This is an appeal from a grant of summary judgment in favor of the defendant, Stuart Hale Company (Stuart Hale) in an action brought by American Home Products Subsidiary Holding Company (American) alleging federal trademark infringement, unfair competition, and dilution of trademark. Additionally, American sought a preliminary injunction, which the district court denied as moot. In the present appeal, American renews these claims, and Stuart Hale cross-appeals the denial of sanctions under Rule 37(a) of the Federal Rules of Civil Procedure. For the following reasons, the judgment of the district court is affirmed in part, reversed in part, vacated in part, and remanded for proceedings in accordance with this opinion.

* BACKGROUND

Appellant American owns the federally registered trademark to "PAM," a nonstick cooking spray sold in nine-ounce aerosol cans in the consumer retail market. American has owned the PAM trademark since 1958 and, with a fifty to seventy-five percent market share, is currently the product leader in the cooking spray industry. Pl.Ex.D. PAM has no registered trademark rights to, and has never marketed, a "lite" or "light" version of its product.

In 1953, five years before American registered the mark "PAM," Stuart Hale had introduced the name "Pan-Lite" to market a nonstick cooking oil. Prior to 1988, however, Stuart Hale had only used the name "Pan-Lite" to market its cooking oil in the wholesale market by the use of fifty-five gallon drums sold directly to bakers; it was not sold in the retail grocery market. Indeed, from 1953 to 1988, Stuart Hale sold the Pan-Lite product to only six customers. In April 1988, Stuart Hale applied for federal trademark registration for the name Pan-Lite to market its nonstick cooking oil as an aerosol spray. This product was a different formulation than the product it had marketed on a wholesale basis earlier. R.64, Plaintiff's Local Rule 12(m) Statement at 2. Prior to the 1988 introduction of the retail aerosol version of Pan-Lite, Stuart Hale had never sought trademark protection for the name Pan-Lite. It is undisputed that the PAM and Pan-Lite products are now in direct competition.

Shortly after Stuart Hale applied for protection of the Pan-Lite name, American initiated the present action as a five-count complaint against Stuart Hale. American alleged federal trademark infringement, federal and common law unfair competition, dilution, and deceptive trade practices.1 American argues that consumers are likely to mistake Pan-Lite for a light version of the PAM product. It notes that there are already three distinct versions of PAM in the market: original, butter-flavored, and olive oil. Additionally, American stresses the proliferation of market products that contain less fat and fewer calories, and submits that consumers have begun to expect light versions of their favorite brand name products. Specifically, American bases its factual claim on the following product similarities between PAM and Pan-Lite: (1) the brand names sound similar; (2) both are sold in predominantly red and yellow aerosol spray cans; (3) both picture fried eggs on their cans; and (4) both cans depict, in "bullet" lettering, claims that the product has "no cholesterol," is "all natural," and is "sodium free."

American retained Dr. Hans Zeisel, now deceased, to survey consumers in order to determine the likelihood that consumers would confuse Pan-Lite with PAM. Dr. Zeisel supervised two surveys in which only primary shoppers over eighteen years of age were questioned. In the first survey, participants were shown photographs of six retail grocery items commonly sold in supermarkets. These items included Pan-Lite but not PAM. After seeing the photographs, participants were asked to recall the brand names they had seen. Forty-five percent of those recalling a brand name for the cooking spray identified Pan-Lite as PAM.2

The second survey was performed as a control because PAM is the product leader in the market and might be identified by consumers for that reason. The control survey showed participants the same six grocery items, but substituted Mazola or Wesson-Lite (also direct competitors of PAM) for Pan-Lite. Seven percent of the participants in the second survey identified the cooking spray as PAM. Dr. Zeisel then subtracted the seven percent figure from the forty-five percent figure to control for the likelihood of misidentification that could reasonably be considered a result of PAM's status as market leader. The result was a net misidentification of thirty-eight percent attributable to the similarities between Pan-Lite and PAM. Dr. Zeisel also submitted by affidavit that it was his expert opinion that the similarities between the names PAM and Pan-Lite caused consumers to misidentify the Pan-Lite product as a PAM brand cooking spray. R.106, Attach. at 4, Zeisel Aff. In addition to the survey evidence compiled by Dr. Zeisel, American submitted to the district court an article in the April 19, 1990 Chicago Tribune stating:

In a red and yellow aerosol can, Pan Lite looks and acts like PAM, a close cousin that shares grocery space.

