Opinion

In Rebney v. Wells Fargo Bank (1990) 220 Cal.App.3d 1117 [ 269 Cal.Rptr. 844 ], an appeal from a judgment entered on a class action settlement, Justice Benson, then a member of this court, began his opinion this way: “According to the parties, the appeals are about either collusion or fantasy. Appellants contend the [class action] settlements were the product of a collusive sellout between class counsel and the banks. Respondents claim the appeals are attributable to appellants’ fantasy vision of wondrously large money judgments.” (Id. at p. 1124.) That wry description might easily fit these appeals as well.

Several “objectors” to a class action settlement agreement reached between the plaintiff class representatives and the corporate defendant, and approved by the Alameda County Superior Court following several evidentiary hearings, appeal from the ensuing judgment, contending in substance that the settlement agreement was the product of collusion among succeeding class counsel, the class representatives, and defendant The Southland Corporation and its counsel; was not fair, adequate, and reasonable; and was *1141 grossly undervalued. The objectors also contend that the national class was improperly certified, and that the attorneys’ fees awarded objectors’ counsel by the trial court were inadequate. Objectors also present a miscellany of satellite claims going to the validity of the settlement agreement and the judgment on it. Current class counsel and defendant Southland have filed briefs defending the settlement as a reasonable outcome to prolonged and heatedly contested litigation, one in which the trial court had before it a sufficient evidentiary basis to support its finding that the settlement was fair, adequate, and reasonable, that the plaintiff class was properly certified as a national class, and that the $2.3 million in fees awarded objectors’ counsel was not unreasonable. We affirm.

I.

Background

Southland is the national franchiser of the 7-Eleven stores; the plaintiff national class is composed of all present and former 7-Eleven franchisees since 1987. The litigation began in 1993, when plaintiffs, represented by California Attorney David Franklin and others (the Franklin group) filed a complaint in the Alameda County Superior Court. In brief, the named plaintiffs alleged Southland had breached its franchise agreements with them by failing to share ratably in certain “returns, discounts and allowances” (RDA’s) over a period of several years. In effect, plaintiffs’ RDA claims presented questions of the interpretation of .a provision of the 7-Eleven franchise agreement. The complaint relied on the following provision of the standard franchise agreement between Southland and its franchisees: “Retailer discounts and allowances (including promotional and display allowances) paid to 7-Eleven and allocated or reasonably traceable to Purchases shall be credited to Cost of Goods Sold, except that 7-Eleven shall retain reimbursements for its expenditures pursuant to vendors’ co-operative advertising or other similar programs where 7-Eleven is partially or wholly reimbursed (or where costs are shared) for advertising expenditure programs.” Plaintiffs’ theory of recovery was that, by failing to credit ratably individual franchise accounts with the RDA’s, Southland had breached this provision of the franchise agreement.

The RDA categories contested by plaintiffs numbered four: advertising allowances paid to Southland by store advertisers; allowances paid South-land by its wholesalers for purchases; equipment RDA’s paid to the company by vendors for equipment furnished to retail outlets; and the “McLane Transition Allowance,” named after the McLane Corporation, a company that had purchased distribution centers from Southland Corporation and which plaintiffs alleged had paid Southland a premium that should have been shared with franchisees as an RDA. Throughout the litigation, Southland *1142 contended that under the text of the franchise agreement provision, it was not required to share with its franchisees any of the four categories of RDA’s at issue, for ¿ variety of reasons.

For plaintiffs, moreover, the California lawsuit was made problematical by parallel litigation filed in 1996 in the District Court of Dallas County, Texas, 14th Judicial District, under the caption Valente v. Southland Corp., No. 96-11972-A (Valente). The Valente suit, also brought by 7-Eleven franchisees against Southland, was filed some three years after the California suit was commenced, and sought comparable relief, including class action certification. Because the bulk of the contested RDA’s had been received in the late 1980’s, the Valente plaintiffs faced a statute of limitations defense interposed by Southland, which had already been the subject of a favorable ruling by the Texas state court. And because some two-thirds of the total number of 7-Eleven franchisees nationally were putative members of the Valente plaintiff class, a ruling by the Texas court upholding Southland’s limitations defense would have been a death knell for recoveries by a majority of the national plaintiff class.

That is, had Southland prevailed on its limitations defense in the Texas litigation, a majority of the franchisees would have recovered little or nothing, even if their view of the merits of the RDA claims proved correct in all respects. The Texas court had upheld Southland’s limitations defense but granted the Valente plaintiffs leave to amend their complaint in an attempt to circumvent the time bar. An amended complaint, together with Southland’s limitations opposition to it, was pending in the Texas trial court at the time the settlement agreement was approved in the California litigation.

Settlement of the California suit came after more than four and a half years of litigation, including voluminous discovery. The settlement proposal itself was preceded by three months of negotiations between the parties and their counsel in the summer and fall of 1997; a compromise proposed settlement was reached in negotiations between company officials and representatives of the class that August, counsel for both sides having been excluded from the talks. After being vetted by attorneys for both sides and following additional negotiations with counsel present, the proposed agreement was presented to the trial court for its tentative approval. There followed a prolonged review by the trial court of the fairness, adequacy, and reasonableness of the proposed settlement at lengthy “fairness hearings.” First, the trial court held three daylong evidentiary hearings on October 10 and November 6 and 7, 1997, to consider the objectors’ claims that the settlement was the product of collusion between class counsel and defendants. At the conclusion of those hearings, the trial judge made oral findings from the bench and later filed a memorandum order and opinion finding no factual basis for the claims of collusion alleged by former class counsel.

*1143 Apace of these developments, other events led to a rupture of the solidarity among class counsel. After former counsel to the Texas Valente plaintiffs joined forces with two California attorneys representing members of the California class action, the Franklin group moved the trial court to discharge both the class representatives and associate Texas class counsel, alleging conflicts of interest and self-dealing. The Franklin group later filed suit against the class representatives (their former clients) and their Texas class counsel colleagues, whom we will refer to as “the Culp group,” after Marc Culp, one of its members. The class representatives rejoined by discharging the Franklin group and retaining the Texas lawyers as succeeding class counsel to represent them in this litigation. These counsel, in turn, asked the trial court to grant them an injunction against litigation the Franklin group had instituted in the San Diego County Superior Court, seeking a judgment increasing the attorneys’ fees awarded them out of the proceeds of the class action settlement. The Franklin group also led the fight to overturn the proposed settlement of class claims with Southland, arguing that it was not the product of an arm’s-length deal but the fruit of collusion among South-land, the class representatives, and the Texas lawyers who had succeeded them as class counsel.

The collusion hearings were followed by extensive briefing and evidentiary submissions on the merits of the proposed settlement agreement. Additional hearings were held by the trial court on December 1, 2, and 3, 1997. At the conclusion of these hearings, the trial court entered a tentative ruling finding the proposed settlement fair, adequate, and reasonable, and the objectors’ objections to it “groundless.” The trial court also certified a national class of current and former 7-Eleven franchisees, entered an order tentatively approving the settlement, and directed that notice of settlement be mailed to all members of the class. A form of class notice was mailed in January 1998, to 2,740 current 7-Eleven franchisees and 2,714 former 7-Eleven franchisees, giving them 75 days in which to opt out of the class, file objections to the settlement or, by doing nothing, assent.

