dissenting.
By their nature, class-certification decisions present complex and important issues. On one hand, the class-action device affords an avenue for relief to large numbers of people who might not otherwise be able to pursue individual claims; on the other hand, the decision to certify a class can have staggering economic consequences. These concerns would, perhaps, justify our interlocutory review of class-certification orders should the Texas Legislature decide to grant us jurisdiction to do so. But as it stands, the Legislature has chosen to limit our jurisdiction over interlocutory appeals like this one unless the court of appeals’ decision conflicts with a prior decision of this court or another court of appeals. Deloitte & Touche, LLP v. Fourteenth Court of Appeals, 951 S.W.2d 894, 396 (Tex.1997); Tex. Gov’t Code §§ 22.001(a)(2), 22.225(b), 22.225(c); Tex. Civ. Prac. & RemlCode § 51.014(a)(3). Frustrated by this constraint, the Court distorts well-established conflicts jurisprudence to usurp the very power that the Legislature has deliberately denied. While I appreciate the Court’s frustration, the important issues this case presents cannot override due respect for precedent and legislative boundaries. Because we do not have jurisdiction to reach the merits of the trial court’s certification order, I dissent.
I
Jurisdiction over interlocutory appeals is generally final in the courts of appeals, absent an express constitutional or legislative grant. Tex. Gov’t Code §§ 22.225(b), 22.001(a)(2); Coastal Corp. v. Garza, 979 S.W.2d 318, 319 (Tex.1998). The Legislature has vested jurisdiction in this Court when, among other instances not pertinent here, “one of the courts of appeals holds differently from a prior decision of another court of appeals or of the supreme court on a question of law material to a decision of the case.” Tex. Gov’t Code § 22.001(a)(2); see also Tex. Gov’t Code § 22.225(c).
We have often noted how “ ‘difficult [it is] to establish conflicts jurisdiction’ ” under this limited legislative grant. Garza, 979 S.W.2d at 319 (quoting Gonzalez v. Avalos, 907 S.W.2d 443, 444 (Tex.1995) (citing Christy v. Williams, 156 Tex. 555, 298 S.W.2d 565, 567 (1957))). While complete factual identity is not required, for this Court to have conflicts jurisdiction “it must appear that the rulings in the two cases are ‘so far upon the same state of facts that the decision of one case is necessarily conclusive of the decision in the other.’ ” Id. at 319 (quoting Gonzalez, 907 S.W.2d at 444). Further, ‘“[t]he conflict must be on the very question of law actually involved and determined ... the test being whether one would operate to overrule the other in case they were both rendered by the same court.’ ” Garza, 979 S.W.2d at 319-20 (quoting Christy, 298 S.W.2d at 568-69 (quoting West Disinfecting Co. v. Trs. of Crosby Indep. Sch. Dist., 135 Tex. 492, 143 S.W.2d 749, 750 (1940))). It is this strict standard that governs our conflicts analysis.
II
The Court concludes that the court of appeals’ opinion conflicts with our decision in Southwestern Refining Co. v. Bernal, 22 S.W.3d 425, 435 (Tex.2000), because the *702trial court’s order does not specifically state how the class members’ damages will be resolved. But the court of appeals’ opinion in this case does not conflict with Bernal on this point.
In Bernal, we considered an order that certified a class of more than 900 personal injury claims arising from a refinery explosion. The order proposed a three-phase trial plan in which the defendant’s liability on various theories, including gross negligence, would be adjudicated first. Id. at 429. Phase II would address punitive damages, and Phase III would determine whether individual class members could show sufficient specific injuries or damages and whether they were proximately caused by the explosion. Id. Punitive damages determined in Phase II would then be reduced in proportion to the number of individuals who could not demonstrate actual damages or proximate cause in Phase III. Id. The court of appeals modified the trial plan to require proof of the class representatives’ actual damages before determining punitive damages. Southwestern Refining Co. v. Bernal, 960 S.W.2d 293, 298-99 (Tex.App.-Corpus Christi 1997), reversed, 22 S.W.3d 425 (Tex.2000). In evaluating the predominance requirement for certification under Rule 42(b)(4), the court of appeals failed to consider whether common issues would predominate throughout the entire trial. Instead, the court looked at each phase separately and determined that common issues would predominate in two phases of the trial, while individual issues would only predominate in determining causation and damages. Id. at 299. Considering this showing sufficient for predominance purposes, and going so far as to suggest that separate juries could be summoned to resolve the individual issues, the court of appeals affirmed the certification order as modified. Bernal, 22 S.W.3d at 429 (citing 960 S.W.2d at 297, 299).
