable loss which plaintiff will sustain in order to enable it to make an intelligent determination of the extent of this loss. The burden is upon the plaintiff to furnish such proof. If he fails in this respect, the jury cannot supply the omission by speculation or conjecture.

[Henne v. Balick, 146 A.2d 394, 396 (Del. 1958) (citations omitted).]

By the same token, the rules for awarding damages for contract breach do not require that the parties attribute some specific value to each of the contract terms. Rather, "[t]he fact that there is some uncertainty as to plaintiff's damage or the fact that the damage is very difficult to measure will not preclude a jury from determining its value." Henne, supra, 146 A.2d at 396. In fact, the

law does not require certainty in the award of damages where a wrong has been proven and injury established. Responsible estimates that lack mathematical certainty are permissible so long as the Court has a basis to make a responsible estimate of damages.

[Bomarko, Inc. v. Int'l Telecharge, Inc., 794 A.2d 1161, 1184 (Del Ch. 1999) (quotations and citation omitted), aff'd, 766 A.2d 437 (Del. 2000).]

Thus, to the extent there is evidence that defendant's actions had prevented TRU from receiving the full extent of its exclusivity bargain, TRU was "entitled to receive the full benefit of the bargain which [it] entered" with defendant. R.M. Williams Co. v. Frabizzio, C.A. No. 90C-MY-10, 1993 Del. Super. LEXIS 55, at *33-*34 (Del. Super. Ct. February 8, 1993).

In R.M. Williams, the court awarded damages to the buyer of real estate because the sellers did not deliver title free of encumbrances as set forth in the agreement. Id. at *36. More specifically, the buyer had gone through with the sale reluctantly when the sellers had purchased an adjacent property after execution of the contract at issue, which purchase encumbered the property sold to the buyer in ways not anticipated in their contract. Id. at *2-*18. In making an award, the court found that the "correct measure" of the plaintiff's damages was "the value of the lots at the time of settlement" had the sellers tendered the property "in conformance with the Agreement[,] less the value of the land as tendered subject to the non-conforming easement." Id. at *36. And the plaintiff was entitled not just to damages for loss in value, but also for "any incidental or consequential damages caused by the breach." Id. at *37. Thus, the key was not that the parties had pre-negotiated or determined the value of the property in case of any subsequent harmful encumbrances, but that it was foreseeable that the sellers' actions could harm the buyer's expectancy. Id. at *38.

And so it was here as well. The parties' ten-year agreement incorporated annual base fees, the amount of which the parties subsequently amended but never eliminated. It is entirely plausible to posit that some portion of those annual amounts was attributable to the exclusivity component of the agreement, otherwise such rights would have been bestowed free of charge and nothing in the record supports that view. Thus, inasmuch as the court found ACT had violated the good faith aspect of the exclusivity promises, TRU was entitled to damages for the difference between the exclusivity bargain it was promised and the compromised exclusivity which it actually received. Otherwise, TRU would be left with a right but no remedy.

Obviously, the calculation of such an amount would be difficult. But, to reiterate, that is no reason to deny TRU damages. Nor is it fatal to plaintiff's claim that the parties never expressly assigned a specific value to "exclusivity." There was, in fact, evidence that some portion of the base fee was intended to include an exclusivity component. Eyler testified that such was the case, and ACT's own representatives even asserted as much during negotiations. In fact, Miller's 2001 internal memo regarding Target supported Eyler's statement in that regard, inasmuch as it linked higher fixed costs to the price for exclusivity. Moreover, when the parties signed the 2002 amendment reducing the annual base fees, TRU also released ACT from claims TRU had regarding the Target-Amazon agreement. Because TRU perceived the Target-Amazon deal as having compromised its exclusivity rights, the fact that plaintiff released its claims in partial return for reduced base fees further suggests the connectiveness between the base fees and exclusivity. And lastly, during the summer 2003 New York City meeting, ACT's representatives, acknowledging at least some linkage, suggested compromising TRU's exclusivity rights in exchange for revenue sharing.

What is more, Hosfield, in part comparing TRU's deal with Target's, estimated TRU's exclusivity losses at more than $65 million. Significantly, the trial court found Hosfield's approach in measuring damages neither improper nor unreliable, but just that the parties never explicitly established an exclusivity value. In this regard, we also disagree with ACT's contention that TRU's own representative, Foster, contradicted Hosfield's position. To be sure, Foster described the base fee, in part, as "a proxy for capital expenditures/opportunity cost, meaning that if Amazon was to use its Georgia building for toys, it couldn't use it for books." Foster added, "[s]o the base fee was meant to be in general a proxy for the fixed costs, either acquired, or that it was unable to use in other parts of its business." Those statements, however, in no way exclude a conclusion that the fixed and opportunity costs included consideration of the exclusivity granted to TRU, much less suggest that Foster himself believed that there was no exclusivity component to the fees charged.

We therefore find error, as a matter of law, in the court's determination that the parties' failure to establish an "exclusivity" value doomed plaintiff's claim for damages to which TRU would otherwise be entitled. We conclude there is sufficient enough evidence for which damages may be assessed, and therefore, reverse and remand for further findings in this regard.

Affirmed in part; reversed and remanded in part. We do not retain jurisdiction.

The base fee was pro-rated for the fourth quarter of 2000, and beginning in 2001 it was $50 million, with annual $5-million escalators through 2004. TRU paid additional amounts in the forms of a "product unit fulfillment fee," basically a price per piece, and a product sales commission, both set in an attached schedule.

"SKU," stock keeping unit, identifies a product for inventory purposes.