According to evidence in the record, the proposed settlement was widely discussed within the national 7-Eleven franchise community. In a court declaration, Eugene Villagrana, one of the original class representatives, recounted a meeting of the national coalition of 7-Eleven franchisees—the franchisee-elected organization formed to promote its members’ interests nationally and of which he was a member—in Universal City, California, in February of 1998. Members of the coalition’s board heard arguments favoring and opposing the settlement presented by Southland and objectors. The meeting closed with a 22-to-0 endorsement of the proposed agreement, one member abstaining.

*1144 Out of a class totaling some 5,500 franchisees, 80 elected to opt out of the settlement; objections were filed by a total of 9 franchisees, 5 of whom were represented by members of the Franklin group. On April 24, 1998, the trial court resumed hearings, this time to consider final approval of the tentative settlement agreement. Objectors, represented by the Franklin group, appeared and renewed the objections made at the previous fairness hearings. The court again concluded the allegations of the objectors were without merit and entered a final order of approval accompanied by findings of fact and conclusions of law. These appeals followed. 1

By the terms of the settlement agreement, Southland agreed to pay to the class of franchisees the sum of $37 million, including a fund of $4.75 million allocated as fees to former and current class counsel. Seven million dollars of the settlement monies would be paid to former franchisees in the form of cash; the remaining $25 million would be used to benefit current franchisees through modifications to the standard franchise agreement, the installation and maintenance of a computerized billing system (known as “Retail Information System” (RIS)) in franchise stores, and related benefits.

II.

Discussion

A. Preliminary Observations.

Before taking up the question of the fairness of the settlement agreement and the circumstances under which it was negotiated and approved by the trial court, we observe that the record before us on this appeal, summarized above, comprises some 50 volumes, running to more than 15,000 pages of briefing, rulings, discovery, and hearings transcripts. We summarize as well a kind of jurisprudence of class action settlements, relying on a brace of California and federal circuit case law defining the standard under which an appellate court reviews such matters.

1. Standard of Review of the “Fairness Hearing.”

Discussing the criteria under which class action settlements are reviewed on appeal, the court in Dunk v. Ford Motor Co. (1996) 48 *1145 Cal.App.4th 1794 [ 56 Cal.Rptr.2d 483 ] {Dunk), wrote “[o]ur task is limited to a review of the trial court’s approval for a clear abuse of discretion. [Citations.] We will not ‘substitute our notions of fairness for those of the [trial court] and the parties to the agreement. [Citations.]’ . . . ‘ “So long as the record ... is adequate to reach ‘an intelligent and objective opinion of the probabilities of success should the claim be litigated’ and ‘form an educated estimate of the complexity, expense and likely duration of such litigation, . . . and all other factors relevant to a full and fair assessment of the wisdom of the proposed compromise,’ it is sufficient.” [Citations.] Of course, such an assessment is nearly assured when all discovery has been completed and the case is ready for trial. [Citations.]’ ” {Id. at p. 1802.)

“Due regard,” the Dunk court continued, “should be given to what is otherwise a private consensual agreement between the parties. The inquiry ‘must be limited to the extent necessary to reach a reasoned judgment that the agreement is not the product of fraud or overreaching by, or collusion between, the negotiating parties, and that the settlement, taken as a whole, is fair, reasonable and adequate to all concerned.’ [Citation.] ‘Ultimately, the [trial] court’s determination is nothing more than “an amalgam of delicate balancing, gross approximations and rough justice.” [Citation.]’ ” {Dunk, supra, 48 Cal.App.4th at p. 1801 .)

A Ninth Circuit Court of Appeals panel has used even stronger language. “Our task on appeal is a very limited one,” it wrote. “ ‘The initial decision to approve or reject a settlement proposal is committed to the sound discretion of the trial judge.’ [Citation.] We are not permitted to ‘substitute our notions of fairness for those of the district judge and the parties to the agreement.’ ” {Class Plaintiffs v. City of Seattle (9th Cir. 1992) 955 F.2d 1268, 1276 .) “Neither the trial court nor this court is to reach any ultimate conclusions on the contested issues of fact and law which underlie the merits of the dispute, for it is the very uncertainty of outcome in litigation and avoidance of wasteful and expensive litigation that induce consensual settlements.” {Officers for Justice v. Civil Service Com’n, etc. (9th Cir. 1982) 688 F.2d 615 , 625 {Officers for Justice).) In other words, “the settlement or fairness hearing is not to be turned into a trial or rehearsal for trial on the merits.” {Ibid.)

Great weight is accorded the trial judge’s views. The trial judge “ ‘is exposed to the litigants, and their strategies, positions and proofs. He is aware of the expense and possible legal bars to success. Simply stated, he is on the firing line and can evaluate the action accordingly.’ ” {City of Detroit v. Grinnell Corporation (2d Cir. 1974) 495 F.2d 448, 454 , quoting Ace Heating & Plumbing Company v. Crane Company (3d Cir. 1971) 453 F.2d *1146 30 , 34.) To merit reversal, both an abuse of discretion by the trial court must be “clear” and the demonstration of it on appeal “strong.” {Dunk, supra, 48 Cal.App.4th at p. 1801 ; Officers for Justice, supra, 688 F.2d at p. 626 [“strong showing [of] clear abuse of discretion”]; Linney v. Cellular Alaska Partnership (9th Cir. 1998) 151 F.3d 1234, 1238 [“we will reverse the district court’s decisions certifying a class only upon a showing that the [trial] court abused its discretion.”].)

2. Criteria for Determining Fairness.

The trial court possesses a broad discretion to determine the fairness of the settlement, a discretion exercised through the application of a handful of identified criteria. Both the federal circuit courts and our Court of Appeal have adopted a mix of relevant considerations, including “[1] the strength of plaintiffs’ case, [2] the risk, expense, complexity and likely duration of further litigation, [3] the risk of maintaining class action status through trial, [4] the amount offered in settlement, [5] the extent of discovery completed and the stage of the proceedings, [6] the experience and views of counsel, . . . and [7] the reaction of the class members to the proposed settlement.” {Dunk, supra, 48 Cal.App.4th at p. 1801 .) The list of factors is not exhaustive and “should be tailored to each case.” {Id. at p. 1801.) According to the Dunk court, “a presumption of fairness exists where: (1) the settlement is reached through arm’s-length bargaining; (2) investigation and discovery are sufficient to allow counsel and the court to act intelligently; (3) counsel is experienced in similar litigation; and (4) the percentage of objectors is small.” {Id. at p. 1802; see also Newberg & Conte, Newberg on Class Actions (3d ed. 1992) § 11.41, p. 11-91.)

Finally, “[i]t cannot be overemphasized that neither the trial court in approving the settlement nor this Court in reviewing that approval have the right or the duty to reach any ultimate conclusions on the issues of fact and law which underlie the merits of the dispute. It is well settled that in the judicial consideration of proposed settlements, ‘the [trial] judge does not try out or attempt to decide the merits of the controversy,’ [citation] and the appellate court ‘need not and should not reach any dispositive conclusions on the admittedly unsettled legal issue.’ ” {City of Detroit v. Grinnell Corporation, supra, 495 F.2d at p. 456 .)

B. The Merits of Objectors’ Appeals.

1. Claim of Evidentiary Inadequacy of the Trial Court’s Conclusion That the Settlement Was “Fair, Adequate and Reasonable.”

Starting from the proposition that the trial court was under a fiduciary duty running to the absent class members to ensure the settlement was “fair, *1147 adequate and reasonable,” objectors assert that in making that determination, the trial court was required to consider the value to the class of the alternative of litigating the case through trial to judgment. Instead of undertaking that calculation, objectors insist the trial court delegated the determination of fairness, adequacy, and reasonableness to the settlement proponents and their counsel. It thus failed to independently evaluate the best possible recovery plaintiffs might have achieved without the proposed settlement, leaving this important function in the hands of the settlement proponents and their lawyers, groups already committed to the settlement and thus irretrievably biased.