This Court has previously held that the differences between personal injury classes and non-personal injury classes are so significant as to defeat conflicts jurisdiction. Coastal, 979 S.W.2d at 321 (“Because this case does not involve certification of personal injury claims, we cannot say that the decisions are in conflict on a material question of law such that this Court has conflict jurisdiction.”). That is because “[pjersonal injury claims will often present thorny causation and damage issues with highly individualistic variables that a court or jury must individually resolve,” such as, in that case, “each class member’s dosage, location, activity, age, medical history, sensitivity, and credibility.” Bernal, 22 S.W.3d at 436, 437. The Court does not even attempt to explain why this distinction that defeated conflicts jurisdiction in Coastal does not apply here. Instead, the Court ignores significant factual and legal differences between this case and Bernal and grounds its conflicts analysis on nothing more than disagreement with the court of appeals’ result. This has never been sufficient to invoke our interlocutory-appeal jurisdiction, until today.
But even if Bernal’s toxic exposure/personal injury context does not distinguish it from this case, the court of appeals’ decision here is not so far upon the same set of facts as to overrule Bernal. In this case, the court of appeals did not endorse the “certify now and worry later” approach that the lower courts followed and we disavowed in Bernal. Instead, it affirmed based upon a detailed examination of the extensive trial court record. The record from the five-day certification hearing consists of twelve bound volumes of documents containing over 180 exhibits, and six volumes of testimony and argument. The trial court made extensive findings enumerating the issues common to the class, *703which center around the same alleged contract breaches, the same alleged software and operational defects, the same misrepresentations, and the same scheme of sending and billing class members for unsolicited software. The court considered Schein’s argument that individual issues, including the determination of consequential damages, would overshadow these common issues, but concluded that the primary damage measure was disgorgement of the software’s purchase price, the amount of which was easily determinable from Schein’s own records. The court recognized that there may be other sources of consequential damages, but concluded that “the damages issue will not require the time-consuming, individualized inquiries that Easy Dental predicts.” 28 S.W.3d at 206. Unlike the court of appeals in Ber-nal, the court here considered the entire trial proceedings, made no suggestion that separate juries might be required to resolve individual damage issues, and determined that “proof-of-claim forms, individual damage hearings, or other manageable means” could be utilized in the event consequential damages were sought on some claims. Id. at 207.
The court of appeals’ conclusion that common issues will predominate may or may not ultimately bear out. But the court of appeals “carefully scrutinized] the predominance standard to ensure that the proposed class is ‘sufficiently cohesive to warrant adjudication by representation,’ ” as Bernal mandates. Bernal, 22 S.W.3d at 435 (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997)). Its decision would not operate to overrule Bernal, and therefore it cannot provide a basis for conflicts jurisdiction.
Ill
Having crafted a conflict upon which to base jurisdiction, the Court is then free to consider all issues in the case. Brown v. Todd, 53 S.W.3d 297, 301 (Tex.2001). Rather than proceeding to do so, however, the Court gratuitously identifies an additional purported conflict with Bernal based upon the review standard the court of appeals articulated in this case. But even a cursory review of the two opinions reveals the weakness of the Court’s conflicts analysis on this point. Moreover, the Court’s unnecessary treatment of this issue, albeit dicta, starkly illustrates its desire to expand our interlocutory-appeal jurisdiction beyond the clear parameters the Legislature has imposed.
In this case, in a section entitled “Standard of Review,” the court of appeals recited the abuse-of-discretion review standard that applies to class-certification decisions. 28 S.W.3d at 201. In the course of that discussion, the court states: “In our review of the trial court’s decision, we view the evidence in the light most favorable to the trial court’s action and entertain every presumption in favor of its judgment.” Id. In Bernal, we identified a variety of less-than-rigorous approaches some appellate courts had taken in evaluating the predominance requirement for class certification, including an approach whereby the court “indulged every presumption in favor of the trial court’s ruling, viewed the evidence in the fight most favorable to that ruling, and frankly acknowledged that if they erred, it would be in favor of certification.” 22 S.W.3d at 434. We rejected this and other “certify now and worry later” approaches to predominance in favor of a more “rigorous analysis.” Id. at 434-35. But in reviewing the trial court’s certification order under this principle, we applied an abuse of discretion standard, just as the court of appeals did here. Id. at 439. There is nothing in the court of appeals’ opinion in this case to suggest that the court was uncertain about predominance *704and erred in favor of certification. To the contrary, the court cited Bernal for the proposition that ‘“a cautious approach to class certification is essential,’ ” and acknowledged that “ ‘[i]f it is not determinable from the outset that the individual issues can be considered in a manageable, time-efficient, yet fair manner, then certification is not appropriate.’ ” 28 S.W.3d at 204 (quoting Bernal, 22 S.W.3d at 435). The court then proceeded to perform an independent predominance analysis, concluding that individualized issues did not predominate. Id. at 204-08.