R.102, Attach. A. American submits that this article further evidences a likelihood of consumer confusion between the two products.

In response, Stuart Hale argues that American offered no material, probative evidence that could establish a likelihood of consumer confusion between PAM and Pan-Lite. In fact, submits Stuart Hale, the real motivation behind American's lawsuit is that it is considering marketing plans for a "light" version of PAM, for which it has yet to seek trademark protection. Appellee's Br. at 2-3. Additionally, Stuart Hale filed a Rule 37 motion for attorney fees on the ground that American had deliberately frustrated the discovery process and should be sanctioned.

American's motion for preliminary injunction and Stuart Hale's motion for summary judgment were referred to a United States magistrate judge for an evidentiary hearing and a Report and Recommendation. See 28 U.S.C. Sec. 636(b)(1)(B) (1988). Subsequently, the district court reviewed the magistrate's report de novo and found it to be "thorough, accurate, and the decision proper." R.122, Order at 1. Consequently, the district court adopted verbatim the recommendations of the magistrate judge and entered summary judgment for Stuart Hale.3 The district court applied the seven-factor "likelihood of confusion" test now common in this circuit.4 With respect to several of these factors, there was no dispute between the parties. As the district court noted, the products are indisputably direct competitors and are similarly advertised in the same market. The strength of the PAM mark was also uncontradicted. However, the court specifically rejected American's theories on four of the likelihood of confusion factors.

The district court first considered the degree of similarity between the marks in appearance and suggestion. With respect to suggestion, the court determined that PAM was a completely fanciful mark, while Pan-Lite has a functional suggestion.5 Similarly, the court found that "[l]ittle homogeneity exists in the products' visual appearance." Rep. and Rec. at 7. The court also rejected American's argument that the fact that Pan-Lite's dominant term "pan" sounds and appears like "PAM," coupled with consumer expectation concerning "light" products, causes a likelihood of source confusion. The court emphasized the fact that American does not currently market a light version of PAM and has no trademark rights to a PAM "light" or "lite." Accordingly, concluded the district court, American's proposition is "nothing more than mere conjecture." Id. at 8.

The district court also rejected American's argument that consumers exercise little care in distinguishing between low-cost grocery items, concluding:[W]ith the prodigious advertising campaigns conducted by corporate America, consumers can not help but to be well aware of new and old products.... [G]iven that all grocery items are relatively inexpensive and sold in a multitude of brands, at the very least, some amount of careful discrimination must be exercised by consumers, if only in an effort to find their favorite brands.

Id. at 9.

On the issue of actual confusion, the district court found that American had submitted no probative evidence. Specifically, the court rejected American's consumer survey proffer that thirty-eight percent of the persons who were shown pictures of Pan-Lite believed it to be PAM, on the grounds that it was "not fairly prepared and, therefore, does not accurately gauge actual confusion among consumers." Id. at 12. Specifically, the court stated:

[T]he survey's major failing is that it does not gather data of actual confusion based upon a showing of PAM and Pan-Lite as found in the marketplace--shelved together. A proper and fair analysis would more accurately simulate consumer perception by showing the products one next to the other, or amongst a group of cooking spray brands. American's survey does not achieve proper marketplace reality.

Id. at 13.

The district court also rejected American's contention that a triable issue of fact existed as to whether Stuart Hale intended to palm-off Pan-Lite as related to the successful PAM product. The court noted that the name Pan-Lite had been in existence five years prior to American's PAM mark and that the Pan-Lite name describes the product's function. The court disposed of the purported significance of the recent entry of Pan-Lite into the retail market as "tenuous support for the proposition that Stuart Hale intended to palm-off." Id. at 14.

The district court also summarily rejected American's state law claims because of the "absence of similarity between the marks." Id. at 16. Finally, without separate analysis of the merits, the district court denied American's motion for a preliminary injunction on the ground that it was moot.