Objectors assert that, given the limitations of the California discovery, the absence of any significant discovery in the parallel Texas Valente case, and gaps in the RDA financial data provided by Southland, the trial court lacked any ability to assess the plaintiff class’s maximum potential recovery, had the decision been made to litigate the case rather than settle it. The only evidence offered at the fairness hearings, according to objectors, was the declaration of Paul Hanson, a Southland auditor-employee, Mr. Culp’s legal opinions based on reading 3,000 pages of depositions, and a stale document summary prepared by Franklin’s office, and “allusions” to undocumented representations made to Mr. Culp by Southland employees during settlement negotiations. But the Hanson declaration, objectors contend, was prepared for purposes of defending Southland’s position in this very litigation and was; moreover, incomplete and inaccurate on its face. At the least, they assert, it is fair to assume Hanson’s revenue estimates were on the conservative side. Moreover, Hanson reviewed only cash receipts records and some documents relating to equipment. Missing from his analysis was any reference to so-called “off-invoice” discounts or other non-cash-receipt RDA’s earned by Southland. And while there was some evidence that Southland took off-invoice discounts without sharing them with the franchisees, the trial court had no hard data—no numbers—with which to estimate the total sum of these unshared discounts. Thus, objectors conclude, a document prepared for purposes of litigation ultimately became the foundation for the settlement agreement.

In addition, Mr. Hanson excluded all individual cash receipts of $20,000 or less. Using his numbers alone, objectors argue, there is evidence of $762 million in RDA credits that should have been included but were not. Indeed, according to objectors, the trial court could not even determine whether the current franchisees were receiving a net positive settlement or giving up more than they were receiving. The same holds for the recoveries by current franchisees versus former franchisees. Objectors contend, in short, that the *1148 proponents of the settlement failed to carry their burden of proving the settlement was fair, adequate, and reasonable.

Finally, objectors argue that the settlement proponents had limited familiarity with the discovery that had been taken. Culp reviewed aged summaries of discovery documents and admitted he had not reviewed any discovery for several months prior to the settlement negotiations. He failed to recognize any documents with which he was confronted on cross-examination at the fairness hearings. And, if anything, the class representatives were even more in the dark than class counsel, owing to a protective order barring their access to the discovery results. The trial court acknowledged Franklin was the most knowledgeable of the class attorneys, but placed no weight on the fact that he had been excluded from the settlement negotiations. These facts, objectors assert, place squarely in question the reliability of the information on which the settlement negotiations and agreement were based. They suggest; moreover, the existence of collusion among succeeding class counsel and Southland and its attorneys in removing the Franklin group as class counsel.

In rebuttal, the proponents of the settlement agreement direct our attention to the overarching standard under which class action settlements are reviewed on appeal. That standard is whether, in concluding the settlement agreement was “fair, adequate and reasonable” to absent class members, the trial court abused its discretion. Applying to this record the seven factors identified by the Court of Appeal in Dunk, supra, 48 Cal.App.4th 1794, 1801 , there is no question, proponents tell us, that the trial court’s decision to approve the settlement was both proper and adequately supported by the evidence before it. The franchisee class, according to their analysis, had pleaded two claims with good-to-high potential merit—the equipment and the wholesale purchases RDA’s. By class counsels’ estimate, if successful, these claims totaled less than $20 million in damages. A third claim, for an estimated $20 million—the Southland Distribution Center (SDC) RDA claim—had less than a 50-50 chance of success, according to class counsels’ analysis. And the fourth and final RDA category—the claim for a share of advertising RDA’s worth an estimated $25 million—had almost no chance of succeeding in counsels’ estimation, a conclusion founded on the view that most 7-Eleven advertising during the relevant period qualified as “cooperative,” i.e., shared between the advertiser and Southland and thus exempt under the franchise RDA provision. More importantly, recoveries on the equipment claim might be lost altogether if Southland’s limitations defense in the Texas litigation proved successful. In addition, trial of the California case was imminent, would probably have lasted for months, and class *1149 counsel were not prepared to prove any of the RDA claims except the first two—the equipment and the purchase commitments.

According to proponents, three sources of evidence supported the settlement agreement: extensive written evaluations of the case by the Culp group as class counsel, and testimony from Paul Hanson, from Culp, and from Eugene Villagrana, a franchisee and member of the national coalition. And notwithstanding objectors, the trial judge took the view that Mr. Culp was in a better position to evaluate the settlement terms than Mr. Franklin because he was the only lawyer on plaintiffs’ side of the case to pull together the separate work of six different attorneys in the California litigation. Culp’s confidential 10-page analysis to the class representatives of April 1997, filed with the trial court by Franklin, was a candid assessment of the case and its prospects.

Even objectors’ counsel agrees that “substantial discovery” occurred “in the California case.” The trial court likewise observed that Mr. Franklin and his associates, as litigation counsel in the California lawsuit engaged in “vigorous, aggressive and exhaustive” discovery against Southland. The company produced over a million pages of documents for inspection and responded to hundreds of interrogatories; plaintiffs’ counsel deposed numerous Southland employees. Moreover, in the course of the settlement negotiations, Southland produced additional RDA financial data for the years 1987 to 1996. These data were compiled by Hanson, the Southland auditor, who swore to their accuracy and who was cross-examined by Franklin at the preliminary approval hearing.

Proponents argue, moreover, that it does not matter that little or no discovery had been conducted in the Texas case since the same RDA’s at issue in the California litigation were also central to the Texas lawsuit. The relevant invoices were received in Southland’s Dallas headquarters, not in the field. The RDA financial data that Southland disclosed in the California litigation were, according to counsel, aggregate nationwide data, which by definition related to the entire country, not just to California. Excerpts from the deposition testimony of a Southland auditor appear to confirm this. The invoices related not only to the California franchisees but to Southland’s national RDA practices. Proponents conclude that the extensive discovery taken in the California case to identify the types of RDA’s that were and were not credited to the franchisees was equally applicable to Valente, and was more than sufficient to analyze the strength of the nationwide claim.

By stipulation of the parties, the discovery depositions taken in the California litigation would be admissible in the Texas litigation. But the *1150 compelling response to objectors’ claim that no discovery had been taken in Texas, proponents argue, is that “Texas counsel could not deliver what [objectors’ counsel] wanted without risking losing the limitations defense” since Southland would not open up discovery without a ruling on the limitations issue. Thus, the only way to improve the Texas discovery was to roll the dice on the limitations defense. That, apparently, is what objectors’ counsel wanted to do. Succeeding class counsel, however, considered the gamble not in the best interests of the class. Given this record, we cannot say they were wrong or, more accurately, that the trial court abused its considerable discretion in so concluding. In any event, that no class action can be settled until the last particle of discovery has been completed and analyzed is not the law. “[I]n the context of class action settlements, ‘formal discovery is not a necessary ticket to the bargaining table’ where the parties had sufficient information to make an informed decision about settlement. . . . [Citation.] .... ‘[Notwithstanding the status of discovery, plaintiffs’ negotiators had access to a plethora of information regarding the facts of their case.’ ” (Linney v. Cellular Alaska Partnership, supra, 151 F.3d at pp. 1239-1240; see also In re Corrugated Container Antitrust Litigation (5th Cir. 1981) 643 F.2d 195 , 211 [‘“It is true that very little formal discovery was conducted and that there is no voluminous record in the case. However, the lack of such does not compel the conclusion that insufficient discovery was conducted’ ” (italics omitted)]; Plummer v. Chemical Bank (2d Cir. 1982) 668 F.2d 654, 660 [“we do not mean to suggest that extensive pre-trial discovery is a prerequisite to approval or that obstructionist tactics by a minority of objecting class members should be permitted to turn a settlement hearing into a full blown trial on the merits”].)