The court of appeals’ description of the standard of review does not conflict with the review standard we recited in Bernal. And even if it did, we have long recognized that inconsistent statements are not sufficient for conflicts jurisdiction. See Gonzalez, 907 S.W.2d at 444 (quoting Christy, 298 S.W.2d at 567); see also Collins v. Ison-Newsome, 73 S.W.3d 178, 185 (Tex.2001) (Jefferson, J., concurring) (citing Gonzalez, 907 S.W.2d at 444 (quoting Christy, 298 S.W.2d at 567)). The court of appeals’ articulation of the review standard simply does not conflict with Bernal “ ‘on the very question of law actually involved and determined,’ ” and thus cannot invoke our jurisdiction over this interlocutory appeal. Bernal, 22 S.W.3d at 430 (quoting Coastal, 979 S.W.2d at 329-20).
IV
Schein also contends that we have jurisdiction based upon a conflict with Checker Bag Co. v. Washington, 27 S.W.3d 625 (Tex.App.-Waco 2000, pet. denied). I disagree. In this case, the court of appeals states in a footnote that “[ejonsumers are not required to prove reliance in order to recover for misrepresentations under the DTPA.” 28 S.W.3d at 206 n. 9. In Checker Bag, the court of appeals observed that a plaintiff in a DTPA laundry-list suit must establish that the defendant engaged in a false, misleading, or deceptive act on which the plaintiff relied. Checker Bag, 27 S.W.3d at 634. Undoubtedly, these two statements are diametrically opposed. But we have long recognized that inconsistent statements are not sufficient to establish conflicts jurisdiction. See Gonzalez, 907 S.W.2d at 444 (quoting Christy, 298 S.W.2d at 567). Instead, our conflicts jurisdiction arises only if one court holds differently from another on a question of law material to deciding the case. Tex. Gov’t Code § 22.001(a)(2). To meet that standard, the two decisions must conflict on “‘the very question of law actually involved and determined.’ ” Garza, 979 S.W.2d at 319 (emphasis added) (quoting Christy, 298 S.W.2d at 568-69); see Collins, 73 S.W.3d at 185 (citing Gonzalez, 907 S.W.2d at 444 (quoting Christy, 298 S.W.2d at 567)); Bland Indep. Sch. Dist. v. Blue, 34 S.W.3d 547, 552 (Tex.2000).
I am not sure that I agree with Schein that the court of appeals’ statement in this case about reliance was material to its decision. In discussing the predominance requirement, the court noted that
[t]he depositions and documentary evidence comprising the bulk of the record directly relate to ... the nature of the defects in Easy Dental’s software, the extent of Easy Dental’s knowledge of those defects, Easy Dental’s alleged uniform misrepresentations about the software and the technical support that it would provide, and Easy Dental’s alleged common scheme of sending and billing class members for unsolicited software [and that] these common issues are the most heavily disputed and will be the focus of most of the trial court’s and parties’ efforts.
28 S.W.3d at 205. The court emphasized that the plaintiffs seek disgorgement of the software’s purchase price as their pri*705mary damages measure and assert fraud and DTPA claims solely to recover exemplary damages. Id. at 206-07. Given that analysis, whether the court’s footnote about rebanee was material to its decision is questionable. '
What is clear, however, is that the Checker Bag court’s statement about reliance was not material to its decision and is dictum. There, the court noted that, to succeed in a DTPA laundry-list suit, a plaintiff must show that “(1) he is a consumer, (2) the defendant engaged in false, misleading, or deceptive acts, (3) on which the plaintiff relied, and (4) these acts constituted a producing cause of the consumer’s damages.” Checker Bag, 27 S.W.3d at 634 (citing Tex. Bus. & Com.Code § 17.50(a)(1)). The defendant challenged the factual sufficiency of the evidence to support reliance, but the court held that the defendant had not preserved error and declined to address defendant’s sufficiency challenge. Checker Bag, 27 S.W.3d at 635. Thus, it was unnecessary for the Checker Bag court to consider, nor did it consider, the reliance element’s relation to the plaintiffs claims in the case. See St. Paul Surplus Lines Ins. Co., Inc. v. Dal-Worth Tank Co., Inc., 974 S.W.2d 51, 53 (Tex.1998) (holding that defendant waived error “and thus [we] need not consider [the substantive issue].”); Keetch v. Kroger Co., 845 S.W.2d 262, 267 (Tex.1992) (holding that, because error was not preserved, “we need not decide whether failure to submit in broad form was reversible error.”). Clearly, the court’s recitation of the DTPA laundry-list elements was immaterial to its decision and was not a holding that would operate to overrule the court of appeals’ decision in this case. Consequently, Checker Bag does not present a conflict with this case for jurisdictional purposes.