We review a district court's grant of summary judgment de novo, taking all inferences in favor of the nonmovant. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S. Ct. 2505, 2510, 91 L. Ed. 2d 202 (1986); Anderson v. Stauffer Chem. Co., 965 F.2d 397, 400 (7th Cir.1992). Accordingly, if there is any genuine material issue of fact, we must remand. See Fed.R.Civ.P. 56.

The linchpin of both common law and federal statutory trademark infringement claims is whether consumers in the relevant market confuse the alleged infringer's mark with the complainant's mark. J. THOMAS MCCARTHY, TRADEMARKS AND UNFAIR COMPETITION Sec. 23:1, at 42-44 (2d ed. 1984). We evaluate the likelihood of confusion in trademark infringement claims by analysis of the following seven factors:

(1) similarity between the marks in appearance and suggestion;

(2) similarity of the products;

(3) area and manner of concurrent use;

(4) degree of care likely to be exercised by consumers;

(5) strength of complainant's mark;

(6) actual confusion; and,

(7) intent of defendant "to palm-off his product as that of another."

Forum Corp. of North Am. v. Forum Ltd., 903 F.2d 434, 439 (7th Cir.1990).6

In addressing the district court's application of this test, American initially asserts that we must reverse unless Stuart Hale, as moving party in this motion for summary judgment, established that there were no material issues of fact on any of the seven elements of the "likelihood of confusion" test. Indeed, submits American, Stuart Hale only attempted to adduce evidence on four of these factors and therefore failed to meet its burden of proof. We cannot accept this argument. Stuart Hale actually conceded that the three factors it did not address favored American in the balance. There was no genuine dispute as to those factors. The district court's ruling simply found that those three factors alone would not be sufficient for a reasonable factfinder to hold Stuart Hale liable. This conclusion was not legal error. None of the seven confusion factors alone is dispositive in a likelihood of confusion analysis:

[T]he plaintiff need not prove each and every factor in order to prevail. However, the converse is also true; neither is it required that the defendant refute each and every factor. The weight and totality of the most important factors in each case will ultimately be determinative of the likelihood of confusion, not whether the majority of the factors tilt the scale in favor of one side or the other.

Schwinn Bicycle Co. v. Ross Bicycles, Inc., 870 F.2d 1176, 1187 (7th Cir.1989).7

Alternatively, American argues that the district court impermissibly weighed and balanced the evidence, instead of taking all evidence presented in the light most favorable to American as the nonmovant. Summary judgment, submits American, should have been precluded because material issues of fact are in dispute. This submission requires extensive attention.

"We have stated a number of times that the trial court's ultimate conclusion on the likelihood of confusion is a finding of fact." Forum Corp., 903 F.2d at 438. Accordingly, a motion for summary judgment in trademark infringement cases must be approached with great caution. Here, at the summary judgment phase, the parties disputed four of the seven factors in the likelihood of confusion analysis. Specifically, the parties disputed (1) the degree of care likely to be used by consumers, (2) the similarity between the marks in appearance and suggestion, (3) the probability of actual confusion, and (4) the intent of Stuart Hale to palm-off Pan-Lite as a PAM product. The only question now before us is whether the differences between the parties with respect to any of these factors create a genuine material issue of triable fact as to the likelihood of confusion. We shall examine these factors in turn.

1. Degree of care likely to be used by consumers

American argues that the district court improperly concluded, on this record, that consumers exercise a high degree of care in purchasing cooking spray. We believe that the district court's conclusion in this regard must be evaluated with some skepticism, especially at this summary judgment stage. Indeed, American attempted to introduce evidence in the form of surveys to suggest that consumers exercise very little care in purchasing low-cost grocery items.8 This submission is relevant to the likelihood of confusion because presumably consumers take less time purchasing low-cost items, and haste increases the possibility of confusion. This court also has observed that

where the product involved [is] a low value item, the risk of confusion is greater "because purchasers are unlikely to complain when dissatisfied, which would bring to light confusion; but rather they are likely simply to avoid all products produced by the company which they believe produced the product which caused the trouble."

Maxim's Ltd. v. Badonsky, 772 F.2d 388, 393 (7th Cir.1985) (quoting Union Carbide Corp. v. Ever-Ready, Inc., 531 F.2d 366, 383 (7th Cir.), cert. denied, 429 U.S. 830, 97 S. Ct. 91, 50 L. Ed. 2d 94 (1976)).