In any event, the merits of the underlying class claims are not a basis for upsetting the settlement of a class action; the operative word is “settlement.” (See, e.g., Officers for Justice, supra, 688 F.2d at p. 625; City of Detroit v. Grinnell Corporation, supra, 495 F.2d at p. 455 [“The fact that a proposed settlement may only amount to a fraction of the potential recovery does not, in and of itself, mean that the proposed settlement is grossly inadequate and should be disapproved.”]; Linney v. Cellular Alaska Partnership, supra, 151 F.3d at p. 1242 [“This court has aptly held that ‘it is the very uncertainty of outcome in litigation and avoidance of wasteful and expensive litigation that induce consensual settlements. The proposed settlement is not to be judged against a hypothetical or speculative measure of what might have been achieved by the negotiators.’ ” (italics omitted)].) All of the class claims were vigorously contested by Southland. At the same time, Southland understandably wanted to put an end to divisive litigation by settling the lawsuit. Given these circumstances, class counsel recommended the litigation be settled for $37 million. The members of the national coalition *1151 association accepted that proposal by a vote of 22 to 0, with one member abstaining. Ninety-nine percent of the members of the franchisee class voted to accept it. As a “fiduciary” of the absent class members, the trial court’s duty was to have before it sufficient information to determine if the settlement was fair, adequate, and reasonable. (Dunk, supra, 48 Cal.App.4th at p. 1802 .)

We turn, then, to a consideration of the record in light of the factors identified by the court in Dunk, supra, 48 Cal.App.4th at page 1801 , tailored to the circumstances of this particular case. 2 Before doing so, however, we note that a “voluntary conciliation and settlement are the preferred means of dispute resolution. This is especially true in complex class action litigation . . . .” (Officers for Justice, supra, 688 F.2d at p. 625) and that, according to the Dunk court, “a presumption of fairness exists where: (1) the settlement is reached through arm’s-length bargaining; (2) investigation and discovery are sufficient to allow counsel and the court to act intelligently; (3) counsel is experienced in similar litigation; and (4) the percentage of objectors is small.” (Dunk, supra, 48 Cal.App.4th at p. 1802 .)

a. Strength of Plaintiffs’ Case.

Mr. Culp presented his assessment of the merits of plaintiffs’ case in testimony at the approval hearings. Counsel thought two of the four RDA claims had a good chance at prevailing if the case went to trial—the equipment claim and the McLane transition allowance claim. These claims Culp estimated had a value of between $15 million and $20 million respectively. But he thought the plaintiff class would lose the SDC RDA claim for the reasons given above, as well as the advertising allowance claim, because the bulk of these were cooperative in nature and thus exempt from sharing under the terms of the RDA franchise provision. Culp also acknowledged the role played by the significant limitations issue pending in the Texas court and that failure there would wipe out a majority of the claims of the non-California class members. Moreover, because of an apparent breach of the attorney-client privilege, the trial court had before it a copy of Culp’s confidential memorandum to the class representatives, authored well before the settlement negotiations had begun, detailing a candid assessment of the merits of the case. And that document was not only not an after-the-fact appraisal, but its conclusions dovetailed with those of the trial judge, who remarked, “I have looked at these contract provisions and I have serious reservations about whether there have been breaches sufficient even to bring these damages issues into account with the singular exception of the equipment allowances.”

*1152 b. The Risk, Expense, Complexity, and Duration of Further Litigation and the Risk of Maintaining Class Action Status Through Trial.

Proponents suggest that trial of the RDA claim might well require an examination of every invoice that Southland’s vendors had provided it for over a decade in order to ascertain the magnitude of the claims at issue. Whether that is true or not, it seems clear enough that trial of the case would have required the preparation and testimony of opposing financial and accounting experts, as well as their depositions and the not insubstantial associated litigation costs. Trial itself would have been lengthy and complex. And, as detailed elsewhere in this opinion, the risks of maintaining class action status and pursuing judgment through trial would have been large.

c. The Amount Offered in Settlement.

Proponents point out that, given the problems with proof, the uncertain fate of the limitations issue in the Texas lawsuit, and an interest on the part of both sides in ending divisive litigation that threatened to harm ongoing business relationships, a settlement fund of $37 million was not unattractive. The litigation interfered with attempts to cooperate in a successful business, especially after recent business reversals suffered by Southland and its franchisees.

d. Stage at Which Settlement Is Reached.

As noted, settlement came only after some four and a half years of litigation, including voluminous discovery and many motions filed and argued by both sides. The trial court was afforded the opportunity both to review many of the materials produced by discovery and to observe at close hand the ability of counsel for all of the parties. (See Dunk, supra, 48 Cal.App.4th at pp. 1802-1803.)

e. The Experience and Views of Counsel.

The resumes of class counsel are a matter of record and disclose that all were experienced litigators, including several with prior class action litigation experience. And, as noted, class counsel recommended settlement on the negotiated terms.

f. The Reaction of the Class Members to the Proposed Settlement.

According to the record, the response of the absent class members to the proposed settlement with Southland was overwhelmingly positive. A mere *1153 80 of the 5,454 national class members elected to opt out and most of them had particularized reasons for doing so that were unrelated to the RDA claims. A total of nine members—five of whom were clients of the Franklin group—objected to the settlement. As the trial court observed, the magnitude of the favorable response was particularly impressive in this case, coming from a class composed of not unsophisticated business owners and operators. Finally, the national coalition of 7-Eleven franchisees voted 22 to 0 in favor of the settlement, an indictor we find particularly telling.

In light of this evaluation, we cannot say the trial court abused its discretion in concluding the settlement was fair, adequate, and reasonable. Indeed, we think the four factors identified by the court in Dunk—arm’s-length bargaining; adequate investigation and discovery; experienced counsel; and a small percentage of objectors—lead to a presumption the settlement was fair, a presumption objectors have failed to overcome. (See Dunk, supra, 48 Cal.App.4th at p. 1801 .)

2. The “Due Diligence” Controversy.

In tentatively approving the settlement agreement, objectors tell us, the trial court relied heavily on representations by class counsel that a postsettlement “due diligence audit” of RDA financial data produced by Southland would take place, observing that “[tjhese [verification] provisions furnish additional insurance that the settlement will be based on adequate information about the potential magnitude of the claims in this case.” The class notice contained similar assurances, advising absent class members that “If the accountant finds significant errors in the [allowance] data . . . the class representatives have the option of terminating the settlement.” The promised “audit” thus became, objectors argue, the pillar supporting the trial court’s determination that the settlement was fair, adequate, and reasonable. In the event, however, objectors argue the due diligence audit metamorphosed into an unsworn farrago of hearsay statements by class counsel in a letter to the class representatives relaying what the accountants allegedly told him they had found. No signature of any accountant or auditor was ever presented to the court; no testimony regarding a real audit was ever taken. The trial court had no opportunity to examine anyone as to findings made or the procedures followed by the accountants. The vaunted “audit,” objectors assert, failed even to qualify as a compilation of documents, the lowest level of accounting work product. The trial court was noticeably uncomfortable at the final approval hearing, objectors assert, and in the end announced it would “not consider” the Culp summary of the accountants’ findings. However, the court then referred to that document several times during the remainder of the hearing.