V
Sehein also argues that the court of appeals’ decision conflicts with Daughety v. National Association of Homebuilders of the United States, 970 S.W.2d 178 (Tex. App.-Dallas 1998, no pet.). In Daughety, the court of appeals affirmed an order denying class certification. Id. at 182. Although the plaintiffs had sought to certify a nationwide class on several causes of action, they argued on appeal that the trial court should have certified a class on a single claim. Id. The court of appeals affirmed the trial court’s order because the plaintiffs had never presented the trial court with an opportunity to rule on a single-claim class. Id.
Sehein contends that this case conflicts with Daughety because the court of appeals “affirmed by effectively narrowing Plaintiffs’ claims, ignoring some, and holding that fraud was ‘ancillary to and subsumed by appellees’ DTPA claims.’ ” But the court of appeals did not certify a class different than the one before the trial court; instead, in considering predominance, it merely reasoned that the majority of the litigants’ efforts would be focused upon breach-of-contract questions. 28 S.W.3d at 206-07. The two opinions do not conflict on “ ‘the very question of law actually involved and determined.’ ” Bernal, 22 S.W.3d at 430 (emphasis added) (quoting Garza, 979 S.W.2d at 319).
VI
Sehein also asserts that the court of appeals’ opinion conflicts with our decision in Intratex Gas Co. v. Beeson, 22 S.W.3d 398 (Tex.2000), because the court “changed the nature of the certification order by ruling that ‘reliance’ was unimportant to the analysis of Rule 42(b)(4) and was not a common question, because the reliance-dependent claims were ‘ancillary to and subsumed by1 the DTPA claim.” But contrary to Schein’s characterization, the court of *706appeals did not rule that reliance was unimportant to the Rule 42(b)(4) analysis or was not a common issue. The court simply concluded that reliance would not be an individualized issue that would defeat predominance. 28 S.W.3d at 206-07. In reaching that conclusion, the court did not redefine the class, which we cautioned against in Intratex. Intratex, 22 S.W.3d at 406. The court of appeals’ opinion does not conflict with our decision in Intratex.
VII
Finally, Schein argues that the court of appeals’ opinion conflicts with Gutierrez v. Collins, 583 S.W.2d 312 (Tex.1979), because sections 6 and 145 of the Restatement (Second) of Conflicts of Law, which we adopted in that case, “do not mention the parties’ contractual relationship as a factor — much less the controlling factor for determining the state with the most significant relationship to the case.” The court of appeals’ decision in this case, however, falls far short of overruling Gutierrez. In Gutierrez, we held that the most-significant-relationship test set out in sections 6 and 145 of the Restatement (Second) of Conflicts would apply in all future tort cases. Gutierrez, 583 S.W.2d at 318. We concluded that the law of Texas would likely apply to a case involving Texas residents, even though the case arose from an accident in Mexico. Id. at 319. In this case, the court applied the most-significant-relationship test set out in the Restatement and Gutierrez, and concluded that Texas law applied to the class members’ contract claims. 28 S.W.3d at 208-09. The court then concluded that Texas law should apply to the class members’ tort claims because those claims arise from the parties’ contractual relationship. Id. at 209. The case before us does not conflict with Gutierrez.
VIII
While Schein raises important issues concerning the trial court’s certification order which may or may not have merit, we cannot ignore the limits the Legislature has placed on our jurisdiction. The author of today’s decision has consistently, but unsuccessfully, advocated a broader approach to conflicts jurisdiction. See Collins, 73 S.W.3d at 185-93 (Tex.2001); Garza, 979 S.W.2d at 322-26 (Hecht, J., dissenting); Wagner & Brown, Ltd. v. Horwood, 53 S.W.3d 347, 349-51 (Hecht, J., dissenting from denial of motion for rehearing of petition for review). A majority of the Court has heretofore resisted the call to abandon precedent and expand the legislative constraints on our jurisdiction. Today, inexplicably, the majority yields. Because I do not believe that our conflicts jurisprudence supports jurisdiction in this case, I dissent.