It may be possible for Stuart Hale to establish that the "prodigious advertising campaigns" of corporate America9 have resulted in consumers' exercising great care in the selection of their favorite brand of cooking spray. However, at this point, American has properly raised a genuine dispute by producing survey evidence tending to show that such care was not used. Accordingly, it was error for the district court to determine, as a matter of law, that consumers exercise some amount of care in choosing a cooking spray.

2. Similarity between the marks in appearance and suggestion

The district court concluded that "the products differ in visual appearance" and that, with respect to suggestion, the "marks are clearly different." Rep. and Rec. at 5. Specifically, the court noted: "In contrast to PAM, the Pan-Lite product has its name cursively written in white; the can's body is virtually all red against a thin yellow plaid outline,10 its cap is white; and a caricature of a chef, also in white, adorns the cap." Id. at 7. American invites our attention to several similarities in the two products' appearance, including the common yellow and red color scheme, the similar depiction of eggs in a frying pan, and similar representations that the products have "no cholesterol" and are "all natural" and "sodium free." Additionally, American emphasizes the phonic and visual similarity between the mark "PAM" and the dominant word "Pan" in the Pan-Lite name. Stuart Hale counters that the terms "Pan" and "Lite" and the depiction of a frying pan containing eggs are descriptive of the function of cooking sprays generally. Further, contends Stuart Hale, American has overplayed the similarity between the mark "PAM" and the prefix "Pan" because marks should be considered as a unit and not dissected into individual elements.

Similar arguments were presented to this court in Scandia Down Corp. v. Euroquilt, Inc., 772 F.2d 1423 (7th Cir.1985), cert. denied, 475 U.S. 1147, 106 S. Ct. 1801, 90 L. Ed. 2d 346 (1986), a case involving two competitors in the bedding industry. On appeal, this court held that the factfinder was entitled to conclude that, although the elements of the mark were descriptive if taken in isolation, as a unit the mark was not merely descriptive:

The words and pictures surely are "descriptive" if taken one at a time. But that is not the right way to take them. The eye sees the combination of words, typeface, and goose as a unit.... The district court therefore must consider each mark as a unit.

Id. at 1430. See also Computer Care v. Serv. Sys. Enters., Inc., 982 F.2d 1063, 1069 (7th Cir.1992) ("Where the plaintiff's overall trade dress is distinctive, the fact that it uses descriptive (or generic) elements does not render it nonprotectable."); Roulo v. Russ Berrie & Co., 886 F.2d 931, 936 (7th Cir.1989), cert. denied, 493 U.S. 1075, 110 S. Ct. 1124, 107 L. Ed. 2d 1030 (1990) (same). In the present case, we do not believe that an argument that the combination of the Pan-Lite name and packaging is confusingly similar to that of PAM should have been foreclosed as a matter of law.

American also submits that the name "Pan-Lite" creates a likelihood of source confusion because it may be considered a "light" version of the PAM product line. The district court rejected this argument on the ground that it was "mere conjecture." Rep. and Rec. at 8. We believe that this conclusion is premature. American's claim is not that it has some form of preemptive right to the potential trademark "PAM-Lite." Rather, it simply argues that there is a likelihood of confusion as to the source of the product. "The 'keystone' of trademark infringement is 'likelihood of confusion' as to source, affiliation, connection or sponsorship of goods or services among the relevant class of customers and potential customers." Sands, Taylor & Wood Co. v. Quaker Oats Co., 978 F.2d 947, 957 (7th Cir.1992) (quoting 2 MCCARTHY Sec. 23:1 at 42-43, 46-47); see also International Kennel Club, 846 F.2d at 1089 (fact that "parties' products are the kind the public might very well attribute to the same source ... provides additional evidence of a likelihood of confusion"). American has raised a genuine issue of triable fact as to whether consumers are likely to mistakenly believe that there is an association or sponsorship connection between PAM and Pan-Lite.