Among other things, TRU's Arthur "was assured . . . that Target toys and baby products would not show up on the Amazon platform," and that if a customer "wanted to see [Target's] selection[,] they would actually have to go to the address bar and type in target.com."

On July 28, 2002, TRU and ACT agreed to a second amendment to the SAA, primarily regarding customer service, and a second consent and release.

For example, ACT's Rothman explained that searching "Bruce Springsteen" in Amazon's search bar would first return links to products on Amazon.com relating to Bruce Springsteen, but also return "sponsored links" to third-party websites in some way related to Bruce Springsteen.

UPC, Universal Product Code, is a bar code on products used for inventory control.

"GUI" stands for "graphical user interface," while 1X1 referred to "one at a time, a manual process."

ACT charged a monthly fee to 1X 1 GUI sellers. The 1X1 GUI format was most typically used by "smaller merchants. Larger merchants typically have the information about their products stored in electronic form already . . . ."

When originally launched in 2003, the 1X1 technology was not implemented in the "Toys, Games and Baby Products" categories.

The preliminary injunction was extended by orders of July 26, 2004 and August 12, 2004, the latter also providing counsel fees to TRU. Following the July 26 extension, ACT began a "24-hour-a-day, 7 day-a-week monitoring and review process" for all new product detail pages through the use of 1X1 GUI so it could "immediately remove product listings for toys, games or baby products" TRU had selected for sale. As of August 2, 2004, Amazon was pursuing the "manual review" of the 1X1 GUI detail pages as well. Following the court's August 12 extension, Rothman told TRU that she had "taken steps to prevent third parties from using 1X1 GUI to sell toys, games and baby products in other product categories." She stated that as those issues were being brought to ACT's attention, ACT was addressing them, but at that time ACT did not have the means to control the overlapping.

During pre-trial, the trial court retained retired Justice Gary Stein as discovery master to sort through documentation of third-party sales to address TRU's claim of breach of the 3.5% provision. Just before trial commenced, Justice Stein issued findings on certain discovery issues, including his conclusion that ACT had been "dilatory in the production of documents and had not complied with his previous orders."

"Drop shipping" involved ACT taking a customer's order and sending it electronically to a "drop shipper" named by TRU, so that the drop shipper's warehouse, as opposed to the ACT fulfillment center, would actually fill and ship the order. Beginning in 2003, ACT had set up a system eventually able to accommodate drop shipping for about forty to fifty drop shippers, but as of trial TRU was only using about twelve drop shipping vendors.

Even by TRU's account, 2005 was expected to be a profitable year with its toy and baby business at toysrus.com up roughly 25% compared to 2004.

Brunner faulted Hosfield's analysis for "over including" products. For example, by including "returns" data for TRU but not for Amazon or third-parties, Hosfield had "[s]ignificantly overstate[d] the percentage calculation" of exclusive product sales. As a result, that aspect of Hosfield's calculation was inflated by $150,853, or about seven percent. And Brunner believed that Hosfield's partial-year 2005 analysis was inappropriate because the 3.5% safe harbor was to be calculated on a "calendar year basis"; thus, part-year analyses would fail to take into consideration such factors as "cyclicality or seasonality of both sales and returns," or Amazon's need to "alter its business practices." Brunner also disagreed with Hosfield's assertion that ACT had not disclosed all of the relevant data, concluding that Hosfield failed to appreciate how ACT had supplemented its disclosures.

ACT also filed suit in Washington State for damages and a declaration of its rights following termination of the SAA. On June 26, 2006, the Washington court denied ACT's application for emergent relief and stayed the action in favor of the matter here.

What is more, the last sentence in Section 12.1.2 barred ACT from creating "a virtual 'tab' or other similarly prominent identifier [on the site] for any Third Party [not TRU] for the sale of Exclusive Products or specifically market or promote the sale of any Selected Exclusive Products sold by any Third Party through [the site] pursuant to subparagraph (c) above." By referencing subsection (c) with its 3.5% cap, such language arguably could have been intended to prohibit the kind of sponsored link marketing ACT subsequently pursued. The evidence was clear that customers "clicked" on sponsored links of TRU's competitors, and accessed selected exclusive products. But the extent of such access, or the number of subsequent sales, was unknown.

The same case is listed without opinion in a table at 530 A.2d 1128 (Del. 1987), affirming the Superior Court. A note in the Lexis decision indicates the opinion had not been "released for publication in the permanent law reports." 1987 Del. LEXIS 1219 at *1.

The order was amended on April 3, 2006.

In fact, ACT's finance manager certified that between June 3, 2006 and June 22, 2006, it deducted from the amount otherwise due TRU the "outstanding balance on its quarterly base fee payment," with such deductions eventually totaling $12,308,664.51.

On the following day, June 23, 2006, the Washington court denied defendants emergent relief, and stayed the action pending the matter here.

The initial order was supplemented by one dated August 12, 2004, providing for a $1000 sanction per violation. On December 4, 2004, the court entered another order imposing additional temporary restraints and finally on December 23, 2004, the court replaced its most recent order with one directing that failure to comply "will result in the Court's making a finding of damages and imposing sanctions."

Punitive measures, however, should be undertaken pursuant to the procedure specified by Rule 1:10-2. Ibid.

Although recognizing that the merits of the issue regarding the 3.5% safe harbor could not be resolved pending trial, the court had consistently taken the position that in the interim ACT was not to permit third parties to sell any selected products.

(continued)

(continued)

14

A-3391-06T2

March 24, 2009