*1154 According to objectors, these circumstances placed the trial court in a Catch-22 situation: if it considered the Culp due diligence letter, that letter failed to provide the promised analysis of financial data; if it did not consider the letter, the trial court was left without any basis for filling the informational gaps it had acknowledged existed at the preliminary approval stage. The trial court was left with no real choice, objectors conclude; it was “ ‘presented with what looks like a fait accompli.’ ” (Quoting In re General Motors Corp. Pick-Up Truck Fuel Tank Products Liability Litigation (3d Cir. 1995) 55 F.3d 768, 789 (Pick-Up Truck Litigation).)

The proponents’ response to this line of argument is to remind us that “no good deed goes unpunished.” While the Franklin group’s discovery efforts may have been ineffective, they did not undermine class counsels’ ability to evaluate the settlement value of the case. According to proponents, the Franklin group had done little to confirm the accuracy of Southland’s figures regarding the magnitude of the RDA’s in controversy. The only information counsel had obtained was Southland’s interrogatory answers to the effect that its figures were substantially accurate. Succeeding class counsel, however, did not believe it was prudent to accept these figures without some degree of independent checking. Accordingly, as part of the settlement negotiations, the parties agreed that following preliminary approval of the settlement by the trial court, class counsel would select accountants to test the validity of Southland’s RDA data by sampling the underlying records, interviewing Southland’s accounting personnel, and taking related “due diligence” measures; Southland would pay $30,000 to defray the cost of the review work; and the settlement would be voidable at plaintiffs’ option if significant errors—defined as understating aggregate dollar amounts by 15 percent or more—were found. As adopted by the negotiators and accepted by Southland and the class representatives, the due diligence provision took the following form:

Pointing to the text of this provision, proponents say the verification procedure was never intended to be an “audit” or to result in a formal report. Instead, the process was meant to give the class representatives and their attorneys additional assurance that the RDA financial data provided by Southland were accurate. An actual audit of the RDA data would have required review of hundreds of millions of dollars in RDA’s that Southland had processed over the preceding 10 years, and would have been time-consuming and costly. Instead, according to class counsel, independent accountants were hired to review Southland’s work and determine if there was any reasonable basis to believe there were substantial errors in the information provided. According to Mr. Culp’s letter to the class representatives, lodged with the trial court, the accountants found no reason to believe Southland’s figures were misstated. We think the text of the “due diligence” provision of the settlement agreement, quoted above, bears out this view of the role of the sampling by the accountants. And while the counsel’s report on the results of the sample might have been handled more deftly, no “audit” was required as a condition of court approval. The due diligence process that was employed by class counsel went beyond any legal requirement for a valid settlement.

3. Claims of Collusive Settlement.

According to objectors, this case is “replete with circumstantial evidence of an unholy alliance, evidenced by a single-minded ignoring of hard questions and of cooperation with and blind reliance upon Southland, between the proposed class representatives, [Texas counsel], some of the California counsel and defendant Southland and its attorneys.” Relying on Pick-Up Truck Litigation, supra, 55 E.3d. at page 789, objectors insist the result of that alliance was to exclude the most knowledgeable attorney for the plaintiff class, Franklin, from the settlement negotiations; to exclude any putative class representatives who even questioned the settlement; and to move the proposed settlement through the process with a momentum and inevitability that made the absent class members accept it as a fait accompli. Moreover, in what objectors term was a “direct violation of the views of the Third and Ninth federal circuits,” the settlement agreement was negotiated in tandem with the proposed award of attorneys’ fees to class counsel. The *1156 Third Circuit, objectors argue, has virtually prohibited the conjunctive negotiation of attorneys’ fees with the terms of class action settlement agreements. (See Pick-Up Truck Litigation, supra, 55 F.3d at p. 790 .) In this case, however, the terms of the proposed settlement and class counsels’ fees were negotiated in a package deal, circumstantial evidence, objectors argue, of a collusive settlement made to the disadvantage of excluded class counsel and the absent class members.

We explore the claim that the simultaneous negotiation of settlement terms and class counsel’s fees is a per se violation at greater length below. (See post at p. 1158.) But as for the claim that momentum for settlement pushed aside substantive concerns over its fairness, serious settlement discussions often induce a kind of “momentum.” And the “momentum” and “fait accompli” language objectors invoke is just that—comments made by courts in “true” settlement-only class actions, making them not pertinent here. (See Mars Steel v. Continental III. Nat. Bank & Trust (7th Cir. 1987) 834 F.2d 677 , 680-681 {Mars Steel) [“where notice of the class action is . . . sent simultaneously with the notice of the settlement itself, the class members are presented with what looks like a fait accompli”]; Pick-Up Truck Litigation, supra, 55 F.3d at p. 789 [“the mere presentation of the settlement notice with the class notice may pressure even skeptical class members to accept the settlement out of the belief that, unless they are willing to litigate their claims individually . . . they really have no choice”].) As we explain later in this opinion, this is not a genuine “settlement-only” class action for the obvious reason that it had been heatedly litigated for over four years before settlement discussions ensued.

Linked to the claim that counsel sold out the class for a mess of pottage is the related claim that James Alberts, a member of the original group of franchisee-plaintiffs who brought the California suit, had been saddled with a judgment for $42,000 in costs taxed against the class representatives when summary judgment was granted against non-Southland defendants on plaintiffs’ failed antitrust causes of action. On that ground alone, objectors argue, Alberts was in a position to collude with Southland by promoting an inadequate settlement in exchange for a forgiveness of the $42,000 costs bill. It is true that the trial court expressed some unhappiness with Mr. Alberts, but what provoked its ire was Alberts’s failure to disclose the potential conflict when the class was certified the preceding November.

Moreover, Mr. Alberts was one of three class representatives. The trial court held a two-day evidentiary hearing to assess these claims on November 6 and 7. It heard from eight witnesses, including principals at the August *1157 settlement negotiations with Southland. So far as we are able to glean from the record, Southland offered Mr. Alberts nothing; there was no agreement to settle his suit filed by Mr. Franklin; the only time the topic came up, Mr. Alberts was told the cost bill issue could not be coupled with the settlement discussions and would have to wait. At the conclusion of the November hearings, the trial court entered a finding that the “charges of fraud, collusion or other wrongdoing by class representatives are entirely groundless and unfounded . . . . ftQ . . . There is no evidence that any of them did anything unethical or improper or contrary to the interests of the class.”

As for Mr. Alberts being the chief instigator of the settlement, the two other class representatives participated in the August negotiations as well; they conferred with class counsel during the talks; and counsel continued to negotiate the terms with Southland lawyers over several additional months. Even if Mr. Alberts were pursuing a personal agenda—a supposition lacking support in the record—the remaining class representatives appear to have been fully capable of protecting the interests of the class members.

The trial court explored the Franklin group’s claim of a collusive settlement at three daylong evidentiary hearings—on October 10 and again on November 6 and 7, 1997—and in the end pronounced them “groundless.” The latter two collusion hearings included the testimony of several witnesses who had participated in the settlement negotiations. Six witnesses testified at the three-day December fairness hearings. Counsel for the objectors submitted over 800 pages of argument and supporting materials to the trial court. The trial court considered these objections again at the daylong final hearing of April 24, 1998. And preceding that, counsel for objectors had filed with the court more than 6,500 pages of materials in opposition to the settlement agreement and fee requests. We cannot, in the face of such a record, say the trial judge abused his discretion in finding the claims of a collusive settlement to be without substance.