American argues that the district court erred in excluding the Zeisel survey results, its principal evidence of actual confusion. The court stated that the survey's "major failing is that it does not gather data of actual confusion." Rep. and Rec. at 13. Yet, this circuit has repeatedly stated that actual confusion need not be demonstrated; the test is likelihood of confusion and actual confusion is only one element of the test. Computer Care, 982 F.2d at 1070; McGraw-Edison Co. v. Walt Disney Prods., 787 F.2d 1163, 1172-73 (7th Cir.1986) (district court erred in rejecting survey evidence using photo comparisons to demonstrate thirty-five percent consumer confusion, method of testing went to weight not admissibility of evidence). Indeed, recently, in Sands, 978 F.2d at 960, this court held that the district court erred in rejecting survey results similar to those in the present case. In Sands, the survey purported to establish that twenty-four percent of those shown the label from the defendant's product believed the product to be the plaintiff's. The court noted that the survey was "precisely the sort of study ... held to be the correct methodology for assessing consumer confusion." Id. In McGraw-Edison, we also reversed the district court's rejection of survey evidence offered to show actual confusion. Specifically, we concluded that "the district court's concern regarding the manner of presentation to the interviewee goes to weight to be accorded to the survey results rather than providing a reason to ignore the survey altogether." 787 F.2d at 1171-73. Similarly, in the present case, we believe that the district court was premature in rejecting the survey evidence proffered by American. While there will be occasions when the proffered survey is so flawed as to be completely unhelpful to the trier of fact and therefore inadmissible, see Delavan, 975 F.2d at 394, such situations will be rare and the present case is not among them. We believe that any shortcomings in the survey results go to the proper weight of the survey and should be evaluated by the trier of fact.11

American also asserts that the district court committed reversible error by concluding that the marks were dissimilar based upon its own side-by-side comparison. Although side-by-side comparison may be marginally helpful at times, see Computer Care, 982 F.2d at 1070; Roulo, 886 F.2d at 937, it must be remembered always that the ultimate focus is on "the purchasing public's state of mind when confronted by somewhat similar trademarks singly presented." International Kennel Club, 846 F.2d at 1088 (quoting James Burrough Ltd. v. Sign of Beefeater, Inc., 540 F.2d 266, 275 (7th Cir.1976)).

4. Stuart Hale's intent to palm-off the PAM mark

American challenges the district court's conclusion that Stuart Hale did not intend to palm-off Pan-Lite as PAM because it had been using the Pan-Lite name five years prior to American's registration of the PAM mark. Stuart Hale's president also asserted that he did not intend to benefit from PAM's market popularity. However, American correctly notes that the district court cannot weigh credibility issues at the summary judgment stage. Wilson v. Williams, 997 F.2d 348, 350 (7th Cir.1993) (quoting Liberty Lobby, 477 U.S. at 255, 106 S.Ct. at 2513-14). We believe that American did raise a genuine issue of triable fact.

American argues that the district court erred in dismissing its state law unfair competition and dilution claims based upon analysis of the federal trademark infringement claim. The state unfair competition claim is analyzed under the likelihood of confusion standard and thus mirrors our infringement analysis. See McGraw-Edison, 787 F.2d 1163.12 Accordingly, we shall not repeat that analysis here.

Under Illinois law, dilution is not dependent upon the likelihood of confusion but on a separate likelihood of dilution inquiry. See Ringling Bros. Barnum & Bailey Combined Shows, Inc. v. Celozzi-Ettelson Chevrolet, Inc., 855 F.2d 480, 482 (7th Cir.1988) (The Illinois Anti-Dilution Act "grants protection to trademarks beyond that provided by the classic 'likelihood of confusion' test under the Lanham Act.... The additional protection prevents the gradual whittling away of trademarks' distinctiveness through use by third parties on non-confusing, non-competing products.") (citations omitted). Nonetheless, Stuart Hale correctly asserts that, under modern state precedent,13 the protection of the Illinois Anti-Dilution statute is not available to competitors under Illinois case law. See EZ Loader Boat Trailers, Inc. v. Cox Trailers, Inc., 746 F.2d 375, 380 (7th Cir.1984); see also Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 786 F. Supp. 182, 214 (E.D.N.Y.1992) (same; applying Illinois law). Accordingly, the district court was correct in dismissing this claim.

American argues that, if we remand for any reason, we should also hold that the district court erred in not entering a preliminary injunction. In reply, Stuart Hale emphasizes that Pan-Lite controls less than one percent of the relevant market, while PAM dominates the market. Accordingly, submits Stuart-Hale, the balance of hardships14 that guides a court's preliminary injunction determination favors denying the motion. American responds that a market leader also has an interest in protecting its reputation as a quality product. Consumers may mistake other brands, which may not be held to the same quality standards, for American products and subsequently avoid purchasing American's entire product line. This course of conduct could result regardless of the alleged infringer's market share.

The district court never reached the merits of this issue; the court dismissed the preliminary injunction motion as moot after granting summary judgment for Stuart Hale. Consequently, we cannot now review it on appeal; instead, we must remand for the district court to make a fact-based determination. See Schwinn, 870 F.2d at 1176.

D. Stuart Hale's Cross-Appeal for Attorney Fees

On cross-appeal, Stuart Hale argues that the district court erred in failing to award it attorney fees because American "systematically obstructed [its] discovery to conceal [American's] real motive behind [the] action, ... to market a new product, PAM-Lite, into a market [without] Pan-Lite." Appellee's Br. at 2-3.

Generally, we review a district court's refusal to impose Rule 37 sanctions only for an abuse of discretion. National Hockey League v. Metropolitan Hockey Club, 427 U.S. 639, 642, 96 S. Ct. 2778, 2780, 49 L. Ed. 2d 747 (1976); Shine v. Owens-Illinois, Inc., 979 F.2d 93, 96 (7th Cir.1992). However, the denial of sanctions with no explanation may constitute an abuse of discretion. Wojan v. General Motors Corp., 851 F.2d 969, 974 (7th Cir.1988); Szabo Food Serv., Inc. v. Canteen Corp., 823 F.2d 1073, 1084 (7th Cir.1987) (If the reasons for denying a motion for sanctions are clear on the record "the judge need not belabor the obvious[;] ... [Otherwise, the parties] and this court are entitled to explanations."). The district court noted:

[T]he federal rules place the burden on the litigants to vigorously pursue appropriate motions to ensure that justice is served. In the instant case, if Stuart-Hale was dissatisfied with the actions of any party or the court, it had at its disposal an arsenal of weapons, including Rule 37, and Local Rule 12(k), to alter the course of the litigation. Stuart-Hale, the proponent of the motion, failed to do so effectively and will not now be heard to complain of the magistrate judge's failure to rule.

R.122, Order at 2 (emphasis added). A failure to pursue a Rule 37 motion more "vigorously" might well justify a district court's decision not to award sanctions despite the merits of the motion and, as we have noted, such a decision is reviewed in this court under a deferential standard of review. However, deferential review does not mean no review at all. Mars Steel Corp. v. Continental Bank N.A., 880 F.2d 928, 936 (7th Cir.1989) (en banc) (citing In re Ronco, Inc., 838 F.2d 212, 217-18 (7th Cir.1988)). Here, the district court's explanation is so conclusory that we cannot review the substance of its decision. Accordingly, its judgment shall be vacated and, on remand, the parties may again address the matter and obtain a more plenary explanation from the district court.

We must address one more issue. Because this case comes to us on summary judgment, a remand to the district court does not automatically invoke the provisions of Circuit Rule 36 that require reassignment for further proceedings in the district court. Here, however, we believe that it would be most compatible with the rationale underlying Rule 36 to require that the case be reassigned. As our discussion indicates, we believe that the district court did not limit itself to a determination of whether there was a genuine issue of fact requiring trial, but rather proceeded to evaluate the evidence as if the proceeding were a trial on the merits. Under the circumstances presented by this case, further proceedings ought to be conducted before a judicial officer not involved in the earlier proceedings.

Finally, we emphasize that our analysis has been limited to whether it was appropriate to grant summary judgment for the appellees. We stress emphatically that we intimate no opinion with respect to the merits of the dispute.

For the foregoing reasons, the judgment of the district court is affirmed with respect to the dismissal of the count under the Illinois Anti-Dilution Statute. It is vacated insofar as it denies the appellee's motion for sanctions; it is reversed and remanded insofar as it granted summary judgment to the appellee. Circuit Rule 36 shall apply. The appellant may recover its costs in the appeal; no costs are awarded on the cross-appeal.

IT IS SO ORDERED.