OPINION OF THE COURT
SCIRICA, Circuit Judge.This dispute arises from an advertising war between major competing producers of over-the-counter antacid remedies. Plaintiff/appellant Johnson & Johnson-Merck Consumer Pharmaceuticals Company (“Johnson-Merck”) alleges that television commercials by defendanVappellee Rhone-Poulenc Rorer Pharmaceuticals, Inc. (“Rorer”), about ■ its product Extra Strength Maalox Plus (“ESMP”), are misleading advertising in violation of section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a) (1988). Johnson-Merck produces and markets a competing product, Mylanta Double-Strength (“Mylanta II”), and claims its sales suffered as a result of the misleading advertising.
Key to the controversy is Rorer’s description of ESMP as “the strongest antacid there is” in television commercials promoting ESMP tablets and liquid. Johnson-Merck contends the claim misleads consumers into thinking that ESMP is superior as a treatment for acid indigestion.
The district court held a five-day evidentia-ry hearing on Johnson-Merck’s motion for a preliminary injunction. Because the record developed in the hearing was so extensive, the court, with the agreement of the parties, converted its Memorandum and Order on the motion into one on the merits of the case and entered final judgment.1 The district court found Johnson-Merck failed to meet its burden of proof with respect to claims that the *127advertising was false or misleading and with respect to damages.2
Johnson-Merck appeals, contending the advertisements should have been enjoined if it showed that Rorer either intended to or in fact did mislead consumers. Johnson-Merck claims it proved both at the hearing. Specifically, Johnson-Merck maintains the trial court erred in failing to evaluate its evidence of Rorer’s intent to mislead and in misevalu-ating its key evidence that the commercials did mislead consumers.
The district court had jurisdiction based on a federal question arising under the Lanham Act. 28 U.S.C. § 1331. We have jurisdiction under 28 U.S.C. § 1291. We review the district court’s conclusions of law in a plenary fashion, its findings of fact under a clearly erroneous standard, and its decision to grant or deny an injunction for abuse of discretion. Castrol Inc. v. Pennzoil Co., 987 F.2d 939, 943 (3d Cir.1993); Sandoz Pharmaceuticals Corp. v. Richardson-Vicks, Inc., 902 F.2d 222, 226 (3d Cir.1990).
I.
A. Acid Neutralizing Capacity and Effectiveness of Antacids
Over-the-counter (“OTC”) antacids are designed to provide symptomatic relief from acid indigestion by neutralizing excess acid in the stomach or esophagus. However, the test on which advertisers rely for claims of strength does not measure the operation of antacids in the human body (in vivo testing); rather, it measures their operation in glass beakers of acid in the laboratory (in vitro testing).
In the 1970’s, the Food and Drug Administration adopted a laboratory test for acid neutralizing capacity (“ANC”) which defines OTC antacids for purposes of labeling and measures their strength. The test determines how much acid is neutralized in a glass beaker by a single dose of antacid over a fifteen-minute period. The ANC test does not measure an antacid’s capacity to neutralize acid in vivo, which depends on other factors in addition to ANC,3 nor does it provide information on an antacid’s ability to relieve the symptoms of acid indigestion. Such information would be more useful to consumers than the results of laboratory studies, but the FDA concluded human testing would be too laborious. ANC information is provided to physicians, but because the FDA thought the consuming public might mistakenly take the ANC rating as a measure of effectiveness, it decided to exclude such information from product labels. It concluded that this “technical information on the consumer label could result more in confusion than enlightenment and could result in unwarranted consumer reliance solely upon this information as an indication of relative effectiveness.” 38 Fed.Reg. 31264 (11/12/73). Ironically, the very information the FDA feared would mislead consumers has become the basis of advertising claims and is called upon to support such claims when they are challenged as false or misleading.
When liquid ESMP and Mylanta II are compared in vitro, ESMP has a higher ANC rating; when they are compared in tablet form, Mylanta IPs ANC rating is higher. However, in vivo tests, conducted by both parties to demonstrate the relative effectiveness of the two brands of antacid, show no advantage for either brand in the human body. The district court found that “[n]ot one of the in vivo studies, performed by either side, demonstrated any statistically or clinically significant difference in the ability of Mylanta II or ESMP to relieve symptoms of acid indigestion.”
*128The district court’s conclusion that neither in vitro nor in vivo tests provides any basis for either antacid manufacturer to claim that its product is more effective at relieving symptoms is well supported, and neither party contests it. The dispute is over whether Rorer caused consumers to think that ESMP was more effective at relieving symptoms by making misleading claims of superior relief in its commercials, thereby violating the Lan-ham Act.