Next, objectors assert the settlement agreement’s establishment of an allowance review committee (ARC), made up of seven franchisees to monitor Southland’s RDA processing for a period of six years, is nothing more than a thinly disguised form of payoff to the class representatives for delivering an undervalued settlement to Southland. It was Alberts, objectors say, who was instrumental in negotiating the ARC provision, which sets aside a budget of $500,000 for six years and provides each committee member with $10,000 a year in compensation. Asserting that the average annual compensation for a 7-Eleven franchisee is $30,000, objectors assail the ARC as a form of additional compensation to the class representatives, a *1158 means of obtaining disproportionate economic benefits for themselves, not shared with the remainder of the class.

The ARC provision of the settlement agreement, however, did not identify anyone as a committee member; the class representatives were to select five of the seven members and Southland the remaining two. And rather than a perquisite, the annual compensation paid the ARC members is in exchange for work performed by them in monitoring Southland’s RDA accounting. We agree with the proponents that the ARC provision of the settlement imposes substantial responsibilities on its members and that the annual compensation is not unreasonable.

4. Claim Class Counsel Had Substantial Conflicts of Interest and Could Not Adequately Represent the Class.

Under this heading, objectors assert the Culp firm originally was retained by the Texas franchisees in 1996 to pursue the Valente litigation. By late 1997, objectors tell us, the firm was pursuing settlement discussions without the knowledge of their clients, and before either class certification or the selection of class counsel had occurred. Eight of the ten class representatives in the Texas litigation asked the Culp group to cease negotiations with Southland until some discovery had been conducted. The upshot of this request, objectors say, was the Culp group’s withdrawal from representing the unhappy clients, and its representation of the remaining two class representatives, on whose behalf it continued to negotiate a settlement of the Texas litigation with Southland. As eventually negotiated with Southland, the California settlement gave the Culp group $1 million in attorneys’ fees and placed limits on the fee recovery of the Franklin group, which had done all the work. It is reasonable to infer, objectors infer, that a collusive trade-off occurred in which the class and former counsel took less and the settlement attorneys took more of the pie.

According to objectors, the Third Circuit has flatly prohibited simultaneous negotiation of attorneys’ fees and settlement terms. That is no longer, however, an accurate statement. While the Third Circuit formerly barred simultaneous negotiation of class settlement terms and counsel fees, in its Pick-Up Truck Litigation opinion, the court acknowledged that the United States Supreme Court’s opinion in Evans v. Jeff D. (1986) 475 U.S. 717, 734-738 [ 106 S.Ct. 1531, 1540-1543 , 89 L.Ed.2d 747 ] “overruled [the Third Circuit’s] strict rule prohibiting simultaneous negotiations” of counsel fees and settlement terms. {Pick-Up Truck Litigation, supra, 55 F.3d at p. 804 ; see Evans v. Jeff D., supra, 475 U.S. at p. 734 [ 106 S.Ct. at p. 1541 ] [“If *1159 [attorneys’] fee waivers cannot be negotiated, the settlement package must either contain an attorneys’ fee component of potentially large and typically uncertain magnitude, or else the parties must agree to have the fee fixed by the court”].) Even for the Third Circuit, the timing of class counsel’s attorneys’ fee negotiations is now merely one factor in evaluating the settlement’s adequacy. (See Pick-Up Truck Litigation, supra, 55 F.3d at p. 804 [“we see no reason why Jeff D. or its underlying policy of avoiding rules that impede settlement preclude us from considering the timing of fee negotiations as a factor in our review of the adequacy of the class’s representation”].) Here, of course, there is no contention that the total fee award is excessive; in point of fact, the Franklin group takes the opposite position, arguing that more in the way of attorneys’ fees should have been granted them.

Objectors’ related claim that the Culp group had a conflict because they withdrew as class counsel in the Texas litigation also falls short. Counsel’s withdrawal was approved by the trial judge in the Valente case and the remaining class representatives were successively represented by members of the Franklin group. Moreover, conflict and division between class counsel and class representatives are sufficiently common in class action litigation to have made the appellate reports; counsel’s duty runs to the class as a whole and the governing rule is, in the event of conflicts, withdrawal is the appropriate course for counsel to take. (See, e.g., Kincade v. General Tire & Rubber Co. (5th Cir. 1981) 635 F.2d 501, 508 [“Because the ‘client’ in a class action consists of numerous unnamed class members as well as the class representatives, and because ‘[t]he class itself often speaks in several voices . . . , it may be impossible for the class attorney to do more than act in what he believes to be the best interests of the class as a whole . . . .’ ”]; Maywalt v. Parker & Parsley Petroleum Co. (2d Cir. 1995) 67 F.3d 1072 , 1077 [“Inherent in any class action is the potential for conflicting interests among the class representatives, class counsel, and absent class members”]; In re Agent Orange Product Liability Litigation (2d Cir. 1986) 800 F.2d 14, 19 [“A motion to disqualify an attorney who has represented the entire class and who has thereafter been retained by a fraction of the class to represent its interests in opposition to a proposed settlement. . . cannot be automatically granted. Rather, there must be a balancing of the interests of the various groups of class members and of the interest of the public . . . .”].) In light of this repeated judicial recognition, we cannot say the Culp group acted improperly in withdrawing as counsel for the Valente plaintiffs who opposed the settlement and in pursuing it on behalf of some of the class representatives.

Finally, objectors claim Mr. Franklin was excluded from the settlement negotiations with Southland officials. The record, however, fails to bear out *1160 this contention. Franklin was asked to allow the class representatives to meet with Southland executives in August of 1997, and he agreed. He advised his clients by telephone during the course of those meetings. And Mr. Franklin and a colleague participated in several subsequent negotiations with South-land executives. We find no merit in this claim.

5. Claim Trial Court Failed to Independently Determine Whether the Proposed National Settlement Class Met Certification Standards.

Objectors complain that the trial court conducted no hearings and observed none of the required formalities under Federal Rules of Civil Procedure, rule 23(a) (28 U.S.C.) 3 in certifying a national class of plaintiff 7-Eleven franchisees. (See Pick-Up Truck Litigation, supra, 55 F.3d at p. 785 [“the procedural formalities of certification are important even if the case appears to be headed for settlement rather than litigation”].) They argue that a fairness determination alone is no surrogate for rule 23 findings, and that even after being reminded of the need to do so at the final settlement proceedings, the trial court conducted no hearings on anything except allegations of fraud, collusion, and the fairness of the proposed settlement. No evidence was presented. And no evidence, objectors charge, supports the trial court’s finding that a national settlement class was appropriate for certification. A review of the five days of fairness hearings demonstrates the complete absence of any evidence on the issue of national class certification. The trial court failed entirely to require a showing of adequate representation by either counsel or the proposed class representatives of the national class. We are urged to hold that this absence is fatal to the proposed settlement.

In rebuttal, proponents assert it would have made little sense to Southland to have settled the class litigation in California only, leaving the pending Texas proceedings to continue to trial. The company had no interest in settling one case while the other continued, because piecemeal settlement would not have eliminated litigation costs or ended the threat of divisiveness between Southland and its franchisees. Indeed, Southland was concerned that such a path would have increased tensions among the class of franchisees—between California 7-Eleven retail operators and those in other states. The question thus became, according to the settlement proponents, whether the California court was the proper forum for a global settlement of the *1161 franchise issues. And indeed, it appears that former class counsel, Mr. Franklin, at one time thought it was. The California case had been filed three years earlier than the Valente litigation in Texas. And the trial court had held an all-day hearing in June of 1997, on the motion of counsel to certify the suit as a California-wide class action, and found, among other matters, that the RDA claims were suitable for class treatment.

