filed a concurring opinion, in which Justice SCHNEIDER joined.
The Court directs the seven designated pretrial judges who ordered discovery of the skim-stock formula used at Firestone’s Decatur plant to vacate their order because “the plaintiffs’ evidence of unfairness ... is essentially no different than the evidence in Continental General Tire.” 979 S.W.2d at 615. While I agree that the plaintiffs in this case did not meet their burden to establish that the skim-stock formula’s disclosure was necessary to fairly adjudicate their claims and that Firestone is entitled to mandamus relief, I cannot join the Court’s opinion because it fails to fairly treat the plaintiffs’ contentions and offers little useful guidance to the bench and bar. Accordingly, I concur.
I
This case presents the Court with the opportunity to build on the foundation laid in Continental General Tire, in which we first described how trial courts should apply Rule 507 of the Rules of Evidence.1 There, we held that
*735the party resisting discovery must establish that the information is a trade secret. The burden then shifts to the requesting party to establish that the information is necessary for a fair adjudication of its claims. If the requesting party meets this burden, the trial court should ordinarily compel disclosure of the information, subject to an appropriate protective order. In each circumstance, the trial court must weigh the degree of the requesting party’s need for the information with the potential harm of disclosure to the resisting party.
In re Cont’l Gen. Tire, Inc., 979 S.W.2d 609, 613 (Tex.1998) (footnote omitted). The Court should use the case before us to expand on that decision and further explain when discovery of a trade secret is “necessary for a fair adjudication.” I recognize that the concept of “necessity” in this context must be somewhat fluid; its application will vary depending upon the circumstances presented. But determining what evidence is necessary for a fair adjudication should be more than ad hoc, as the Court’s opinion suggests. Courts in other jurisdictions have provided guidance on the subject, and some general principles can be distilled from those decisions.
Any analysis of the necessity of trade secret information to a fair adjudication must begin by examining the relationship between the trade secret information sought and the material elements of the parties’ claims and defenses. See Bridgestone/Firestone, Inc. v. Superior Court, 7 Cal.App.4th 1384, 9 Cal.Rptr.2d 709, 714-15 (1992); Coca-Cola Bottling Co. v. Coca-Cola Co., 107 F.R.D. 288, 289 (D.Del.1985).2 In the often-cited Coca-Cola Bottling Co. case, for example, the district court considered whether to allow discovery of the formula for Coca-Cola, “one of the best-kept trade secrets in the world.” Coca-Cola Bottling Co., 107 F.R.D. at 289. The plaintiffs, various bottlers, had sued Coca-Cola alleging that the company was contractually obligated to sell them syrup for the newly introduced diet Coke under the same terms as their contracts covering regular Coke syrup. The bottlers essentially complained that the two products were so similar that the syrup for diet Coke was “bottler’s syrup” within their contracts’ terms, while Coca-Cola contended that Coke and diet Coke were separate and distinct products. Id. at 296. The court noted, first, that the formula was relevant to the issue of product identity, “one of the primary issues” in the case. Id. Production of the formula was necessary for a fair adjudication, the court reasoned, to allow the plaintiffs “to address the product identity issue by comparing the ingredients of the various soft drinks involved,” and to allow them to “respond to the assertions of defendant’s experts that diet Coke and Coca-Cola are two products.” Id. at 298. Plaintiffs could not fully cross-examine Coca-Cola’s experts without the formula, nor could they prove product identity without disclosure of the secret ingredients. Id. at 297-98. Furthermore, the information was available from no other source, and no adequate substitute existed. Id. at 298.
*736Similarly, in Compaq Computer Corp. v. Packard Bell Electronics, Inc., 163 F.R.D. 329, 338-39 (N.D.Cal.1995), a district court allowed Packard Bell to discover information about another computer company’s use of recycled components in computers marketed as “new.” Packard Bell defended its use of recycled components on the ground that the practice was standard within the industry. The court reasoned that other manufacturers were not likely to voluntarily disclose information about a potentially controversial practice. Id. Thus, unless the court ordered disclosure, Packard Bell’s expert testimony about industry practice would be based on nothing “more than mere surmise.” Id. Likewise, in a patent infringement case the Federal Circuit affirmed a district court order that denied discovery of the alleged infringer’s confidential sales data because the party seeking discovery already had access to nonconfidential sales data that would serve the same purpose. Am. Standard, Inc. v. Pfizer, Inc., 828 F.2d 734, 742-43 (Fed.Cir.1987).
We can derive several guiding principles from these cases. First, trade secret information is generally discoverable when not allowing discovery would significantly impair a party’s ability to establish or rebut a material element of a claim or defense. A party’s ability is significantly impaired when the information is unavailable from any other source and no adequate alternative means of proof exist. Discovery is also necessary when the party seeking trade secret information could not knowledgeably cross-examine opposing witnesses without it, or when the party’s experts would be unable to formulate opinions supported by an adequate factual foundation. On the other hand, information that is merely cumulative is not necessary for a fair adjudication.
II
Applying these principles, I agree with the Court that the plaintiffs have not demonstrated that disclosure of Firestone’s skim-stock formula is necessary for a fair adjudication of the claims and defenses in this case. I disagree, however, that the evidence in this case “is essentially no different than the evidence in Continental General Tire.” 979 S.W.2d at 614. The plaintiffs here assert, among other claims, that Firestone’s tires were defectively manufactured.3 In order to establish a manufacturing defect, the plaintiffs must prove that “the finished product deviates, in terms of its construction or quality, from the specifications or planned output in a manner that renders it unreasonably dangerous.” Am. Tobacco Co., Inc. v. Grinned, 951 S.W.2d 420, 434 (Tex.1997). Unlike in Continental General Tire, where we specifically left the question open because plaintiffs had provided no evidence to support the theory, 979 S.W.2d at 615, the plaintiffs here offered evidence attempting to establish that disclosure of Firestone’s skim-stock formula was necessary to prove that the finished tires deviate from specifications, either because of skim-stock contamination during the manufacturing process or because ingredients were used in proportions that deviate from the formula. Robert Ochs, a former Michelin engineer with extensive experi*737ence in tire-failure analysis, testified by affidavit:
I believe there is a rubber compounding problem with the Firestone ATX and Wilderness tires. I would like to obtain these compound formulas to compare the substances which are supposed to be in the tire against those which are actually found in the tire. The best way to be able to determine what is supposed to be in the tire is to know the compounds, formulas, and specifications for the tire; one can then compare the recipe with what is actually in the tire.... The compounds and formulas for the tire components are highly probative and relevant in that they allow for a comparison between the actual, individual substances found in the failed tires and the combinations called for in the manufacturer’s formula. While the tire curing process will chemically change some of the ingredients, a comparison can still be made to determine the propriety of the materials actually detected in the cured tire.
(Emphasis added). The plaintiffs provided a similar affidavit from another engineer, Dennis Carlson. The Court quotes only a fraction of the evidence presented and characterizes it as “tautological.” Fairly read, the affidavits offer a compelling necessity for discovery that is not so easily dismissed. But while the plaintiffs here presented a stronger case for disclosure than in Continental General Tire, they nevertheless did not establish that disclosure of the skim-stock formula is necessary for a fair adjudication of their manufacturing defect claims.
First, the plaintiffs have not alleged that the skim-stock formula comparison method is necessary, or is even the superior way, to establish their manufacturing defect claim; rather, the plaintiffs have produced evidence that the formula would be “highly probative and relevant.” Trade secret information cannot be subject to disclosure based on mere relevance without rendering Rule 507 of the Texas Rules of Evidence meaningless. Cont’l Gen. Tire, 979 S.W.2d at 611 (citing Bridgestone/Firestone, 9 Cal.Rptr.2d at 712).
Second, although proving that finished tires contain material different from that called for in the skim-stock formula would be one way to prove a manufacturing defect, plaintiffs could also establish a defect by showing that the composition of tires manufactured at Firestone’s Decatur plant differed from that of tires manufactured at its other plants. See Bridgestone/Firestone, 9 Cal.Rptr.2d at 714 (citing Barker v. Lull Eng’g Co., 20 Cal.3d 413, 143 Cal.Rptr. 225, 573 P.2d 443, 454 (1978)). Moreover, the record contains evidence that Ford, without access to the skim-stock formula, was able to determine that the manufacturing process at Firestone’s Decatur plant produced flaws that contributed to tire failures. The plaintiffs also have evidence from Firestone’s own internal investigation of the tire failures that the manufacturing process at the Decatur plant allowed contamination that caused reduced cohesion between the tire’s components. Finally, Firestone’s motion to reconsider the trial court’s discovery order specifically admitted that the manufacturing process at the Decatur plant resulted in contamination of the skim stock. The plaintiffs did not argue here or in the trial court that the discovery was necessary to fairly adjudicate their defective design claims, or any other claims they have asserted.
Ill
For these reasons, I concur in the Court’s holding that the plaintiffs did not meet their burden to establish that the skim-stock formula was necessary to fairly *738adjudicate their manufacturing defect claim.
. Rule 507 provides
A person has a privilege, which may be claimed by the person or the person’s agent or employee, to refuse to disclose and to prevent other persons from disclosing a trade secret owned by the person, if the *735allowance of the privilege will not tend to conceal fraud or otherwise work injustice. When disclosure is directed, the judge shall take such protective measure as the interests of the holder of the privilege and of the parties and the furtherance of justice may require.
Tex.R. Evid. 507.
. We noted in Continental General Tire that California had adopted a rule based on the same Supreme Court Standard 508 that forms the basis of our Rule 507. In re Cont’l Gen. Tire, 979 S.W.2d at 610-11. The federal courts’ treatment of trade secrets is consistent with that standard. Id. at 611-12.
. Although the plaintiffs also apparently assert defective design claims, the trial court’s comments at the hearing on Firestone's motion to quash indicate that the trial court considered the trade secret information necessary to their manufacturing defect claim: "One thing that — I’m reading as you’re talking, and it just jumps out at me, is how — the plaintiffs make the argument here that ... how could they ever make an argument of manufacture defect if they don’t really know what the design intent was in the first place?”