B. Commercial Claims
Rorer began airing television commercials in July 1989, claiming that ESMP is “the strongest antacid there is,” or “the strongest antacid I can [could] buy.” 4 Some commercials included the statements, “The doctor told me it was strongest,” or ‘Tour doctor will tell you they’re strongest.” Others contained weaker medical claims, such as “My doctor recommended it,” or “Its doctor recommended formula neutralizes more, more than any leading antacid.” The district court focused on two television commercials, first aired in 1991, the “Firefighter” commercial, which advertised ESMP liquid, and the “Minty Tablets” commercial, which advertised ESMP tablets. Although there were minor differences in these and other commercials aired at different times, the versions described below are typical of the advertisements objected to by Johnson-Merck.
The Firefighter commercial which aired on NBC and CBS in the summer of 1991 showed a fireman saying, “To survive in today’s world, you have to follow certain basics. You have to work, and you have to eat. And both play havoc with your stomach. That’s why I take Maalox.” The screen then showed liquid ESMP, with the voice saying, “Maalox is the strongest antacid there is.” Superimposed on the screen was the text, “Dose for Dose Based on Lab Tests.” The screen showed the fireman again, who went on to say, “And who knows more about stomach problems than Maalox? You know it’s a funny thing. I started taking Maalox because the doctor told me it was strongest. I keep on taking it because my stomach tells me it’s fast.” The screen then showed Maalox again, with the text superimposed, “The doctor told me it was strongest. My stomach tells me it’s fast.”
The Minty Tablets commercial which aired on ABC and NBC in August 1991 started with pictures of a bottle of ESMP liquid. The superimposed text stated, “Dose for Dose Based on Lab Tests of Acid Neutralization,” while an announcer said, “Extra Strength Maalox Plus, the strongest antacid there is, ... ” The picture changed to show a bottle of ESMP tablets as the announcer continued, “... now comes in tablets. New Extra Strength Maalox Plus Tablets.” The screen showed Turns and Rolaids tablets as the announcer went on to say, “Maalox strength is stronger than Turns; Maalox strength is stronger than Rolaids.” A woman’s voice said, “My doctor told me it’s the strongest,” as the screen showed ESMP tablets. The woman then appeared, saying, “My stomach tells me it’s fast.” The remaining pictures were of bottles of ESMP tablets as the announcer’s voice said, “Extra Strength Maalox Plus, now in Minty Tablets.” Both the announcer and a superimposed text said, “Your doctor will tell you they’re strongest. Your stomach will tell you they’re fast.”
Neither commercial was being aired in the form described at the time of the hearing. The Firefighter commercial had been dropped and the Minty Tablets commercial was running in modified form without the claim, “Your doctor will tell you they’re strongest.” At the hearing, counsel for Rorer represented that it intended to continue its advertising campaign for ESMP using the claim “the strongest antacid there is,” but agreed it would not use the claim that a doctor told, or would tell, someone that ESMP was strongest without further substantiation. Therefore, the district court did *129not address the “doctor” claim.5 Slip op. at 9.
Johnson-Merck maintained Rorer’s claim that ESMP products are strongest was false or misleading or both. It argued that Rorer’s claim in its Minty Tablets advertisement that ESMP tablets are the strongest was literally false because Mylanta II tablets have a higher ANC rating than ESMP tablets. The district court, however, found that the Minty Tablets commercial explicitly compared ESMP tablets not with Mylanta II tablets but with the leading antacid tablets, Turns and Rolaids, each of whose ANC was lower than that of ESMP tablets. The court concluded that Mylanta II tablets were not within the scope of comparison in the commercial and therefore the advertisement was not literally false. Slip op. at 24-25. Johnson-Merck did not pursue this issue on appeal.
Johnson-Merck also asserted, and continues to assert on appeal, that the commercials are misleading. It contends that it met its burden of proof in two ways: first, it presented evidence that Rorer intended to mislead the public into thinking “strongest” meant strongest in providing relief; and second, it commissioned consumer surveys that it claims show Rorer succeeded in misleading the public. It argues the district court made an error of law in failing to consider its evidence on intent to deceive or mislead and clearly erred in finding it had not met its burden of proof.
II.