A comparison of the complaints filed in both cases confirms that the claims pleaded by the Valente plaintiffs were not materially different from the California RDA claims. Moreover, the breakaway Texas plaintiffs had sought national (excluding California) class certification in the Texas litigation. The trial court’s findings from the June 1997 certification hearing appear equally applicable to the Valente litigation. Additional plaintiffs were added to the group of class representatives in the California suit in the June 1997 hearing. Last, the trial court also assessed the qualifications of class counsel in its order of May 8, 1998. Notably, unlike the typical class action in which a finding of representational adequacy is made early in the proceedings, the finding here came only after the trial judge had had the opportunity to observe class counsel, both in action and in cross-examination regarding the adequacy of their representation.

Writing of the so-called “settlement-only” class action, a prominent federal appellate jurist has said “where notice of the class action is . . . sent simultaneously with notice of the settlement itself, the class members are presented with what looks like a fait accompli.” (Mars Steel, supra, 834 F.2d at pp. 680-681.) Yet, although criticized, “every court that has addressed the question . . . has declined to hold that the procedure is a per se violation of Rule 23 [of the federal rules] and therefore requires that the settlement automatically be disapproved.” {Id. at p. 681.) However, the federal circuits, following the United States Supreme Court’s ruling in Amchem, supra, 521 U.S. 591 , have made it clear that certifications in such settlement-only class actions must have given “ ‘undiluted, even heightened, attention’ to class certification requirements in a settlement context” in the interest of blocking unwarranted or overbroad class definitions. {Hanlon v. Chrysler Corporation (9th Cir. 1998) 150 F.3d 1011 , 1019, quoting Amchem, supra, 521 U.S. at p. 620 [ 117 S.Ct. at p. 2248 ].)

The Hanlon court elaborated on the Amchem requirement of “heightened attention” to certifications in the settlement context and the reasons behind it. “The Amchem Court also noted the problem of counsel ‘not prepared to try a case,’ ” it wrote, “Such counsel is, almost by definition, inadequate because an inability or unwillingness to try a case means the class loses all *1162 of the benefits of adversarial litigation. . . . [The trial court] must be skeptical of some settlement agreements put before them because they are presented with a ‘bargain proffered for . . . approval without benefit of an adversarial investigation.’ ” (Hanlon Chrysler Corporation, supra, 150 F.3d at p. 1021, quoting Amchem, supra, 521 U.S. at p. 623 [117 S.Ct. at pp. 2249-2250].)

We think this case falls between two stools. It does not qualify as a “true” settlement-only class action, for the obvious reason that it was extensively litigated for several years preceding national certification. Moreover, in December of 1997, the trial court entered findings of fact and conclusions of law approving a conditional settlement class. Among those findings and conclusions was a determination that a nationwide settlement of the class claims was appropriate because the principal claims were common to members of the national class. The trial court also made findings which can be read as satisfying the other criteria of Federal Rules of Civil Procedure, rule 23(a) (28 U.S.C.): numerosity, typicality, and adequacy of representation. To the extent objectors contend these findings were insufficient, the short answer was given by the Dunk court: “No findings were required . . . .” ( Dunk, supra, 48 Cal.App.4th at pp. 1805-1896.) But we also think that if there was error, it was harmless under the circumstances disclosed by this record. The rule 23 certification criteria had been extensively briefed by the parties prior to the daylong hearing on June 13, 1997; the trial court entered a California-only class certification order on July 23. It is clear from the record that these same criteria were applicable to the national class. Moreover, the case had been vigorously litigated for over four years, removing any concerns that class counsel negotiated a settlement without the benefit of “adversarial investigation.” Under these circumstances, we conclude no reversible error was committed.

6. Claims of Discrimination Within the Class.

Objectors next assert the settlement improperly discriminates in favor of current franchisees at the expense of former franchisees. They point out that current franchisees receive $25 million in future benefits, while former operators must settle for $7 million in cash. There is, however, no legal requirement that all members of the class must participate equally in any settlement. While intraclass disparities may be a signal of unfairness, the inference is rebuttable by a showing that differences in treatment “are rationally based on legitimate considerations.” (Holmes v. Continental Can Co. (11th Cir. 1983) 706 F.2d 1144, 1148 ; see also Kincade v. General Tire & Rubber Co., supra, 635 F.2d at p. 506, fn. 5 [no requirement that class *1163 members must participate equally in settlement]; Curtiss-Wright Corp. v. Helfand (7th Cir. 1982) 687 F.2d 171, 175 [“When a . . . judge approves a class action settlement. . . , he almost always overrides the wishes of some class members for a bigger share of the pie”].)

The distinction between the two groups of franchisees made in the settlement agreement is a legitimate one because they are differently situated. Former franchisees can only be compensated by money; current franchisees can be compensated by other means over time and through contractual changes. It was thus rational to compensate the group of former franchisees with cash payments and the latter with benefits conferred through contract modifications. The record also demonstrates that greater aggregate benefits for current franchisees is rational because they have been with Southland longer than former franchisees; former franchisees get immediate cash payments; and Southland has an ongoing business relationship with its current franchisees. Objectors also argue the settlement allocations fail to discriminate, by giving all current franchisees the same benefits, notwithstanding differences in duration of operations and sales volume. But this failure to distinguish is rational too, or at least not irrational for settlement purposes. Southland regarded identical treatment of current franchisees as less disruptive than differentiating benefits according to duration of ownership, for example.

7. Claim the Notice to the Class Was Defective.

Objectors argue the notice of settlement mailed to the class members was biased, emphasizing the favorable terms of the agreement and burying its negative components in fine print or leaving them out altogether. As examples from a laundry list of omissions, objectors say that the notice fails to disclose that each month of delay in the computerized billing system saves Southland substantial sums; that a “clarification” of the franchise definition of “cost of goods” in the settlement agreement will have the effect of costing the franchisees millions in lost income over the next 10 years; that the agreement imposes a two-year period of repose during which franchisees must assert RDA claims even if concealed by Southland; that the notice failed to mention the range of potential recovery found by the trial court; that it failed to mention that most franchisees have more freedom now to reduce operating hours than under the settlement agreement; that it failed to disclose the creation of the ARC committee and its substantial annual budget and compensation to its members; and that it failed to disclose that mediation procedures provided in the settlement agreement will prove so expensive that no one will use them. The notice, moreover, was mailed to absent class members before any independent due diligence audit had taken place.

*1164 Two standards guide our resolution of objectors’ defective notice claim. The notice must “express no opinion on the merits of the settlement” and, in determining its particulars, the trial court “has virtually complete discretion as to the manner of giving notice to class members.” (Handschu v. Special Services Div. (2d Cir. 1986) 787 F.2d 828 , 833.) Having reviewed the text of the notice in this case, we think it was adequate to “ ‘fairly apprise the prospective members of the class of the terms of the proposed settlement and of the options that are open to them in connection with [the] proceedings.’ [Citations.]” (Weinberger v. Kendrick (2d Cir. 1982) 698 F.2d 61, 70 .) The arguments of objectors that additional materials should have been included are just that, arguments that would have been inappropriate in a content-neutral advisory. The trial court did not abuse its discretion in approving the notice.