A. The Lanham Act
Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a) (1988), creates a cause of action for false or misleading description or representation of a product.6 To prevail on its claim of unfair competition under Section 43(a), we have said a plaintiff must prove by a preponderance of the evidence:
(a)(1) Any person who, on or in connection with any goods or services ..., uses in commerce any ... false or misleading description of fact, or false or misleading representation of fact, which—
(B) in commercial advertising or promotion, misrepresents the nature, characteristics, [or] qualities ... of his or her or another person's goods, services or commercial activities, shall be liable in a civil action by any person who believes that he or she is or is likely to be damaged by such act.
1) that the defendant has made false or misleading statements as to his own product [or another’s]; 2) that there is actual deception or at least a tendency to deceive a substantial portion of the intended audience; 3) that the deception is material in that it is likely to influence purchasing decisions; 4) that the advertised goods travelled in interstate commerce; and 5) that there is a likelihood of injury to the plaintiff in terms of declining sales, loss of good will, etc.
U.S. Healthcare, Inc. v. Blue Cross of Greater Phila., 898 F.2d 914, 922-23 (3d Cir.1990), cert. denied, 498 U.S. 816, 111 S.Ct. 58, 112 L.Ed.2d 33 (1990) (quoting Max Daetwyler Corp. v. Input Graphics, Inc., 545 F.Supp. 165, 171 (E.D.Pa.1982)). A plaintiff must prove that the claim is false or misleading, not merely that it is unsubstantiated. San-doz, 902 F.2d at 228.
If a plaintiff proves a challenged claim is literally false, a court may grant relief without considering whether the buying public was misled. Sandoz Pharmaceuticals Corp. v. Richardson-Vicks, Inc., 735 F.Supp. 597, 600 (D.Del.1989), aff'd, 902 F.2d 222 (3d Cir.1990). A determination of literal falsity rests on an analysis of the message in context. Id. If a plaintiff does not prove the claim to be literally false, he must prove that it is deceptive or misleading, which depends on the message that is conveyed to consumers. U.S. Healthcare, 898 F.2d at 922. Public reaction is the measure of a commercial’s impact. Sandoz, 902 F.2d at 228-29; American Brands, Inc. v. R.J. Reynolds Tobacco Co., 413 F.Supp. 1352, 1357 (S.D.N.Y.1976). As the district court noted, the success of the claim usually turns on the persuasiveness of *130a consumer survey. Slip op. at 23 (citing Johnson & Johnson-Merck Consumer Pharmaceuticals Co. v. Smithkline Beecham Corp., No. 91 Civ. 0960, 1991 WL 206312 (S.D.N.Y., Oct. 1, 1991), aff'd 960 F.2d 294 (2d Cir.1992); Coca-Cola Co. v. Tropicana Products, Inc., 690 F.2d 312, 317 (2d Cir.1982)). The factfinder must determine whether the public was, in fact, misled. Drawing a distinction between the FTC plaintiff and the Lanham Act plaintiff, this court held in Sandoz that the FTC plaintiff could rely on its own determination of deeep-tiveness, but the Lanham Act plaintiff
bears the burden of proving actual deception by a preponderance of the evidence. Hence, it cannot obtain relief by arguing how consumers could react; it must show how consumers actually do react.
We hold that it is not sufficient for a Lanham Act plaintiff to show only that the defendant’s advertising claims of its own drug’s effectiveness are inadequately substantiated under FDA guidelines; the plaintiff must also show that the claims are literally false or misleading to the public.
902 F.2d at 228-29.
B. Intent to Deceive or Mislead
Johnson-Merck contends its evidence shows Rorer intended to convey a message of superior relief by making the claim that ESMP is the strongest antacid. It cites a decision by the United States Court of Appeals for the Second Circuit, Johnson & Johnson-Merck Consumer Pharmaceuticals Co. v. Smithkline Beecham Corp., 960 F.2d 294 (2d Cir.1992), which adopts a presumption for Lanham Act cases shifting the burden of proof to the defendant when a plaintiff has presented evidence of defendant’s intent to deceive the public. Johnson-Merck argues that it has presented evidence of Rorer’s intent to deceive or mislead and asks us to adopt the burden-shifting presumption and require the district court to consider its evidence on intent.
As evidence of intent to mislead, Johnson-Merck introduced Rorer’s strategy document, for ESMP.7 Under the heading “Consumer Promise,” the document states: “To convince the target audience that Extra Strength Maalox Plus Liquid provides superior relief of acid indigestion/stomach problems.” Next, under the heading “Reason Why,” the document states, “Contains a special, doctor recommended (Extra Strength) formula that provides the highest level of acid neutralizing capacity (ANC) of any leading brand.”
In addition, Johnson-Merck cited the testimony of Patricia Bowman, Senior Brand Manager for Maalox liquids from October 1988 through August 1990. On direct examination, Johnson-Merck asked her whether another ESMP commercial called “Date,” first aired in July 1989, was intended to convince consumers that ESMP provided superior relief:
Q So “Date” was the implementation of a copy strategy of superior relief due to Maalox’s highest acid neutralizing capacity? •
A Yes.
Q So you were intending to communicate superior relief by saying that ESMP was the strongest antacid there is?
A No, that’s not true. We didn’t communicate superior relief. We communicated strongest.
Q Just so I understand, again in the copy strategy, you say that you want to convince the target audience — I take it that’s consumers — that ESMP provides superior relief. How were you going to go about convincing them:
A By communicating that the product is the strongest.
Ms. Bowman also testified about a memorandum to her from Ann Bernheim, an advertising executive responsible for the ESMP account, listing the “Communication Objectives” for commercials introducing new flavors of ESMP liquid. The memo stated:
*131Please confirm your agreement with the following Communication Objective for Ranch [commercial for new flavors of ESMP] to be submitted to DRI [Diagnostic Research, Inc.]:
To communicate the news of the introduction of New Cherry Creme and New Mint Creme Extra Strength Maalox Plus in a manner that establishes them as:
• Great tasting new flavors
• Providing superior relief
• Is the strongest antacid
• Is doctor recommended
When questioned on the meaning of the memo, Ms. Bowman insisted that “it was never an intent to communicate superior relief,” and explained the references to “superi- or relief’ and “strongest antacid” in this way:
[S]uperior relief meant the halo effect that I’m trying to get from my advertising to mate my consumers choose — choose my brand to differentiate it. And strongest antacid literally means strongest antacid.
In responding to the meaning of “halo effect,” Ms. Bowman said:
I think — halo effect — what I would say our advertising is trying to do is to give your brand — you know, make your brand different in the consumer’s mind. Give consumers a reason to choose you over another product.
Although Ms. Bowman tried to distinguish “intent to communicate” and attempted “halo effect,” we see this as a distinction without a difference. One can intend to create a misleading halo effect.
While acknowledging that “there was conflicting testimony regarding the original ESMP copy strategy,” Rorer insists that it changed its copy strategy in 1990 to a promise of “fast relief’ or “instant relief’ supported by the “strongest antacid” claim, and that thereafter, the commercials, including the Firefighter and Minty Tablet commercials “were indisputably the product of a strategy that did not intend to promise consumers superior relief.” That is “indisputably” true only in the sense that the copy strategy documents changed so that the “consumer promise” was no longer stated to be superior relief. From this record, we are not convinced that Rorer no longer intended or expected to communicate the halo effect of superior relief to which Ms. Bowman referred.
Ann Berriheim testified that Rorer lawyers had made clear to the advertising agency that it could not claim superior relief. Therefore, Rorer knew when it began running the “strongest antacid” commercials that it could not properly make such an explicit claim in the commercials. The intent to communicate, or to promise superior relief appeared only in its internal documents, and that was purged in 1990.
For us to accept Rorer’s contention that after 1990 its strategy changed so that the claim of “strongest antacid” was no longer intended to promise the consumer superior or strongest relief, but instead was intended to promise only “fast” or “instant” relief, would take us into an Alice-in-Wonderland world of meaning. We cannot see how the change of stated consumer promise in copy strategy documents and the addition of a claim of fast relief altered the halo effect of superior relief that attached to the phrase “strongest antacid there is” and that Ms. Bowman’s testimony clearly shows was intended. Rorer’s rewriting of its intent is therefore unconvincing. But even if we agree that Johnson-Merck has presented evidence of Rorer’s intent to mislead, this alone is insufficient to qualify for the burden shifting presumption set forth in Smithkline Beecham.
As the United States Court of Appeals for the Second Circuit stated:
[W]e have held that “where a plaintiff adequately demonstrates that a defendant has intentionally set out to deceive the public,” and the defendant’s “deliberate conduct” in this regard is of an “egregious nature,” a presumption arises “that consumers are, in fact, being deceived.”
Smithkline Beecham, 960 F.2d at 298-99 (quoting Resource Developers, Inc. v. Statue of Liberty-Ellis Island Foundation, Inc., 926 *132F.2d 134, 140 (2d Cir.1991)).8 Johnson-Merck contends that the district court’s failure to follow Smithkline Beecham and to consider Rorer’s intent was error. But from this record, Johnson-Merck’s evidence only-shows an intent to mislead, or to create a misleading halo effect, of a kind that is unfortunately common in the antacid industry. Significantly, we see no evidence of “deliberate conduct” of an “egregious nature,” the second part of the Smithkline Beecham test.
Painting a picture of the industry, Rorer asserts that “every major antacid brand historically has made comparative strength claims based on ANC test data,” including Mylanta, Turns and Rolaids. Before Johnson-Merck acquired Mylanta in early 1990, Stuart, Mylanta’s former owner, advertised: “Try Double Strength Mylanta-II with nearly double the acid neutralizing power of any leading brand. That’s real relief, with the double strength difference.” It also advertised, “Introducing Mylanta-II Roll Packs. For double strength antacid relief wherever you are.” Indeed, the names of both products here refer to strength, and suggest superior relief: Extra-Strength Maalox Plus and Mylanta Double-Strength.
Both the earlier Mylanta advertising and the Maalox advertising have tried to exploit the consumer confusion the FDA feared between the results of ANC tests and symptom relief. The advertisements tout the ANC strength, promise symptom relief, and invite consumers to make the connection. Although there is evidence of intent to mislead, it is of a kind regrettably pervasive throughout the antacid industry and does not reach the egregious proportions that would warrant a presumption shifting the burden of proof. Therefore, we need not decide whether to adopt the Smithkline Beecham presumption in case of clear and egregious conduct. In this case, the district court’s failure to consider Johnson-Merck’s evidence as to Rorer’s intent was not error.
C. Consumer Surveys
Johnson-Merck concedes that liquid ESMP has a higher ANC rating than liquid Mylanta II; that is, it is stronger in laboratory tests. The claim that ESMP is the “strongest antacid there is” is therefore not literally false.9 The question is whether it is misleading. Specifically, did consumers understand the commercials to mean that ESMP was better at providing relief?
Johnson-Merck offered five consumer surveys, which it asserts show that consumers bought the claims of superior relief. The district court focused on two surveys, the Firefighter and Minty Tablets commercials, which it considered most relevant to the case.10 On appeal, Johnson-Merck focuses on the Minty Tablets survey which, it contends, presents its strongest evidence of misleading advertising.
In each survey, researchers travelled to busy shopping malls and searched out men and women who had, within the previous six months, used an antacid product in either tablet or liquid form.11 In each survey, a *133researcher showed the respondent the commercial and asked a number of questions. The purpose of the surveys was to discover whether consumers understood various claims made in each commercial, particularly the phrase “strongest antacid there is,” to mean that the antacid provided superior relief.
In initial surveys, consumers were asked to identify the product advertised and then asked what, if any, message they had gotten from the commercial: for example, “What ideas did the advertiser try to get across about [the product] in the commercial?” This open-ended question was followed by one or more “probes,” such as, “What other ideas did they try to get across?” or, simply, “What else?” “Anything else?” On occasion, interviewers followed the initial question with numerous probes.
In some surveys, the researcher showed consumers the commercial a second time and asked the meaning of a particular statement, such as: “In the commercial you just saw, they said Maalox is the strongest antacid there is. What does that mean to you?” The researcher then asked, “According to the commercial you just saw, what do you think Maalox is the strongest at doing?” Slip op. at 16-17.
Not surprisingly, the parties’ experts differed in their interpretations of the survey results. Rorer’s expert, Dr. Seymour Lieberman, criticized Exhibits 43A and 48B, which contained verbatim consumer responses in a survey of a Minty Tablets commercial.12 Dr. Lieberman discounted a number of responses, either because he considered the questions invalid or because he interpreted the responses as not referring to comparative relief. He disqualified responses to leading questions, such as “What do you think Maalox is strongest at doing?” and responses that he concluded were elicited only after repeated probes.13 He also criticized questions on specific messages not preceded by filter questions: for example, the question asking what the statement ESMP was “strongest” meant was not preceded by a question filtering out those subjects who had not said that the commercial claimed ESMP was strongest.
In the end, Dr. Lieberman considered valid only answers to open-ended questions, such as “what ideas did the advertiser try to get across,” which were not prompted by repeated probes. And even for those questions he considered valid, he found only a few responses linked comparative strength with comparative relief. He did not characterize responses such as ESMP works “well,” or “fast,” or is a “good” product as comparative responses. Nor did he equate speed with relief and he thereby discounted statements like “works faster.” He also distinguished and discounted the response “works best” from those linking relief with strength. His assessment of the Minty Tablets survey, the survey Johnson-Merck contends comprises its key evidence, was that only 7.5% of the responses linked strength to superior relief *134in response to open-ended questions. This is an insufficient number to show that, under the Lanham Act, the advertising tends to deceive or mislead “a substantial portion of the intended audience.” U.S. Healthcare, 898 F.2d at 922.14
Johnson-Merck’s expert, Dr. Susan S. McDonald, criticized Dr. Lieberman’s restrictive analysis of the surveys. Dr. McDonald found that well over half the responses in the Minty Tablets survey reported a comparative relief claim. She reached this result by interpreting responses of “faster,” “stronger,” and “strongest,” as well as responses of “better” and “best,” to refer to greater relief. Dr. McDonald did not criticize or discredit responses to repeated probes. Interpreting the consumer responses in this way, Dr. McDonald found in the Minty Tablets survey that over 110 out of 200 consumers, or over 60%, responded in terms of comparative relief. Nonetheless, the district court found Dr. Lieberman’s evaluation of the Minty Tablets surveys represented in Exhibits 43A and 43B more persuasive than Dr. McDonald’s, although it did not accept all of his criticisms of them.
On appeal, Johnson-Merck claims the district court clearly erred in evaluating the survey evidence, which Johnson-Merck contends shows a majority of consumers understood the “strongest antacid” claim meant “strongest” relief, and that those misled far exceeded the “substantial portion of the intended audience” we have said is required under the Lanham Act. U.S. Healthcare, 898 F.2d at 922. Johnson-Merck contends the district court failed to discuss explicitly what it regards as its key piece of evidence, Exhibit 43A, consumers’ verbatim responses in a survey of a Minty Tablets commercial. Johnson-Merck claims Exhibit 43A’s initial question, “What ideas did the advertiser try to get across about Maalox tablets in the commercial?” was open-ended, meeting all Dr. Lieberman’s criticisms, and the answers prove that consumers were misled by the “strongest antacid there is” claim. Johnson-Merck also alleges that Dr. Lieberman mis-characterized the answers to the first open-ended question as using primarily non-comparative words (“gives relief,” “is strong,” “it works”) when in fact a substantial number of responses used comparative or superlative words (“works better,” “works faster,” “it’s the best”).
As the district court noted, “The probative value of a consumer survey is a highly fact-specific determination and a court may place such weight on survey evidence as it deems appropriate.” Weight Watchers Int’l, Inc. v. Stouffer Corp., 744 F.Supp. 1259, 1272 (S.D.N.Y.1990), citing Universal City Studios, Inc. v. Nintendo Co., 746 F.2d 112, 118 (2d Cir.1984); Coca-Cola Co. v. Tropicana Products, Inc., 690 F.2d 812, 317 (2d Cir.1982). A survey is not credible if it relies on leading questions which are “inherently suggestive and invite guessing by those who did not get any clear message at all.” Slip op. at 25, citing American Home Products Corp. v. Johnson & Johnson, 654 F.Supp. 568, 581 (S.D.N.Y.1987).
The court agreed with Dr. Lieberman that a well-designed consumer survey first asks “communication” questions to see what messages the viewer got and to “filter” or separate those viewers who received certain messages from those who did not. In the next step, the survey asks those who received a particular message, such as the message that Maalox is the “strongest antacid” there is, “comprehension” questions to determine what the viewers thought the message meant. The court noted that distortions can occur at either stage.
*135The district court found that “[t]he consumer surveys conducted for [Johnson-Merek] were not objective because they asked very leading questions.”15 Slip op. at 26. It did, however, credit the non-leading questions. Specifically, the court found, with respect to the Firefighter and Minty Tablets surveys:
They suffered from repetitive and leading questions and no filter mechanism. The first “communication” questions were most probative. For the second series of questions, the survey failed to filter out those respondents who recorded a message of superiority on first viewing. By flagging Rorer’s “strongest claim” (“in the commercial you just saw, they said Maalox is the strongest antacid there is. What does that mean to you?”), the “comprehension” question colored the answers. The next questions, (3a) in the “Minty Tablets” survey and (4) in the “Firefighter” survey, were even more suggestive. By asking what ESMP was strongest at doing, they called for the answer “relief.”
The technique of punctuating open-ended questions with repeated probes is questionable but did not discredit the responses to questions (2a), (2b), and (2c). This series reliably recorded consumer impressions of the commercials. According to the tabulations, consumers thought the commercial’s main idea was that ESMP works well but not that it works better than other antacids. The responses to questions (2a) and (2b) alone, the only non-leading questions showed that in the “Firefighter” survey, only 8% (NBC/CBS) and 14% (CBS) and in the “Minty Tablets” survey, only 23% responded in the category of “good/better/best.” Of these, only a subset answered with the [comparative or] superlative[ ] “better” or “best” that linked strength with superior relief.
Although the second part of the surveys were flawed, Dr. Lieberman’s criticisms were exaggerated. His classification scheme mistakenly excluded responses like “works best” from the category of comparative statements linking relief with strength. However, even if some of the responses Dr. Lieberman left out were counted, the number of consumers misled by the commercials was not substantial.
Slip op. at 27-29.
Although not mentioning Johnson-Merck’s Exhibit 43A by name, the district court discussed extensively the survey in that exhibit and Dr. Lieberman’s evaluation of the questions and responses in the survey. The court also relied in part on the testimony of the survey’s designer, A. Spencer Bruno, but not on that of Dr. McDonald, in its finding that an insufficient number of the responses linked strength with comparative relief to show that a substantial percentage of consumers were misled.16 Mr. Bruno classified the responses to questions 2a and 2b in the Minty Tablets survey and found that 23% of the responses were in the category of “good/better/best,” but that only a subset of the 23%, those that responded “better” or “best,” linked strength with superior relief. Mr. Bruno did not break down his figures into comparative and non-comparative subsets. Dr. Lieberman performed the breakdown and concluded that only 7.5% of consumers who were shown the Minty Tablets commercial linked strength to symptomatic relief. The court noted that, in reaching that conclusion, Dr. Lieberman included only the most explicit statements that were not prodded by repeated probes.
With respect to the survey evidence as a whole, the district court accepted most, but not all, of Dr. Lieberman’s criticisms. It agreed with him that only a few of the survey questions were probative, and that decision was within the sound discretion of the district court. In evaluating responses to the *136non-leading questions, the court stated that Dr. Lieberman’s criticisms were exaggerated and that he “mistakenly excluded responses like ‘works best’ from the category of comparative statements linking relief with strength.” Slip op. at 28-29. However, it concluded that even if additional responses were included in the totals to correct for his exaggerated criticisms, the number of consumers misled by Rorer’s “strongest antacid” claim was not substantial. We cannot say the district court’s finding was clearly erroneous.
Dr. Lieberman’s professional opinion was that the consumer surveys were seriously flawed and had not proved that ESMP’s commercials misled consumers. Interestingly, his personal opinion was that consumers understood the claim in the Maalox commercials, “strongest antacid there is,” to refer to the product’s ability to give them relief. Even if we were inclined to agree with his personal opinion, Dr. Lieberman’s personal opinion is not the legal standard by which the courts must determine whether consumers were misled. Sandoz, 902 F.2d at 228-29. If Rorer’s commercials do mislead consumers into thinking that ESMP promises superior relief, well-designed consumer surveys will show that they do. Absent such evidence, neither the district court nor this court can conclude that consumers were misled.
The district court’s finding that the consumer surveys in this case did not prove that a substantial number of consumers were misled was not clear error. Because the district court did not err in its finding that plaintiff had not sustained its burden of proving that defendant’s advertising was deceptive or misleading, we need not reach the issues of damages or the denial of the injunction.
III. CONCLUSION
For the foregoing reasons, the district court’s findings of fact are not clearly erroneous, it was correct in rejecting Johnson-Merck’s legal claims, and it did not abuse its discretion in failing to issue an injunction. The judgment of the district court will be affirmed.
. The parties supplemented the record with two additional exhibits for purposes of the final judgment.
. The court noted that, because it found plaintiff failed to meet its burden with respect to liability, it did not need to reach damages, but it went on to consider whether Johnson-Merck showed it had suffered a decline in sales as a result of the challenged advertising, as it claimed. The court found Johnson-Merck failed to meet its burden in proving that its sales had suffered as a result of Rorer’s commercials. Johnson & Johnson-Merck Consumer Pharmaceuticals Company v. Rhone-Poulenc Rorer Pharmaceuticals Company, No. 91-7099, slip op. at 29, 1993 WL 21239 (E.D.Pa. Jan. 29, 1993).
. Some of the other factors affecting the efficacy of antacid remedies are rate of gastric emptying, rate of secretion of acid by the stomach, and degree of mixing of antacid with gastric contents.
. The district court chronicled Johnson-Merck’s unsuccessful attempts to convince the major television networks to discontinue Rorer's advertisements containing the "strongest antacid there is” claim. The networks permitted the commercials to continue so long as they contained superimposed texts stating: "based on lab tests” (CBS/NBC) or "based on lab tests of acid neutralization” (ABC). Slip op. at 10-11, n. 3.
. Rorer's voluntary relinquishment of the "doctor” claim did not extend to weaker "doctor" statements such as "My doctor recommended it,” contained in other commercials.
. Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a) (1988), as amended, provides in part:
. The document is a one-page summary giving tile target audience, major competition, tone and message of the advertisements, and other infer-mation.
.The burden-shifting presumption was not, however, applied, in either of those cases, in one because the plaintiff had not proved intent to deceive and in the other because the commercial for which deceptive intent was claimed was not at issue. In Smithkline Beecham, Johnson-Merck sued the manufacturer of a competing product. Turns, for falsely implying that aluminum in other antacids was harmful. Plaintiff claimed that by advertising Turns as "aluminum free,” the manufacturer suggested that antacids that were not aluminum free were harmful by playing into a popular misconception about ingested aluminum. Plaintiff introduced evidence of deceptive intent (controverted by defense witnesses), but the commercial in question had been withdrawn from the air and was not the subject of the suit. Therefore, the court declined to apply the presumption. 960 F.2d at 299.
. The district court found the statement was not literally false with respect to tablet form ESMP either, although Mylanta II tablets have a higher ANC rating than ESMP tablets. As noted supra, the advertisement claiming that ESMP tablets are the “strongest antacid there is” compared ESMP explicitly with Turns and Rolaids; the district court thereby excluded tablets from the scope of the comparison.
. In addition to the Firefighter and Minty Tablet surveys, there were two surveys of a "Date” commercial, and one of an “Up to Here” commercial, that the district court only mentioned.
. Respondents who already used antacids were chosen because the primary objective of most advertising in this market is to convince current users to change brands, not to persuade nonusers to try over-the-counter antacid remedies.
. The Minty Tablets survey asked the following questions, some of which were followed by additional probes:
2a. What ideas did the advertiser try to get across about Maalox tablets in the commercial?
2b. What other ideas did they try to get across?
2c. What did they show in this commercial about Maalox tablets? What else? Anything else?
3a. In the commercial you just saw, they said Maalox tablets are the strongest. What does that mean to you?
3b. What is the commercial saying that Maalox tablets are the strongest at doing?
5 a. The commercial you just saw contained the statement, "Your doctor will tell you they're the strongest,” referring to Maalox tablets. What does that statement mean to you? Verbatim responses appear at Plaintiff's Exhibits 43A and 43B.
. Exhibits 43A and 43B, which were reviewed by both plaintiffs' and defendants’ experts, were typescripts of responses and not the questionnaires themselves. The probes were not reproduced in the typescript, except for the ones that were listed as survey questions, i.e., 2b and 2c, so it was difficult for anyone evaluating answers to other questions to separate responses that were given spontaneously from those that were responses to repeated questions. Where it appeared that a respondent was elaborating upon an answer in a string of responses, Dr. Lieberman credited only the first response. He testified that the use of repeated probes yielded artificially slanted answers. Johnson-Merck’s expert, Dr. Susan McDonald, considered valid all the responses to repeated probes.
. With regard to what constitutes a substantial or significant number of consumers who are misled, the court cited to several cases which suggest that 20% would be sufficient. Stiffel Co. v. Westwood Lighting Group, 658 F.Supp. 1103, 1114 (D.N.J.1987) (potential that between 22% and 57% of consumers will be misled is not insubstantial), R.J. Reynolds Tobacco Co. v. Loew's Theatres, Inc., 511 F.Supp. 867, 876 (S.D.N.Y.1980) ("deception rate” of between 20% and 33% sufficient to warrant preliminary injunctive relief); McNeilab Inc. v. American Home Products Corp., 501 F.Supp. 517, 527 (S.D.N.Y.1980) (23% not insubstantial number of consumers). While Johnson-Merck disputed the court’s evaluation of the consumer surveys, it did not challenge the court's assumption of what constitutes a substantial or significant number of misled consumers for purposes of a Lanham Act violation. Therefore, we do not review that assumption.
. Examples of the questions to which the district court refers appear supra at note 12. In that list, questions 2a and 2b are neither leading nor do they contain repeated probes. Question 2c is not leading, but it contains three probes in addition to the probe in question 2b. Questions 3 a, 3b and 5 a fit the district court's description of “very leading” questions.
. See supra note 14 for the district court's discussion of what constitutes a sufficient number of misled consumers for purposes of a Lanham Act violation.