8. Claim the Award of Attorneys’ Fees to the Franklin Group Was Inadequate.

An appellate court reviews an award of attorneys’ fees in the settlement of a class action under an abuse of discretion standard. (Dunk, supra, 48 Cal.App.4th at p. 1809 ; Hanlon v. Chrysler Corporation, supra, 150 F.3d at p. 1029.) As noted ante, the settlement agreement allocated the sum of $4.75 million to current and former class counsel as attorneys’ fees and costs. The agreement made the following specific allocations among current and former class counsel:

The numbers themselves tell the story. First, the Franklin firm received the highest individual award of attorneys’ fees—$1.2 million. Moreover, by the terms of the settlement agreement, any reduction in the fees awarded an attorney goes to the class members, not to other counsel. The short of the matter is that, were we to accept Mr. Franklin’s claim that the fees awarded him were inadequate, the effect would be to unwind a $37 million settlement agreement, reached after almost five years of litigation, in which Mr. Franklin and his associates were awarded half of the $4.75 million fund reserved for attorneys’ fees. (Mr. Franklin argues that he should have been awarded $8.75 million in fees.) Second, if the claim were to succeed and the settlement derailed, the case would go to trial. As the trial court stated at the final April 1998 hearing, “If the California case is tried, based upon what I *1165 have seen in this Court, it would be a disaster for the plaintiffs . . . . 0Q ... A major disaster.”

Third, in negotiating the settlement agreement, settlement counsel were denied access to the Franklin group’s time records. Some attorneys of the Franklin group eventually did turn over their time records, although two of those were post hoc efforts; a third attorney had billed substantial hours on what settlement counsel regarded as “wasteful” efforts in the failed antitrust claims. Last, the trial court determined the amount of the fees was reasonable, given considerations of quantity and quality. It appears that Franklin billed for much of the unsuccessful antitrust work and for the time spent opposing the settlement.

These flaws in former class counsel’s performance were perhaps best summed up by the trial judge who commented on objectors’ efforts to overturn the settlement at the close of the December 3 hearing: “This proceeding, like the one that we had previously on the issue of whether . . . the settlement was procured by fraud or collusion, also was an unfocused and ineffective presentation .... I have gone over every document that’s been presented. And it is true, there are documents here . . . which, by centering ... on one or another sentence . . . can justify real suspicions .... But we have had four years of opportunity to translate those suspicions into a sound case, and I have yet to see that sound case presented on behalf of the plaintiffs.”

9. Other Claims of Objectors.

Objectors present a miscellany of additional attacks on the settlement agreement. First, they assert that a settlement benefit based on Southland’s assumption of the costs of the RIS computer billing system was “illusory” because the company was going to assume those substantial costs—valued at $25 million—in any case. The franchise agreements, however, permitted Southland to impose the costs on franchisees. The company, it is true, had underwritten certain pilot costs of the project, but did not plan to assume the costs in the future. The trial court, in any event, explored these claims of illusory benefits at length in the fairness hearings. We cannot say the outcome represented an abuse of discretion.

Next, objectors contend the trial court erred by placing the burden of proof on those opposing the settlement agreement. This appears to be in the nature of an ipse dixit argument, without support in the record. It is clear from the record that the trial court, at least, understood the settlement proponents had the burden of proof. As the Dunk court explained in rejecting a similar claim, *1166 “[Objectors] misapprehend[] [the class representatives’] burden. [They] made a sufficient showing which [objectors] failed to adequately rebut.” {Dunk, supra, 48 Cal.App.4th at p. 1800 .)

in.

Conclusion

Of the contemporary American class action, Judge Richard Posner of the Seventh Circuit Court of Appeals has written that it “differ[s] from ordinary lawsuits in that the lawyers for the class, rather than the clients, have all the initiative and are close to being the real parties in interest. This fundamental departure from the traditional pattern in Anglo-American litigation generates a host of problems well illustrated” by efforts to settle such litigation. {Mars Steel, supra, 834 F.2d at p. 678.) That observation is doubly borne out by the voluminous record in this case. It is troubling to pore over pages of transcripts of hearings before the trial judge, troubling to see the extent to which the class action has become the vehicle for prolonged and bitter conflict between attorneys representing the same “client.”

The absence of much substantive law to apply in such cases is equally troubling. Courts and commentators have pointed out that “the use of settlement classes . . . risks transforming the courts into mediation forums,” with a “natural hydraulic pressure” leading to settlement. {Pick-Up Truck Litigation, supra, 55 F.3d at p. 790 ; see also Coffee, The Corruption of the Class Action, Wall Street Journal (Sept. 7, 1994) p. A15 [“Trial judges are being converted from neutral umpires, adjudicating factual disputes, into problem-solving bureaucrats dispensing social justice”].) The tendency is to convert the class action lawsuit into a judicialized administrative process, with the trial judge serving as a kind of glorified mediator, resolving the endless disputes that arise between class counsel, shepherding the negotiation process, overseeing the mailing of notices, and so on. Objectors have presented a torrent of criticisms of the Southland settlement and the circumstances leading up to it. Underlying and driving these complaints is former counsels’ dissatisfaction with the quantity of attorneys’ fees awarded them by the trial court and preceding events that led to their discharge as class counsel.

Given this context, one aspect of the settlement process does appear salutary: the recognition by the appellate courts that their review function in such circumstances is gross at best and, given that “so many imponderables enter into the evaluation of a settlement” {Mars Steel, supra, 834 F.2d at p. 682), an abuse of discretion standard of appellate review is singularly *1167 appropriate. The trial judge in this case spent untold hours as a class action administrator and, on the whole, acquitted himself well. Having reviewed the lengthy record, we are satisfied that the settlement is fair, adequate, and reasonable, that a national class was properly certified, and that the settlement agreement was not the product of fraud or collusion. The judgment is affirmed; respondents to recover their costs.

Hanlon, P. J., and Woolard, J., * concurred.

A petition for a rehearing was denied January 29, 2001, and appellants’ petition for review by the Supreme Court was denied April 11, 2001.

In addition to this appeal, the 7-Eleven franchise litigation provoked several other appeals. These include an appeal from a summary judgment granted in favor of Southland and several soft drink bottlers on plaintiffs’ state antitrust claims; an appeal from a judgment dismissing an effort by plaintiffs to join as party defendants in this suit several Japanese citizens with ownership interests in Southland; and an appeal from an order of the Alameda County Superior Court preliminarily enjoining members of the Franklin group from continuing to prosecute two civil actions filed in the San Diego County Superior Court against the class representatives and class counsel seeking additional attorney’s fees. By the terms of the settlement agreement, all of the appeals save the injunction against the San Diego litigation are stayed to abide the outcome of this, appeal.

Dunk, supra, 48 Cal.App.4th at page 1801 .

Under the class action provisions of the Federal Rules of Civil Procedure, to which California courts look in the absence of state authority (Vasquez v. Superior Court (1971) 4 Cal.3d 800, 821 [ 94 Cal.Rptr. 796 , 484 P.2d 964 ]), rule 23(a) (28 U.S.C.) requires that before a suit can be certified as a class action, four criteria must be met: (1) numerosity, (2) commonality, (3) typicality, and (4) adequacy of representation. (See, e.g., Amchem Products, Inc. v. Windsor (1997) 521 U.S. 591 [ 117 S.Ct. 2231 , 138 L.Ed.2d 689 ] {Amchem).)

Judge of the San Francisco Superior Court, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution.