ABM Industries Overtime Cases ( 2018 )


Menu:
  • Filed 12/11/17; Certified for Publication 1/10/18 (order attached)
    IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA
    FIRST APPELLATE DISTRICT
    DIVISION FOUR
    ABM INDUSTRIES OVERTIME CASES                                    JCCP No. 4502
    A132387, A133077 & A133695
    (City & County of San Francisco
    Super. Ct. No. CJC-07-004502)
    Respondent ABM Industries, Inc. (collectively with related respondents, ABM) is
    a large facility services company with employees throughout the United States, including
    thousands of janitorial workers at hundreds of job sites in California. Appellants
    (referred to herein as plaintiffs) are present or former ABM janitorial employees. On
    behalf of themselves and similarly situated Californians, plaintiffs filed their complaint in
    this coordinated proceeding in September 2007, alleging that ABM violated California
    labor laws by, among other things, failing to properly record and compensate employees
    for meal breaks; requiring employees to work split shifts without appropriate
    compensation; and failing to ensure that employees were reimbursed for expenses
    incurred when traveling between work sites. In June 2010, plaintiffs moved for class
    certification of a general class of ABM workers and various subclasses of such workers
    who had been subjected to particular wage and hour violations. After briefing and
    argument, the trial court found plaintiffs’ expert evidence inadmissible and indicated
    orally that it was denying the class certification motion. In response, plaintiffs filed a
    1
    motion pursuant to Code of Civil Procedure section 473, subdivision (b) (the 473(b)
    motion), attempting to supplement the evidence previously provided with respect to the
    qualifications of their expert. By order dated June 29, 2011, the trial court denied
    plaintiffs’ 473(b) motion. Thereafter, on September 1, 2011, the trial court issued its
    written order, formally denying plaintiffs’ class certification motion. We conclude that
    the trial court’s wholesale exclusion of plaintiffs’ expert evidence in this case was error.
    We further determine that the trial court’s refusal to grant class certification on these facts
    was an abuse of discretion, and therefore reverse.
    I.       BACKGROUND
    A.     Facts Underlying the Consolidated Complaint
    ABM’s numerous California janitorial employees work at customers’ workplaces
    scattered throughout the state. ABM’s job sites in California are organized into two
    regions (Northern California and Southern California), various branches within a region,
    and dozens of distinct districts within a branch. A district is a number of buildings within
    a geographic area. Each branch is under the supervision of a different branch manager.
    Employees report to an individual site supervisor, who in turn reports to the branch
    manager. According to ABM, the site supervisor is responsible for “the daily operations
    of the location, including assurance that employees are paid properly and provided with
    their meal and rest breaks . . . .” However, it appears that ABM’s wage and hour policies
    are controlled centrally and thus applied uniformly throughout all janitorial job sites. In
    addition, ABM pays all of its employees through use of a single software application, the
    Labor Management System (LMS).
    ABM provides janitorial services to clients under contracts obtained through
    competitive bidding. According to ABM, “[t]he low cost of entry in the facility services
    business has led to strongly competitive markets comprised of a large number of mostly
    regional and local owner-operated companies, primarily located in major cities
    throughout the United States.” In order to compete, ABM provides various contracts at
    agreed-upon prices. For instance, ABM provides a fixed price contract where “the client
    agrees to pay a fixed fee every month over a specified contract term.” Under the cost-
    2
    plus arrangement, “the clients reimburse [ABM] for the agreed-upon amount of wages
    and benefits, payroll taxes, insurance charges and other expenses associated with the
    contracted work.” Given the fixed-price nature of these contracts, it is ABM, not the
    customer, who is responsible for higher labor costs if their employees cannot finish their
    assigned work within budgeted timeframes. As Faisal Algaheim, ABM’s Regional
    Operations Manager for Northern California, testified: “The customer paid the
    contracted amount; the contracted price. And if it is a fixed job—which means the
    customer will only pay us a contracted amount—whether we work more or less, it is our
    problem to maintain the cleaning specifications, and pay the employees currently.”
    All of ABM’s non-exempt janitorial employees, who provide the services under
    these contracts, are entitled to the benefits prescribed by California’s labor laws and the
    related wage orders promulgated by the Industrial Welfare Commission (Wage Orders).
    For instance, “[p]ertinent meal period provisions require that ‘[n]o employer shall employ
    any person for a work period of more than five (5) hours without a meal period of not less
    than 30 minutes . . . .’ (Cal. Code Regs., tit. 8, § 11050, subd. 11(A).) ‘[A]n employer’s
    obligation is to provide a first meal period after no more than five hours of work and a
    second meal period after no more than 10 hours of work.’ (Brinker [Restaurant Corp.
    v. Superior Court (2012)] 53 Cal.4th [1004,] 1049 [Brinker].) To qualify as a lawful
    meal break under California law, an employee must be relieved of all duties for an
    uninterrupted 30 minutes. (Id. at p. 1040; Cal. Code Regs., tit. 8, § 11050, subd. 11(A).)
    If an employer fails to comply with these requirements it must pay one hour of pay at the
    employee’s regular rate ‘for each workday that the meal period is not provided.’ (Cal.
    Code Regs., tit. 8, § 11050, subd. 11(B); see Lab. Code, § 226.7, subd. (c).)” (Alberts
    v. Aurora Behavioral Health Care (2015) 
    241 Cal.App.4th 388
    , 400 (Alberts).) We refer
    to any extra hours of wages potentially due to employees under the labor laws as
    premium pay.
    Similarly, pursuant to Wage Order 5-2001(4)(C): “When an employee works a
    split shift, one hour’s pay at the minimum wage shall be paid in addition to the minimum
    wage for that workday . . . .” (See Lab. Code, § 1197 [“The minimum wage for
    3
    employees fixed by the commission or by any applicable state or local law, is the
    minimum wage to be paid to employees, and the payment of a lower wage than the
    minimum so fixed is unlawful.”]; see also id., §§ 1194, subd. (a) & 1194.2 [allowing civil
    action for recovery of unpaid wages].) For purposes of this requirement, “split shift” is
    defined to mean “a work schedule which is interrupted by non-paid non-working periods
    established by the employer, other than bona fide rest or meal periods.” (Wage Order 5-
    2001(2)(R).) Although the Wage Order does not define “bona fide meal period,” the
    Division of Labor Standards Enforcement (DLSE) has historically taken the position that
    a bona fide meal period “is one that does not exceed one hour (60 minutes) in length.”
    (DLSE Of Counsel H. Thomas Cadell, Jr., letter to Paul K. Schrieffer, Dec. 11, 2002.)1
    Finally, California law requires employers to fully reimburse employees for
    expenses actually and necessarily incurred in the discharge of their duties, including
    automobile expenses. (Lab. Code, § 2802; Gattuso v. Harte-Hanks Shoppers, Inc. (2007)
    
    42 Cal.4th 554
    , 569 (Gattuso).) This right to reimbursement cannot be waived.
    (Gattuso, supra, 42 Cal.4th at p. 561.) However, an employer can discharge its
    reimbursement obligation in a number of different ways, including through
    reimbursement for actual expenses or mileage, or through lump sum payments. (Id. at
    pp. 567–571.)
    On September 19, 2007, plaintiffs filed their consolidated class action complaint
    in this matter (Complaint). The Complaint alleges numerous violations of California’s
    labor laws and Wage Orders, including violations related to missed meal periods, failure
    to provide mandatory split shift premium pay, and failure to compensate ABM
    employees for travel expenses incurred when travelling between job sites. The
    Complaint additionally alleges unfair competition under Business & Professions Code
    section 17200, based on the asserted labor law violations. Finally, it contains a claim
    1
    Although the DLSE is responsible for enforcing California’s labor laws,
    including Wage Orders, its interpretations of Wage Orders—while entitled to
    consideration and respect—are not binding. (Aleman v. Airtouch Cellular (2012)
    
    209 Cal.App.4th 556
    , 573 (Aleman).)
    4
    under the Private Attorneys General Act of 2004, Labor Code section 2698 et seq.
    (PAGA), to collect penalties based on ABM’s alleged systemic wrongdoing.
    B.     Class Certification Motion
    After a number of years of discovery and other preliminary matters, plaintiffs filed
    their motion for class certification on June 14, 2010. The motion sought certification of a
    general class described as “[a]ll non-exempt janitorial employees and former non-exempt
    janitorial employees employed by ABM in the State of California at any time from April
    6, 2002 to the present” (ABM Workers) (italics omitted). This putative class was
    estimated as of 2007 to include approximately 35,000 ABM janitorial employees. In
    addition, the motion proposed seven subclasses of ABM Workers, the following four of
    which are relevant here: (A) “ABM Workers who . . . suffered an automatic deduction of
    a half-hour although the employee actually worked through the deducted meal
    period . . .” (Unpaid Time/Meal Period Subclass); (B) “ABM Workers who were not
    paid premium meal period wages when they (1) worked shifts of at least five hours
    without an uninterrupted meal period of at least 30 minutes, (2) worked shifts of at least
    10 hours without a second uninterrupted meal period of at least 30 minutes, or (3) were
    provided a first meal period after the fifth hour of work” (Unpaid Meal Premium
    Subclass); (C) “ABM Workers who were scheduled or required in a workday to work
    two or more shifts separated by a period of time that was not a bona fide meal period, but
    were not paid an additional hour of wages for each split shift” (Unpaid Split-Shift
    Premium Subclass); and (D) “ABM Workers who were not reimbursed for expenses that
    were necessary to carry out their duties, including (1) the use of their own vehicles to
    travel between jobsites, or transport ABM supplies or equipment” (Reimbursement
    Subclass).2
    Plaintiffs argued that class certification was warranted because, among other
    reasons, common legal and factual issues predominated. For instance, plaintiffs alleged
    2
    Plaintiffs declined to appeal from the denial of class certification for their other
    three proposed subclasses—the Unpaid Rest Premium subclass, the Unpaid Reporting
    Time subclass, and the Paystub subclass.
    5
    that ABM applied a uniform payroll policy which compensated employees according to
    anticipated work schedules rather than for hours actually worked, leading to
    uncompensated time. In particular, according to plaintiffs, the LMS, ABM’s payroll
    system, automatically deducted 30 minutes of work time for a meal period whenever an
    employee was scheduled for a shift of five or more hours, without sufficient documentary
    evidence that those meals were actually taken. In addition, plaintiffs averred that analysis
    of the LMS disclosed a company policy of never paying statutorily required premium
    wages for missed meal periods or split shifts, despite the fact that some employees were
    scheduled to work split shifts and, reportedly, many routinely missed meals if they
    otherwise had insufficient time for cleaning. Finally, plaintiffs claimed that, although
    ABM scheduled route workers to provide janitorial services at different locations within
    the same workday—and required them to travel between sites—the LMS disclosed very
    few instances in which employees were reimbursed for expenses.
    According to plaintiffs, the legality of these common practices could most
    appropriately be decided on a classwide basis, and ABM’s computerized payroll records
    could be used both to identify violations and to establish common policies. In support of
    their motion, plaintiffs submitted declarations from 50 ABM Workers, including four
    named plaintiffs, stating that the schedules under which employees were paid often bore
    little relationship to the hours actually worked. For instance, they often worked through
    meal periods because there was too much work to do. In addition, plaintiffs provided
    evidence of company practices from various ABM supervisors and officials. Finally,
    plaintiffs also submitted expert declarations from Aaron Woolfson, a provider of database
    services who analyzed certain timekeeping and payroll data maintained by ABM with
    respect to its employees. For example, Woolfson determined that, of the 1,141,903 shifts
    greater than five hours that failed to show any time-out/time-in entries during the
    scheduled workday, 1,070,517 of those shifts (94 percent) nevertheless showed an
    automatic 30-minute meal period deduction. Further, there was no indication in the
    records that premium pay was ever provided for missed meal periods. In addition,
    although Woolfson identified 6,331 employees for whom ABM reported at least one shift
    6
    containing shift segments separated by more than one hour, there was no indication in the
    payroll records that split shift premium pay was ever provided. Finally, as stated above,
    analysis of the payroll records disclosed very few instances in which employees were
    reimbursed for travel expenses (12,834 checks to 826 employees out of the 6,396
    employees who worked 155,485 shifts at more than one job site).
    ABM opposed plaintiffs’ class certification motion, claiming that plaintiffs had
    failed to offer any “common evidence of a pattern or practice of wrongdoing.” Rather,
    ABM asserted, it promulgated its written meal policy both in its employee handbook and,
    as of late 2006, on timecards used by some employees.3 ABM also had a policy for
    travel reimbursement, and stated that its practice was not to schedule split shifts. With
    respect to payroll, ABM acknowledges that there is no data in the LMS that describes
    when a meal period is taken. Rather, the LMS shows the hours scheduled for each
    employee, by listing the scheduled start and end time for each shift. In addition, the LMS
    automatically deducts a 30-minute meal period when warranted due to the length of the
    scheduled shift. According to ABM, when employees have worked their regularly
    scheduled shifts, they are paid according to their schedule as listed on the LMS. In
    contrast, if an employee worked additional time, including through a meal break, the site
    supervisor was required to submit an exception report for input into the payroll system,
    showing that the employee worked different hours than scheduled.4 Under these
    3
    Specifically, the employee handbook stated: “If you are a non-exempt employee
    . . . you may receive at least one half hour time off as a meal period. Your supervisor
    schedules meal . . . periods.” The timecards, as of late 2006, stated more directly: “State
    law requires that you take a meal break of at least thirty (30) minutes whenever you work
    five consecutive hours or more in a day. The meal period must begin before you exceed
    five hours of work and you must sign in and out for your meal period.” According to the
    plaintiffs, however, timecards that recorded meal breaks were used by less than
    15 percent of ABM employees.
    4
    According to Woolfson, however, despite ABM’s “timesheet maintenance”
    policy, of the 1,836,083 time entries in the data he reviewed, only 5,625 (0.3 percent)
    contained any adjustments to pay. Moreover, at least one ABM manager testified that
    exception reports did not list missed meal breaks and that, in fact, supervisors were not
    required to report missed meals.
    7
    circumstances, ABM argued that class treatment was inappropriate because resolution of
    plaintiffs’ claims would turn on multiple individualized inquiries, such as whether and
    when each employee took lunch breaks, why an employee failed to take a lunch break,
    how many miles a particular employee drove between work sites, whether a split shift
    employee received total wages for that day less than the minimum wage that they would
    otherwise have been owed, and whether the employee requested a split shift.
    In support of its opposition, ABM submitted declarations from 14 current
    employees as well as excerpts from the depositions of certain of plaintiffs’ declarants.
    According to ABM, the deposition testimony of plaintiffs’ declarants cast “serious
    doubts” on their credibility. With respect to expert testimony, ABM did not provide its
    own expert, but argued generally that Woolfson’s expert declaration should not be
    considered. ABM also requested that the court take judicial notice of certain deposition
    testimony provided by Woolfson in another case.
    After hearing on April 19, 2011, the trial court issued its oral ruling denying class
    certification. As a preliminary matter, the court opined that the evidence submitted by
    Woolfson was inadmissible because his declarations failed to qualify him as an expert on
    anything material to the class certification motion. Although the court did not strike the
    Woolfson material, it concluded that it was not admissible “because it doesn’t prove
    anything.” When asked about the validity of the many factual findings set forth in the
    Woolfson declarations, the trial court responded that the question at hand was “whether
    or not a class should be certified” and that, in this regard, it was “not sufficient to ferret
    out individualized common questions.” With respect to Woolfson himself, the court
    found many of the statements regarding his expertise conclusory and thus believed that
    he had not “demonstrated that this court should accept him as a person with particular
    background, experience, skills, [or] expertise to differentiate him from the rest of the
    world so he should be accepted by this court as an expert.”5 Since the trial judge rejected
    5
    Although there was evidence in the record that Woolfson had previously
    qualified as an expert in both state and federal court, it was attached to an attorney
    declaration rather than incorporated into Woolfson’s own declaration. Under these
    8
    Woolfson as an expert, he was not qualified “to present to me opinions that are not
    generally understood by the rest of the world and to allow him to present hearsay material
    to rely on and to give me opinions.”
    On the merits, the trial court found certification inappropriate due to issues with
    the subclass definitions. In particular, the court appeared concerned that the subclasses
    were defined in terms of individuals who had been harmed, making class members
    unascertainable until the conclusion of the case. In addition, it concluded that the
    plaintiffs had failed to meet their burden to show that common issues of fact and law
    predominated over individual questions, “given the employment structure and the variety
    of circumstances that each worker finds him or her under.” The trial court further noted
    with regards to predominance that the number of declarations submitted by plaintiffs
    disclosing labor code violations was insufficient standing on its own to establish
    commonality. Rather, it believed “evidence besides declarations would have to be
    submitted to show a common practice.” However, when asked about whether the
    existence of ABM’s auto-deduct policy for meal periods was evidence showing a
    common issue sufficient to support certification, the trial court responded: “Your class
    definition is wrong. It’s not up to me to ferret through what you present and to see if I
    can craft a class somehow from what you are arguing.” In the end, the trial court opined:
    “[T]he concept is not whether we can find a common question here. There are plenty of
    common questions, but the question is whether the common questions predominate over
    individual questions so that it would be appropriate to utilize the class action mechanism.
    And it’s just a procedural device for the convenience and efficiency of the court and for
    the efficiency and economic self-interests of the litigants . . . . [¶] . . . Just because you
    might have a common question in here somewhere doesn’t mean it’s appropriate to have
    this case proceed as a class action.”
    Plaintiffs filed notices of appeal with respect to this oral ruling in June 2011 (case
    No. A132387).
    circumstances, the court found it to be hearsay and indicated, regardless: “I feel it’s my
    job to figure out whether somebody is an expert.”
    9
    C.     Motion for Relief Under Section 473, Subdivision (b)
    In the meantime, on May 11, 2011, following the oral denial of their class
    certification motion, plaintiffs filed their 473(b) motion, asking to augment the record
    with further evidence of Woolfson’s credentials and expertise. Plaintiffs sought
    introduction of this additional evidence in hopes that the trial court would accept
    Woolfson as an expert, reassess its ruling that Woolfson’s declarations were hearsay, and
    reconsider its class certification decision in light of Woolfson’s expert findings and
    conclusions.6 In support of their motion, plaintiffs argued that it was excusable neglect
    not to have made a more thorough demonstration of Woolfson’s qualifications prior to
    the hearing on the class certification motion because ABM had given no indication that it
    was raising a serious challenge to those qualifications. Specifically, ABM had failed to
    lodge a formal objection to the evidence, move to strike the declaration, depose Woolfson
    on his credentials or conclusions, and/or provide their own contrary expert opinion.
    Instead, ABM simply made a brief argument in its opposition papers that Woolfson was
    not qualified to analyze the data in question and that the opinions he offered were
    conclusory and based on common experience.7
    The trial court denied plaintiffs’ 473(b) motion after hearing on June 8, 2011.
    According to the court, plaintiffs had not shown grounds for relief under that statute. In
    particular, the trial court opined: “Well, the problem with your certification motion was
    discussed in great detail at the hearing on the motion, and the problem was multifaceted.
    It covered a full range of matters, none of which falls into the category of a technical
    6
    Code of Civil Procedure section 473, subdivision (b), provides in relevant part:
    “The court may, upon any terms as may be just, relieve a party or his or her legal
    representative from a judgment, dismissal, order, or other proceeding taken against him
    or her through his or her mistake, inadvertence, surprise, or excusable neglect.”
    7
    This was apparently in contrast to a previous case where counsel for ABM
    (representing a different party) had vigorously challenged the same Woolfson declaration
    by submitting written objections, filing rebuttal declarations by defense experts, and
    extensively cross-examining Woolfson at deposition. According to plaintiffs’ 473(b)
    motion, these efforts failed and Woolfson was nevertheless qualified as an expert in that
    case.
    10
    failing by the lawyer; in other words, the idea that the problem here is that the lawyers
    did something wrong and therefore I should relieve the parties from the lawyers’ mistake
    is not what happened here.”
    According to the trial court, the real issue in the case was that ABM Workers were
    not a group of workers that were susceptible to being treated as a class, and thus there
    were no predominant questions of fact or law. The trial court also reiterated its problems
    with the plaintiffs’ proposed class definitions. In the end, the court indicated that it had
    reviewed the supplemental evidence provided by plaintiffs and that—even if it agreed to
    consider it—it would not change the court’s view on certification. As the court opined:
    “Whether or not this witness is qualified to give the opinions, the opinions are not
    material to this case . . . .” A written order memorializing the trial court’s denial of the
    473(b) motion was filed on June 29, 2011, and a timely notice of appeal with respect to
    that order was filed on August 29, 2011 (case No. A133077).
    D.     Order Denying Class Certification
    The trial court’s written order denying class certification was ultimately filed on
    September 1, 2011. The court first reiterated its conclusion that the Woolfson
    declarations submitted by plaintiffs in support of their class certification motion were
    inadmissible, stating “there is no evidence that Mr. Woolfson is an expert in any area that
    is material to this case.” As discussed above, the record did contain two orders (from
    state and federal courts) certifying Woolfson as an expert. The trial court, however,
    found that the facts set forth in those orders were hearsay and concluded, regardless:
    “Whether Mr. Woolfson was accepted as an expert in state and federal court is
    immaterial: the Court does its own work regarding the admissibility of expert
    testimony.”
    With respect to the merits, the trial court first concluded that plaintiffs’ class
    definition was unworkable. It found plaintiffs’ general class definition permissible: “All
    non-exempt janitorial employees and former non-exempt janitorial employees employed
    by ABM in the State of California at any time from April 6, 2002 to the present.”
    However, noting that courts have rejected class claims when the class definition is simply
    11
    shorthand for persons possibly wronged by the defendant, the trial court found fault with
    the plaintiffs’ seven subclasses, opining that “defining the proposed class(es) by reference
    to the alleged injury or injuries sustained is a fatal defect, because the members of the
    class cannot be ascertained until the lawsuit is concluded.” According to the trial court,
    under such circumstances, “it is impossible to identify who is a member of the putative
    class, which makes it impossible to provide them with notice of the lawsuit, and which
    therefore also makes it impossible to determine who will be bound by the judgment.”
    The trial court also found fault with the fact that the sum of the seven subclasses did not
    add up to the entire general class.
    In addition to these ascertainability issues, the trial court also concluded that class
    treatment of plaintiffs’ claims was inappropriate because plaintiffs failed to demonstrate
    that common questions predominate over individual inquiries. In particular, the court
    found that plaintiffs did not provide sufficient evidence of a common scheme with
    respect to the negotiation of contracts which, by their terms, led to ABM employees
    being underpaid. In addition, the court determined that the declarations submitted by
    plaintiffs regarding claimed labor law violations were insufficient in number, “without
    additional evidence,” to demonstrate a common practice. In sum, the court opined that
    “consideration of all the factors relevant to class certification demonstrates that individual
    inquiries will predominate in determining all of the putative class members’ claims, and
    therefore class certification is not a superior method of resolving the instant case.”
    Following entry of the trial court’s written order denying class certification,
    appellants filed a third notice of appeal (case No. A133695). By order dated January 26,
    2012, the three cases were consolidated for all future proceedings in this court. In
    addition, at the parties’ request, we stayed the matter pending issuance by the Supreme
    Court of its decision in Brinker, supra, 
    53 Cal.4th 1004
    . Once the Supreme Court’s
    opinion in Brinker was final, a briefing schedule was set, and the matter is now before us
    for decision.
    12
    II.        DISCUSSION
    A.      Admissibility of Expert Evidence
    As a preliminary matter, we must address the trial court’s decision to disregard the
    declarations of plaintiffs’ expert, Woolfson, in making its class certification
    determination. As both parties have accurately asserted, we review a trial court’s ruling
    on the admissibility of expert evidence for abuse of discretion. (Garrett v. Howmedica
    Osteonics Corp. (2013) 
    214 Cal.App.4th 173
    , 187.) “ ‘However, the discretion to admit
    or exclude evidence is not unlimited. “The discretion of a trial judge is not a whimsical,
    uncontrolled power, but a legal discretion, which is subject to the limitations of legal
    principles governing the subject of its action, and to reversal on appeal where no
    reasonable basis for the action is shown.” ’ ” (Kotla v. Regents of University of
    California (2004) 
    115 Cal.App.4th 283
    , 291–292 (Kotla).) This is especially true when,
    as here, a trial court’s exercise of discretion “implicates a party’s ability to present its
    case.” (Sargon Enterprises, Inc. v. University of Southern California (2012) 
    55 Cal.4th 747
    , 773 (Sargon Enterprises); see Brown v. Colm (1974) 
    11 Cal.3d 639
    , 647 (Brown)
    [“the exclusion of the sole expert relied upon by a party because of an erroneous view of
    his qualifications is, in a case where expert testimony is essential, an abuse of discretion
    as a matter of law requiring reversal”].) Indeed, in this context, courts must “be cautious
    in excluding expert testimony” as the trial court’s gatekeeping goal “is simply to exclude
    ‘clearly invalid and unreliable’ expert opinion.” (Sargon Enterprises, at p. 772.)
    Should we determine in this case that an abuse of discretion has occurred, that
    conclusion alone is not sufficient to support reversal of the trial court’s certification
    decision. Rather, the “judgment of the trial court may not be reversed on the basis of the
    erroneous admission of evidence, unless that error was prejudicial.” (Grail
    Semiconductor, Inc. v. Mitsubishi Electric & Electronics USA, Inc. (2014)
    
    225 Cal.App.4th 786
    , 799 (Grail Semiconductor); see Code Civ. Proc., § 475.) Article
    VI, section 13, of the California Constitution further provides that “a judgment may not
    be set aside based on the erroneous admission of evidence ‘unless, after an examination
    of the entire cause, including the evidence, the court shall be of the opinion that the error
    13
    complained of has resulted in a miscarriage of justice.’ ” (Grail Semiconductor, supra,
    225 Cal.App.4th at p. 799; see Evid. Code, § 353.) “In civil cases, a miscarriage of
    justice should be declared only when the reviewing court, after an examination of the
    entire cause, including the evidence, is of the opinion that it is reasonably probable that a
    result more favorable to the appealing party would have been reached in the absence of
    the error.” (Grail Semiconductor, at p. 799.) Thus, our task on appeal is to determine
    whether an abuse of discretion has occurred and, if so, whether it is reasonably probable
    that a result more favorable to plaintiffs would have been obtained absent the error.
    As detailed above, the trial court in the present case based its decision to exclude
    Woolfson’s expert declarations on two separate grounds—that Woolfson had not
    properly established himself as an expert and that, regardless, the information that he
    presented via expert declaration was not material to this case. With respect to
    Woolfson’s expert qualifications, although the trial court acknowledged in its order
    denying class certification that Woolfson indicated an expertise “in creating, managing
    and analyzing large databases,” it rejected him as an expert, finding no evidence that he
    had “formal training or degrees that would qualify him as an expert to review the
    timekeeping and payroll data at issue.” The trial court also noted that “Mr. Woolfson’s
    declaration does not set forth any evidence that he holds certificates, has obtained any
    kind of college or other professional degree, belongs to any professional organizations,
    has published any articles, taught or has ever testified as an expert witness at trial.”
    Further, at the April 2011 hearing denying class certification, the trial court indicated that
    it believed the qualification information supplied by Woolfson in his expert declaration
    was too general to establish him as an expert. For example, the trial court stated:
    “[Woolfson] says he has extensive experience in creating, managing, and analyzing large
    databases, including, and then he lists a number of things. I have no idea what the term
    ‘extensive’ means. It looks to me like a conclusion that he hasn’t explained in any way.
    He doesn’t say how many years, how many assignments, what the nature of the
    assignments were, what the nature of his tasks were or anything of the like.” Similarly,
    the trial court noted that “without any detail” Woolfson stated that he “has provided
    14
    payroll and timekeeping database analysis for attorneys in numerous wage-hour cases.
    That does not communicate any specific facts of the type that is usually relied upon to
    qualify an expert.”
    While the better course of action in this case clearly would have been to provide
    the trial court with a more extensive explanation of the specifics of Woolfson’s expertise,
    we believe that, under the circumstances, the trial court erred by refusing to qualify
    Woolfson as an expert in database management and analysis based on the materials
    before it. “A person is qualified to testify as an expert if he has special knowledge, skill,
    experience, training, or education sufficient to qualify him as an expert on the subject to
    which his testimony relates.” (Evid. Code, § 720, subd. (a), italics added.) “Expertise, in
    other words, ‘is relative to the subject,’ and is not subject to rigid classification according
    to formal education or certification.” (People v. Ojeda (1990) 
    225 Cal.App.3d 404
    , 408.)
    Rather, an expert’s qualifications can be established in any number of different ways,
    including “a showing that the expert has the requisite knowledge of, or was familiar with,
    or was involved in, a sufficient number of transactions involving the subject matter of the
    opinion.” (Howard Entertainment, Inc. v. Kudrow (2012) 
    208 Cal.App.4th 1102
    , 1115
    (Howard Entertainment).) In sum, with respect to expert qualification, “[t]he
    determinative issue in each case must be whether the witness has sufficient skill or
    experience in the field so that his testimony would be likely to assist the jury in the search
    for the truth, and no hard and fast rule can be laid down which would be applicable in
    every circumstance.” (Brown, supra, 11 Cal.3d at p. 645; see Howard Entertainment, at
    p. 1115.)
    Once this threshold has been met, questions regarding the degree of an expert’s
    knowledge go more to the weight of the evidence presented than to its admissibility. (See
    People v. Tuggle (2012) 
    203 Cal.App.4th 1071
    , 1079–1080; see also Jordan v. Allstate
    Insurance Co. (2004) 
    116 Cal.App.4th 1206
    , 1217 [where expert declaration was
    sufficient to demonstrate “ ‘special’ ” knowledge of the subject matter, the “weight and
    value” of the expert opinion was a matter for the trier of fact].) Finally, the ability of an
    expert witness to testify as to either facts or opinions is limited to matters that are not
    15
    common knowledge. Thus, for example, “[e]xpert testimony as to facts may be
    necessary where the facts from which conclusions are to be drawn are peculiarly within
    the expert’s knowledge and are not a matter of common knowledge as to which an
    ordinary witness may competently testify.” (1 Witkin, Cal. Evid. (5th ed. 2012) Opinion
    Evidence, § 27, p. 638.) Similarly, expert opinion should be excluded “ ‘ “when ‘the
    subject of inquiry is one of such common knowledge that men of ordinary education
    could reach a conclusion as intelligently as the witness.’ ” ’ ” (Kotla, supra,
    115 Cal.App.4th at p. 291; see Evid. Code, § 801, subd. (a).)
    Here, Woolfson provided a declaration indicating that he was a founder of
    TelSwitch, Inc., a company which “builds and develops telephonic database service for
    several major telecommunications companies to manage their billing, as well as
    calculating and maintaining extensive databases related to the accurate calculation of the
    rates and rounding mechanisms used on telecommunications services.” Woolfson further
    declared that he was a managing partner of Merkt-Woolfson, a company which
    “produces billing and database mechanisms for banks to keep track of the paperwork that
    banks require to maintain mortgage and loan origination” and also provides “extensive
    database management services to both government and private industries,” including “the
    largest banks, military contractors, and publicly held telecommunications carriers where
    accuracy and accountability are a necessity.” Moreover, according to Woolfson, a
    “typical transaction load” for an “average database” maintained by his company was
    approximately one million records a day; he was “accustomed to, and comfortable with,
    working with a large amount of data across a variety of industries, including for litigation
    purposes”; and he had “extensive experience in creating, managing, and analyzing large
    databases,” including specifically timekeeping databases.
    Woolfson’s expert declaration additionally indicated that he had provided “payroll
    and timekeeping database analysis for attorneys in Northern and Southern California
    involving numerous wage and hour class action cases.” He then described the
    timekeeping records he had received from ABM—including, for example, “1,836,083
    Time Entries in Microsoft ExcelTM files covering 27,183 employees who performed
    16
    1,500,175 shifts of work at 5380 Job Site locations from 12-08-02 through 07-18-07”—
    and walked through his step-by-step analysis of those records. Indeed, Woolfson went so
    far as to set forth the specific Structured Query Language (SQL) queries he used to
    extract relevant information from ABM’s records.8
    We reiterate that additional information regarding the specifics of Woolfson’s
    expertise in matters relevant to this case would clearly have been preferable.9 However,
    8
    As stated above, in addition to Woolfson’s expert declarations, the plaintiffs
    submitted an attorney declaration that, among other things, attached two court orders
    from cases in which Woolfson had reportedly been qualified as an expert under similar
    circumstances. (See Avalos v. La Salsa, Inc. (Super. Ct. Santa Barbara County, 2010,
    JCCP No. 4488 (Avalos) [order dated May 17, 2010, granting in part and denying in part
    plaintiff’s motion for class certification]; see also Hines v. KFC U.S. Properties, Inc.
    (S.D. Cal., Oct. 22, 2010, No. 09-cv-02422-JM (POR) (Hines) [order granting in part and
    denying in part motion for class certification, which states at page 6 that “Mr. Woolfson
    is an expert in the compilation and analysis of databases, based upon his declaration
    which sets forth his qualifications and the methods and procedures adopted to analyze the
    data.”].) At the April 2011 hearing on class certification, the trial court initially indicated
    that it believed there was evidence in the record that Woolfson had not previously
    qualified as an expert. When plaintiffs’ counsel mentioned the two cases cited above as
    instances where Woolfson had been qualified, the court opined that the referenced court
    orders were hearsay and that, regardless, it chose not to take judicial notice of them
    because it believed it was the court’s job “to figure out whether somebody is an expert.”
    The court’s written order denying class certification reiterated this sentiment, stating that
    whether Woolfson was “accepted as an expert in state and federal court is immaterial; the
    Court does its own work regarding the admissibility of expert testimony.” We agree with
    both the trial court and ABM that one court is not required to adopt another court’s
    conclusion that an individual is an expert in a particular matter, although it may do so.
    (See Mora v. Big Lots Stores, Inc. (2011) 
    194 Cal.App.4th 496
    , 513–514.) However, that
    a proposed expert has been previously qualified seems, at the very least, relevant to
    another court’s subsequent qualification analysis and we question the trial court’s perhaps
    overly technical application of the hearsay rule when establishing prior expert
    qualification via judicially noticed court order. Nevertheless, we need not finally reach
    the issue, as we would find error here regardless of whether Woolfson’s history as a
    qualified expert is considered.
    9
    For instance, in a supplemental declaration filed in connection with plaintiffs’
    473(b) motion, Woolfson clarified that he had over 24 years of experience developing
    highly accurate database applications for companies such as Japan Telecom America,
    Experian, Bank of America, and JP Morgan Chase; that his database techniques were
    17
    we conclude that the materials submitted in advance of the April 2011 hearing on class
    certification in this case were sufficient to qualify Woolfson as an expert in database
    management and analysis, and that the trial court’s conclusion to the contrary was an
    abuse of discretion. In particular, we find that the trial court’s emphasis on formal
    education and membership in professional organizations was misplaced with respect to
    Woolfson’s stated expertise, given his clear familiarity with numerous, highly complex
    transactions in that subject matter. (See Howard Entertainment, supra, 208 Cal.App.4th
    at p. 1115.) Indeed, while admittedly not detailed, Woolfson’s declaration did indicate
    that he had “extensive experience” in database management and analysis, including
    statements that he held leadership positions in two database companies which serviced
    the “largest banks, military contractors, and publicly held telecommunications carriers”;
    that his company handled typical transaction loads of approximately one million records
    per day on “average” databases it maintained; and that he had previously provided
    payroll and timekeeping database analysis in numerous wage and hour class action cases
    in California. Moreover, although ABM did argue briefly before the trial court that
    Woolfson was not qualified to analyze the data at issue and that certain of his conclusions
    were overly broad and lacked sufficient factual foundation, ABM did not challenge the
    veracity of any of Woolfson’s qualifications as set forth in his declaration, nor did it
    contest even a single one of the myriad factual findings made by Woolfson during the
    course of his analysis.
    used by the federal government, including by the Department of Justice and the Patriot
    Missile Defense Training System; that he had a number of relevant professional
    certifications; that his authored works included an analysis of the merger between
    Continental Airlines and United Airlines that was presented to the Senate Judiciary
    Committee in 2010; that he had qualified as an expert in Avalos and Hines, both wage
    and hour class actions in which the courts relied on his analysis in granting class
    certification; and that he had been retained as an expert in over 40 cases (90 percent class
    actions) by both plaintiffs and defendants to analyze “timekeeping, payroll records,
    telephone call records, credit card records, reimbursement records, and travel records
    (e.g., gps data and locations where employees worked).”
    18
    Under these circumstances, plaintiffs’ evidence supporting Woolfson’s expert
    qualifications showed that he had “sufficient skill or experience” in the field of database
    management and analysis such that his declarations should have been considered by the
    trial court. (See Brown, supra, 11 Cal.3d at p. 645.) Nevertheless, the trial court chose to
    reject all of the information provided by Woolfson, despite the fact that, as we discuss
    further below, Woolfson’s analysis of the ABM database was central to plaintiffs’ class
    certification motion, and thus the trial court’s decision effectively foreclosed plaintiffs’
    ability to put on their case. (See Sargon Enterprises, supra, 55 Cal.4th at p. 773; Brown,
    supra, 11 Cal.3d at p. 647.) This was error.
    While we do not here pass on the admissibility of every opinion reached by
    Woolfson based on his manipulation of ABM’s database, we find the many facts
    generated by Woolfson’s analysis clearly admissible as matters beyond the common
    knowledge or experience of an ordinary witness.10 (See Business Objects, S.A.
    v. MicroStrategy, Inc. (Fed. Cir. 2005) 
    393 F.3d 1366
    , 1368 [noting that SQL requires the
    user to “understand the structure and content of the relational database as well as the
    complex syntax of the specific query language” and that “[t]hese complexities generally
    prevent laypersons from drafting queries in query languages.”].) Moreover, as evidence
    of ABM’s common wage and timekeeping practices, Woolfson’s results would
    unquestionably aid a jury in its search for the truth regarding any alleged classwide wage
    or hour violations in this case. (See Brown, supra, 11 Cal.3d at p. 645; Howard
    Entertainment, supra, 208 Cal.App.4th at p. 1115; see also Brinker, 
    supra,
     53 Cal.4th at
    p. 1033 [“Claims alleging that a uniform policy consistently applied to a group of
    employees is in violation of the wage and hour laws are of the sort routinely, and
    properly, found suitable for class treatment.”].)
    10
    As just one example, Woolfson identified ABM workers scheduled to work shift
    segments separated by more than one hour on the same day, without any indication of
    premium pay, through use of the following SQL queries: “1. select count (*) from
    workdata where SplitShiftViolation=‘YES’ and isWork =‘YES’ [¶] 2. select distinct
    reference from workdata where SplitShiftViolation =‘YES’ and iswork =‘YES’.”
    19
    Frankly, we are somewhat mystified by the trial court’s wholesale exclusion of the
    entirety of Woolfson’s evidence in this matter. Upon review, it appears that the trial
    court’s conclusions regarding the admissibility of the Woolfson materials were
    impermissibly tainted by its strong views with respect to the underlying merits of
    plaintiffs’ class certification motion—that is, that class certification was improper due to
    the individualized inquiries that would be required to establish which ABM employees, if
    any, had been harmed in this matter. This determination, moreover, appears to have been
    based, at least in part, on the mistaken notion that database analysis of timekeeping and
    payroll records cannot be used as a means to show common practices for purposes of
    class certification. Indeed, at the June 2011 hearing on plaintiffs’ 473(b) motion, through
    which plaintiffs were attempting to bolster Woolfson’s expert credentials, the trial court
    opined that ABM workers were “not susceptible to be treated as a class, period” and that
    the “basic problem” in the case was that “individualized analysis of working situations”
    would be needed “to understand why a worker may not have been given a lunch break.”
    Thus, in the opinion of the trial court: “Whether or not this witness is qualified to give
    the opinions, the opinions are not material to this case.” (Italics added.)
    In sum, it was error for the trial court to completely disregard plaintiffs’ proffered
    expert evidence of common practice, rather than accepting it for what it was and
    weighing it against the existence of any individualized inquiries that might properly have
    defeated plaintiffs’ request for class certification. Moreover, as we discuss in detail
    below, the trial court’s decision clearly prejudiced plaintiffs, as it left them without any
    evidence of systemic wrongdoing other than the information contained in the declarations
    and deposition testimony submitted in connection with their class certification motion,
    materials which the trial court found insufficient in number to demonstrate predominant
    common questions. Indeed, the trial court expressly stated: “The number of declarations
    [submitted by plaintiffs] compared to the number of employees by itself would not be
    sufficient [to demonstrate predominance]. . . . What I’m suggesting is that evidence
    besides declarations would have to be submitted to show a common practice.” (Italics
    added.) Yet this was precisely the evidence that the trial court excluded. Since we find it
    20
    reasonably probable that a result more favorable to the plaintiffs would have been
    reached in the absence of this error, the trial court’s order denying class certification
    cannot stand. (See Kotla, supra, 115 Cal.App.4th at p. 294.)11
    B.     Class Certification Issues
    1.     Rules Governing Class Actions and Standard of Review
    The requirements for class certification are well established and were recently
    summarized by our high court in Brinker, 
    supra,
     
    53 Cal.4th 1004
    , 1021: “Originally
    creatures of equity, class actions have been statutorily embraced by the Legislature
    whenever ‘the question [in a case] is one of a common or general interest, of many
    persons, or when the parties are numerous, and it is impracticable to bring them all before
    the court . . . .’ [Citations.] Drawing on the language of Code of Civil Procedure section
    382 and federal precedent, we have articulated clear requirements for the certification of
    a class. The party advocating class treatment must demonstrate the existence of an
    ascertainable and sufficiently numerous class, a well-defined community of interest, and
    substantial benefits from certification that render proceeding as a class superior to the
    alternatives. [Citations.] ‘In turn, the “community of interest requirement embodies
    three factors: (1) predominant common questions of law or fact; (2) class representatives
    with claims or defenses typical of the class; and (3) class representatives who can
    adequately represent the class.” ’ ”
    “ ‘[T]his state has a public policy which encourages the use of the class action
    device.’ ” (Sav-On Drug Stores, Inc. v. Superior Court (2004) 
    34 Cal.4th 319
    , 340
    (Sav-On).) Further, whether class certification should be granted is a procedural
    question, and not a question of whether the action is “ ‘legally or factually meritorious.’ ”
    (Id. at p. 326.) As a general matter, “ ‘a class action is not inappropriate simply because
    11
    Because we conclude that the trial court erred in failing to consider Woolfson’s
    expert declarations when making its class certification determination at the hearing in
    April 2011 (as memorialized by the court’s order denying plaintiffs’ motion for class
    certification filed on September 1, 2011), we need not reach the issue of whether the trial
    court also erred in refusing to grant plaintiffs’ 473(b) motion so that additional evidence
    of Woolfson’s expert qualifications could be brought before the court.
    21
    each member of the class may at some point be required to make an individual showing
    as to his or her eligibility for recovery or as to the amount of his or her damages.’ ” (Id.
    at p. 333.)
    In addition, with respect to the superiority of the class action mechanism—and as
    is pertinent to our present inquiry—we have previously noted that “[c]ourts regularly
    certify class actions to resolve wage and hour claims. [Citations.] In this arena the class
    action mechanism allows claims of many individuals to be resolved at the same time,
    eliminates the possibility of repetitious litigation and affords small claimants with a
    method of obtaining redress for claims which otherwise would be too insignificant to
    warrant individual litigation. [Citation.] Moreover, the issues slated for contest are
    primarily common issues involving common evidence. It would not be efficient or fair to
    relegate these complaints to multiple trials.” (Bufil v. Dollar Financial Group, Inc.
    (2008) 
    162 Cal.App.4th 1193
    , 1208 (Bufil); see Brinker, 
    supra,
     53 Cal.4th at p. 1033.)
    Indeed, as our high court elaborated in Brinker, a theory of liability that an
    employer “has a uniform policy, and that that policy, measured against wage order
    requirements, allegedly violates the law—is by its nature a common question eminently
    suited for class treatment.” (Brinker, 
    supra,
     53 Cal.4th at p. 1033.) “[I]n the general case
    to prematurely resolve such disputes, conclude a uniform policy complies with the law,
    and thereafter reject class certification . . . places defendants in jeopardy of multiple class
    actions, with one after another dismissed until one trial court concludes there is some
    basis for liability and in that case approves class certification. [Citation.] It is far better
    from a fairness perspective to determine class certification independent of threshold
    questions disposing of the merits, and thus permit defendants who prevail on those
    merits, equally with those who lose on the merits, to obtain the preclusive benefits of
    such victories against an entire class and not just a named plaintiff.” (Id. at p. 1034.)
    “California courts consider ‘pattern and practice evidence, statistical evidence,
    sampling evidence, expert testimony, and other indicators of a defendant’s centralized
    practices in order to evaluate whether common behavior towards similarly situated
    plaintiffs makes class certification appropriate.’ ” (Jaimez v. Daiohs USA, Inc. (2010)
    22
    
    181 Cal.App.4th 1286
    , 1298 (Jaimez).) Other relevant factors include “ ‘whether the
    class approach would actually serve to deter and redress alleged wrongdoing.’ ” (Ibid.)
    Moreover, in the wage and hour context, “[w]e have recognized that retaining one’s
    employment while bringing formal legal action against one’s employer is not ‘a viable
    option for many employees,’ ” and thus a class action may be appropriate as “a current
    employee who individually sues his or her employer is at greater risk of retaliation.”
    (Gentry v. Superior Court (2007) 
    42 Cal.4th 443
    , 459, abrogated on other grounds as
    stated in Iskanian v. CLS Transportation Los Angeles, LLC (2014) 
    59 Cal.4th 348
    , 359–
    360; see also Williams v. Superior Court (2017) 
    3 Cal.5th 531
    , 558, citing Gentry.) And,
    in Gentry our high court noted that class actions may be particularly useful for immigrant
    workers with limited English language skills, as illegal employer conduct might
    otherwise escape their attention. (Gentry, at p. 461.)
    A ruling on class certification is reviewed for abuse of discretion. (Brinker, supra,
    53 Cal.4th at p. 1022; Sav-On, 
    supra,
     34 Cal.4th at p. 326.) Under this standard, “ ‘[a]
    certification order generally will not be disturbed unless (1) it is unsupported by
    substantial evidence, (2) it rests on improper criteria, or (3) it rests on erroneous legal
    assumptions.’ ” (Brinker, 
    supra,
     53 Cal.4th at p. 1022; see Bufil, supra, 162 Cal.App.4th
    at p. 1204 [noting that while “[t]rial courts enjoy wide discretion with regard to class
    certification,” we will nevertheless reverse and order denying class certification “if the
    order is based on improper criteria or incorrect assumptions”].) Moreover, “[a]n appeal
    from an order denying class certification presents an exception to customary appellate
    practice by which we review only the trial court’s ruling, not its rationale. If the trial
    court failed to conduct the correct legal analysis in deciding not to certify a class action,
    ‘ “an appellate court is required to reverse an order denying class certification . . . , ‘even
    though there may be substantial evidence to support the court’s order.’ ” ’ [Citation.] In
    short, we must ‘ “consider only the reasons cited by the trial court for the denial, and
    ignore other reasons that might support denial.” ’ ” (Alberts, supra, 241 Cal.App.4th at
    p. 399.)
    23
    On the issue of predominance, a trial court’s finding is generally reviewed for
    substantial evidence. (Brinker, 
    supra,
     53 Cal.4th at p. 1022.) Thus, “[w]e must
    ‘[p]resum[e] in favor of the certification order . . . the existence of every fact the trial
    court could reasonably deduce from the record . . . .’ ” (Ibid.) However, since the focus
    of this type of certification dispute “is on what type of questions—common or
    individual—are likely to arise in the action, rather than on the merits of the case
    [citations], in determining whether there is substantial evidence to support a trial court’s
    certification order, we consider whether the theory of recovery advanced by the
    proponents of certification is, as an analytical matter, likely to prove amenable to class
    treatment. [Citations.] ‘Reviewing courts consistently look to the allegations of the
    complaint and the declarations of attorneys representing the plaintiff class to resolve this
    question.’ ” (Sav-On, 
    supra,
     34 Cal.4th at p. 327, italics added.)
    In the instant case, plaintiffs contend that the trial court abused its discretion, both
    in concluding that their proposed subclasses are not ascertainable and in determining that
    common issues do not predominate over individual inquiries. We consider each claim in
    turn.
    2.     Ascertainability
    As stated above, the trial court refused to certify this matter as a class action
    because, among other reasons, it believed the subclasses proposed by appellants were not
    ascertainable. In particular, the trial court opined that defining the proposed subclasses
    by reference to the alleged Labor Code violations sustained was a “fatal defect,” because
    the putative subclass members could not be identified without a determination on the
    merits of each class member’s case. The court reasoned that, when the “class definition
    is simply shorthand for persons possibly wronged by the defendant,” it is impossible to
    identify putative class members until the lawsuit is concluded, making it impossible both
    to provide appropriate notice and to determine who will be bound by the judgment. The
    trial court also found fault with the fact that the sum of the seven subclasses did not add
    up to the entire general class of non-exempt janitorial employees, because some members
    of the general class may not have suffered any harm. Unsurprisingly, ABM agrees with
    24
    the trial court on appeal, arguing that “the only means by which the many subclasses
    could be ascertained was by a trial on the merits, requiring the testimony of each and
    every putative class member on each and every claim, a concept that is antithetical to the
    very concept of class litigation.” In our opinion, however, both the trial court and ABM
    have fundamentally misapprehended the concept of ascertainability as it applies to the
    circumstances of this case.
    “Ascertainability is achieved ‘by defining the class in terms of objective
    characteristics and common transactional facts making the ultimate identification of class
    members possible when that identification becomes necessary.’ ” (Bomersheim v. Los
    Angeles Gay & Lesbian Center (2010) 
    184 Cal.App.4th 1471
    , 1483; see Nicodemus
    v. Saint Francis Memorial Hospital (2016) 
    3 Cal.App.5th 1200
    , 1212 (Nicodemus);
    Aguirre v. Amscan Holdings, Inc. (2015) 
    234 Cal.App.4th 1290
    , 1300 (Aguirre).) “In
    determining whether a class is ascertainable, the trial court examines the class definition,
    the size of the class and the means of identifying class members.” (Bufil, supra,
    162 Cal.App.4th at p. 1207, italics added.) Thus, a plaintiff is not required to establish
    the identity of class members at the class certification stage of the proceedings. (Reyes
    v. Board of Supervisors (1987) 
    196 Cal.App.3d 1263
    , 1274.)
    Moreover, “[w]hile often it is said that ‘[c]lass members are “ascertainable” where
    they may be readily identified without unreasonable expense or time by reference to
    official records’ [citations], that statement must be considered in light of the purpose of
    the ascertainability requirement.” (Medrazo v. Honda of North Hollywood (2008)
    
    166 Cal.App.4th 89
    , 101 (Medrazo).) “ ‘Ascertainability is required in order to give
    notice to putative class members as to whom the judgment in the action will be res
    judicata.’ ” (Aguirre, supra, 234 Cal.App.4th at p. 1300.) Therefore, “[t]he goal in
    defining an ascertainable class ‘is to use terminology that will convey “sufficient
    meaning to enable persons hearing it to determine whether they are members of the class
    plaintiffs wish to represent.” [Citation.] “ . . . Otherwise, it is not possible to give
    adequate notice to class members or to determine after the litigation has concluded who is
    barred from relitigating.” ’ ” (Id. at pp. 1300–1301; see also Medrazo, at p. 101
    25
    [ascertainability requirement is satisfied if “the potential class members may be identified
    without unreasonable expense or time and given notice of the litigation, and the proposed
    class definition offers an objective means of identifying those persons who will be bound
    by the results of the litigation”].)
    In sum, a class is ascertainable if a plaintiff supplies a reasonable means of
    identifying potential class members and the class is defined in terms of objective
    characteristics and common transactional facts sufficient to allow a class member to
    identify himself or herself as having a right to recover based on that description. So long
    as these requirements are met, a class is ascertainable “even if the definition pleads
    ultimate facts or conclusions of law.” (Hicks v. Kaufman and Broad Home Corp. (2001)
    
    89 Cal.App.4th 908
    , 915–916 (Hicks); see Faulkinbury v. Boyd & Associates, Inc. (2013)
    
    216 Cal.App.4th 220
    , 226, 240–241 (Faulkinbury) [directing certification of subclasses
    based on meal break, rest break, and overtime violations]; Jaimez, supra,
    181 Cal.App.4th at pp. 1291–1292, 1295–1296 [directing certification of classes found
    ascertainable by the trial court, including a meal break class “based upon the failure to
    permit or authorize meal breaks and the failure to pay one hour of wages for each meal
    break violation” and an overtime class “based upon the failure to pay overtime to the
    class”]; Ghazaryan v. Diva Limousine, Ltd. (2008) 
    169 Cal.App.4th 1524
    , 1529, 1539
    [directing certification of two subclasses based on failure to pay earned wages and
    overtime and failure to provide mandatory rest breaks].) Under this established analytical
    framework—and when one considers the data supplied by Woolfson—the trial court’s
    conclusion that the proposed subclasses in this case are unascertainable due to the need
    for individualized merit determinations is simply not defensible.
    Indeed, we recently considered and rejected a similar argument in Nicodemus,
    supra, 
    3 Cal.App.5th 1200
    . In that case, the plaintiff filed an action alleging that she was
    overcharged for copies of her patient medical records, which were sought in anticipation
    of litigation by her attorney pursuant to Evidence Code section 1158. The named
    defendants were the plaintiff’s hospital (Saint Francis) and HealthPort Technologies,
    LLC (HealthPort), a company that, during the relevant timeframe, provided Saint Francis
    26
    with patient medical record release of information services pursuant to a contract.
    (Nicodemus, at pp. 1205, 1207.) The plaintiff moved for certification of a class
    comprised of all patients who requested medical records from a California medical
    provider through an attorney prior to litigation and who were charged by HealthPort more
    than the statutory maximum set forth in Evidence Code section 1158. (Id. at pp. 1205–
    1206, 1208.) The trial court concluded that the plaintiff’s proposed class was
    unascertainable. (Id. at p. 1210.) Although HealthPort tracked all attorney requests using
    a separate billing code in its database, the trial court concluded that the data set was over-
    inclusive because the plaintiff “had not presented a mechanism for determining whether
    attorneys’ requests were submitted ‘ “prior to litigation” . . . without individualized
    inquiry, for example, by asking’ each attorney.” (Ibid.)
    On appeal, we concluded that the trial court erred as a matter of law in finding that
    the proposed class was not ascertainable. (Nicodemus, supra, 3 Cal.App.5th at pp. 1213–
    1217.) Because it is highly relevant to the case at hand, we set out our reasoning in some
    detail: “[E]ven assuming the attorney request data set does include some unknown
    number of requests that were submitted after litigation was commenced (or after
    defendants’ first appearance) or for reasons unrelated to litigation, this fact would not
    defeat ascertainability. HealthPort argued, and the trial court concluded, that a class is
    not ascertainable if the class members who are entitled to recover from the defendants
    cannot be identified without an individualized inquiry. That is not, however, the standard
    for determining whether a class is ascertainable. As noted, ‘[a]scertainability is required
    in order to give notice to putative class members as to whom the judgment in the action
    will be res judicata. [Citations.] . . . As long as the potential class members may be
    identified without unreasonable expense or time and given notice of the litigation, and the
    proposed class definition offers an objective means of identifying those persons who will
    be bound by the results of the litigation, the ascertainability requirement is met.’
    (Medrazo, supra, 166 Cal.App.4th at p. 101.) Plaintiff here has identified the class in
    terms of objective characteristics, tracking the provisions of section 1158; if it is
    determined later in the litigation that the ‘07’ data set includes requests not made
    27
    pursuant to section 1158, ‘those [persons] can be eliminated from the class at that time.’
    (Aguiar v. Cintas Corp. No. 2, supra, 144 Cal.App.4th at p. 136 (Aguiar); see also Sav-
    On, 
    supra,
     34 Cal.4th at p. 333 [‘ “a class action is not inappropriate simply because each
    member of the class may at some point be required to make an individual showing as to
    his or her eligibility for recovery” ’]; Bell v. Farmers Ins. Exchange (2004)
    
    115 Cal.App.4th 715
    , 743 [class of all employees in certain job categories ascertainable
    even though some employees may not have worked overtime and thus may not be
    entitled to any recovery].) Nor should a court ‘decline to certify a class simply because it
    is afraid that insurmountable problems may later appear at the remedy stage.’ ”
    (Nicodemus, supra, 3 Cal.App.5th at p. 1214.) Thus, contrary to the trial court’s belief,
    possible over-inclusiveness in the method proposed for identifying potential class
    members does not defeat ascertainability.
    In reaching our conclusion in Nicodemus, we distinguished Hale v. Sharp
    Healthcare (2014) 
    232 Cal.App.4th 50
     (Hale)—a case relied on by ABM here—in which
    a class was decertified after nearly three years of litigation, discovery, and notice to
    potential class members. (Id. at pp. 53–55; see Nicodemus, supra, 3 Cal.App.5th at
    pp. 1215–1216.) Hale involved an allegation that a class of persons who self-paid for
    emergency room treatment were overcharged when compared to insured persons. (Hale,
    at p. 53.) In moving to decertify, the defendant argued that the class was not
    ascertainable because the defendant did not keep records in such a way “as to reasonably
    and readily identify those included in the class definition without individualized
    inquiries.” (Id. at p. 55.) The trial court agreed with the defendant and the appellate
    court affirmed, opining with respect to ascertainability that “[i]t is the inability to
    reasonably distinguish those individuals [later determined to qualify for coverage] from
    individuals who were actually uninsured and then to identify any disparity in amounts
    paid that make it unreasonable to ascertain the defined class.” (Ibid.) In Nicodemus, we
    distinguished Hale, both because of its “distinctive procedural posture” and because,
    under Hale’s facts, “it was indisputably demonstrated that there was simply no way to
    28
    avoid a complicated individualized inquiry to determine not just eligibility for damages
    but to prove liability.” (Nicodemus, supra, 3 Cal.App.5th at p. 1216, italics added.)
    In contrast, we found Bufil, supra, 
    162 Cal.App.4th 1193
    , instructive. In Bufil,
    which involved meal and rest break claims, “[t]he proposed class was defined as
    employees for whom the defendant’s records showed a meal period not taken because the
    employee was the only person in the store or was the only person present except for a
    trainee.” (Nicodemus, supra, 3 Cal.App.5th at p. 1216.) “Although employees who
    missed a meal period could be identified from the defendant’s records, employees who
    missed a rest period could not.” (Ibid.) However, Bufil submitted evidence that the
    defendant had a policy that hourly employees who were working alone or only with a
    trainee were not allowed to go off duty for any type of break, and argued that the records
    identifying class members who missed meal periods for the reasons specified thus also
    identified those who missed rest breaks. (Bufil, at pp. 1206, 1208.) Under these
    circumstances, the appellate court reversed the trial court’s denial of class certification,
    concluding that “the class was ascertainable from the defendant’s records.” (Nicodemus,
    at p. 1216.) “In doing so, the court rejected the defendant’s ‘speculation’ that an
    employee who missed a meal break nonetheless might have received a rest break,
    observing ‘speculation that goes to the merits of ultimate recovery [was] an inappropriate
    focus for the ascertainability inquiry.’ (Ibid., citing Medrazo, supra, 166 Cal.App.4th at
    p. 101 [defendant’s sales records offered an objective means of identifying potential class
    members, and plaintiff’s inability at the class certification stage to identify precisely
    which buyers qualified as class members was “irrelevant”] and Harper v. 24 Hour
    Fitness, Inc. (2008) 
    167 Cal.App.4th 966
    , 976 [“the need to individually examine each
    member’s contract to ultimately determine whether he or she qualifies for inclusion in the
    class does not . . . demonstrate a lack of ascertainability or manageability”].)
    Adopting this analysis in Nicodemus, we concluded that potential class members
    could be readily identified by reference to HealthPort’s attorney request data set, and the
    “speculation” that the data set might be over-inclusive went “to the merits of each class
    member’s recovery” and thus “was an inappropriate focus of the ascertainability inquiry.”
    29
    (Nicodemus, supra, 3 Cal.App.5th at p. 1216.) Our analysis of ascertainability in the
    present case mirrors our conclusions in Nicodemus. Here, as established by Woolfson,
    the potential subclass members are all readily identifiable by reference to ABM’s own
    employment and payroll records.
    For instance, the subclass of ABM Workers who “suffered an automatic deduction
    of a half-hour although the employee actually worked through the deducted meal period”
    can be identified through ABM’s timekeeping and payroll records showing numerous
    instances where a meal deduction was made for a shift without any corresponding time
    entry indicating that a meal period was taken. The subclass of ABM Workers who were
    not paid premium meal period wages when they worked shifts of a particular length
    without a recorded meal period can similarly be ascertained through reference to the
    same records, reviewed to determine whether any required premium wages were paid
    where no meal period was recorded. The Unpaid Split-Shift Premium Subclass— “ABM
    Workers who were scheduled or required in a workday to work two or more shifts
    separated by a period of time that was not a bona fide meal but were not paid an
    additional hour of wages for each split shift”—can be identified by examining ABM’s
    timekeeping and payroll records to determine which employees worked two or more
    shifts in the same day separated by more than an hour, but were not paid premium wages
    related to the split shift(s).12 Finally, members of the Reimbursement Subclass—ABM
    Workers who were not reimbursed for expenses related to the use of their own vehicles
    for travel between jobsites—can be identified by searching ABM payroll records to
    determine which employees worked at multiple jobsites separated by a certain baseline
    number of miles during the same workday, but did not receive reimbursement for
    travel.13
    12
    As discussed above, the DLSE has historically taken the position that a bona
    fide meal period is one that does not exceed one hour in length. (See ante at p. 4 &
    fn. 1.)
    13
    Woolfson opined below that the ABM databases contained location information
    for each worker’s shift and that from this information, along with the addresses of the
    30
    In addition, the subclasses are all defined using objective characteristics and
    common transactional facts sufficient to allow a potential class member to identify
    himself or herself as having a right to recover pursuant to that subclass. For example, the
    nonexempt ABM workers who would receive notice as part of the general class would all
    be aware whether they worked though meal periods, failed to receive reimbursement for
    their travel expenses between worksites, or otherwise fell within the articulated
    subclasses. Under these circumstances, ABM’s speculation that some potential class
    members identified in the data may ultimately not be entitled to relief—because, perhaps,
    they actually took an otherwise unrecorded meal, or were not entitled to a split shift
    premium on a particular day, or did not drive themselves between job sites—goes to the
    merits of each class member’s recovery and, as such, was an inappropriate focus of the
    trial court’s ascertainability inquiry. (See Sav-On, 
    supra,
     34 Cal.4th at p. 338 [class can
    be certified based on partial commonality, meaning not every single member of the
    proposed class needs to be exposed to the wrongful practice nor does the practice have to
    be unlawful or lawful as to every class member].)14
    3.     Predominance.
    Having determined that the plaintiffs have proposed ascertainable classes, we must
    next address the trial court’s conclusion that class certification was inappropriate in this
    matter because individual inquiries predominate over common questions. As mentioned
    above, “[t]he ‘ultimate question’ the element of predominance presents is whether ‘the
    issues which may be jointly tried, when compared with those requiring separate
    adjudication, are so numerous or substantial that the maintenance of a class action would
    work locations, it would be possible to calculate the mileage each worker traveled each
    day.
    14
    In this regard, we note additionally that “if necessary to preserve the case as a
    class action, the court itself can and should redefine the class where the evidence before it
    shows such a redefined class would be ascertainable.” (Hicks, supra, 89 Cal.App.4th at
    p. 916.) Thus, as this action progresses, the trial court should be open to making
    modifications to the class definitions as necessary to avert developing certification
    problems or to otherwise enhance the efficiencies of the class certification model.
    31
    be advantageous to the judicial process and to the litigants.’ [Citations.] The answer
    hinges on ‘whether the theory of recovery advanced by the proponents of certification is,
    as an analytical matter, likely to prove amenable to class treatment.’ [Citation.] A court
    must examine the allegations of the complaint and supporting declarations [citation] and
    consider whether the legal and factual issues they present are such that their resolution in
    a single class proceeding would be both desirable and feasible. ‘As a general rule if the
    defendant’s liability can be determined by facts common to all members of the class, a
    class will be certified even if the members must individually prove their damages.’ ”
    (Brinker, supra, 53 Cal.4th at pp. 1021–1022, fn. omitted.) Indeed, “at the class
    certification stage, as long as the plaintiff’s posited theory of liability is amenable to
    resolution on a classwide basis, the court should certify the action for class treatment
    even if the plaintiff’s theory is ultimately incorrect at its substantive level, because such
    an approach relieves the defendant of the jeopardy of serial class actions and, once the
    defendant demonstrates the posited theory is substantively flawed, the defendant
    ‘obtain[s] the preclusive benefits of such victories against an entire class and not just a
    named plaintiff.’ ” (Hall v. Rite Aid Corp. (2014) 
    226 Cal.App.4th 278
    , 293–294, italics
    omitted.)
    In short, when analyzing the element of predominance for purposes of class
    certification “the focus must be on the policy the plaintiffs are challenging and whether
    the legality of that policy can be resolved on a classwide basis.” (Lubin v. The
    Wackenhut Corp. (2016) 
    5 Cal.App.5th 926
    , 940.) Thus, for example, in Morgan v. Wet
    Seal, Inc. (2012) 
    210 Cal.App.4th 1341
    , the Court of Appeal affirmed the trial court’s
    denial of class certification in a case alleging that the company required employees to
    purchase company clothing to wear to work but failed to reimburse such purchases.
    Because there were no clear companywide policies requiring employees to purchase
    company clothing as a condition of employment or describing what an employee was
    required to wear, the trial court determined there was no common method to prove the
    fact of liability on a classwide basis. Rather, individualized inquiries would need to be
    made regarding, among other things, what employees were told by store managers about
    32
    wardrobe, how employees interpreted any such discussion, and what each store manager
    actually required employees to purchase. (Id. at pp. 1356–1357.)
    In contrast, numerous other cases have held that individualized issues regarding
    proof of the amount of damages class members may recover does not defeat a class action
    so long as there are common questions of liability amenable to class resolution. (See,
    e.g., Faulkinbury, supra, 
    216 Cal.App.4th 220
    , 232–240 [common issues of fact
    predominated for subclasses related to meal, rest, and overtime violations because
    liability could be determined classwide based on uniform policies, or lack thereof;
    individual issues, such as whether individuals took rest breaks, went to the issue of
    damages and did not preclude class certification]; Jones v. Farmers Ins. Exchange (2013)
    
    221 Cal.App.4th 986
    , 997 [a uniform policy denying compensation for preshift work
    presented predominantly common issues of fact and law because liability depended on
    the existence of the uniform policy, rather than individual damages determinations];
    Benton v. Telecom Network Specialists, Inc. (2013) 
    220 Cal.App.4th 701
    , 726 [theory
    that defendant violated wage and hour requirements by failing to adopt meal and rest
    break policies is amenable to class treatment; whether employee was able to take required
    breaks goes to damages].)
    The common theme in these cases is that the plaintiff’s theory of liability could be
    determined based on common uniform policies applicable to the class as a whole. (See
    also Department of Fish & Game v. Superior Court (2011) 
    197 Cal.App.4th 1323
    , 1356
    [“Class treatment is not barred where a single wrongful act has different effects on
    different claimants such that some may have claims while others may not. ‘ “In such
    cases, the Courts will generally certify a class if the defendant’s action can be found to be
    wrong in the abstract even if no individual person has been damaged. [Citations.] These
    situations are distinguishable from situations where the Court cannot determine the
    wrongfulness of an action without reference to individuals.” ’ ”].)
    In line with this precedent, and of particular relevance to the case at hand, is the
    Second District’s opinion in Jaimez, supra, 
    181 Cal.App.4th 1286
    . In Jaimez—a case,
    like this one, involving claims of various wage and hour violations, including meal break
    33
    issues—the appellate court concluded that the “trial court misapplied the criteria [for
    determining whether a class should be certified], focusing on the potential conflicting
    issues of fact or law on an individual basis, rather than evaluating ‘whether the theory of
    recovery advanced by the plaintiff is likely to prove amenable to class treatment.’ ” (Id.
    at p. 1294.) Since the plaintiff’s theory of recovery focused on uniform policies and
    practices (such as the defendant’s failure to compensate employees for missed meals, rest
    breaks, and earned overtime), it was “more amenable to class treatment than individual
    disposition.” (Id. at p. 1300.) Indeed, in Jaimez, the defendant had a policy and practice
    of automatically deducting 30 minutes per shift for each employee’s meal break
    regardless of whether that meal break was actually taken, and the appellate court
    expressly found that this policy raised common legal and factual issues. (Id. at pp. 1294,
    1304.)
    Further, in balancing these common issues against any individual inquiries
    necessary, the Jaimez court rejected the trial court’s notion that common questions of fact
    and law did not predominate because the defendant had submitted declarations indicating
    that some employees did, in fact, get meal breaks, rest breaks, and proper pay stubs and
    thus there was a “ ‘strong indication that there could be conflicting testimony regarding
    whether these employees have common factual issues to be presented at trial.’ ” (Jaimez,
    supra, 181 Cal.App.4th at p. 1296.) Specifically, the appellate court declared that the
    trial court had improperly “focused on the merits of the declarations, evaluating the
    contradictions in the parties’ responses to the company’s uniform policies and practices,
    not the policies and practices themselves.” (Id. at p. 1300.) Unfortunately, the trial court
    in this case fell prey to the same errors that infected the trial court’s certification decision
    in Jaimez.
    Specifically, instead of identifying the principal legal issues presented in this
    matter and determining whether those “operative legal principles, as applied to the facts
    of the case, render the claims susceptible of resolution on a common basis’ ” (Alberts,
    supra, 241 Cal.App.4th at p. 399), the trial court here improperly focused on the minutiae
    of each individual janitor’s personal situation. This was a legal error and appears also to
    34
    have been the reason the trial court found Woolfson’s evidence irrelevant to the class
    certification inquiry. However, when the merits of the ultimate damages issues are set
    aside and Woolfson’s analysis of ABM’s payroll practices is considered, along with the
    other evidence submitted by plaintiffs, it becomes clear that numerous common issues
    predominate in this matter, rendering class certification appropriate.
    For instance, the legality of ABM’s uniform payroll policy—which assumes each
    employee works his or her scheduled shift and takes any legally required meal breaks
    absent some type of exception report—is a legal question that can be determined by
    reference to facts common to all class members. Certainly, the evidence provided by
    Woolfson that a mere 5,625 of the 1,836,083 time entries for ABM Workers he
    investigated (0.3 percent) contained adjustments to pay calls into question the efficacy of
    ABM’s asserted “timesheet maintenance” procedure, as does the evidence presented by
    plaintiffs that ABM does not generate exception reports for missed meals periods.
    Moreover, the legality of ABM’s auto-deduct policy for meal breaks in light of the
    recordkeeping requirements for California employers is also an issue amenable to
    classwide resolution. (See Cal. Code Regs., tit. 8, § 11050, subd. (7)(A)(3).) In addition,
    ABM’s apparent uniform practice of never providing premium pay to its employees,
    either for split shifts or missed meal breaks, is susceptible to classwide treatment.
    Moreover, ABM’s defenses with respect to split shift premium pay—that voluntary split
    shifts are not compensable and that class members are paid more than the threshold under
    which premium pay is mandated—are also susceptible to common proof. (See Saechao
    v. Landry’s, Inc. (N.D. Cal. Mar. 15, 2016, No. C 15-00815 WHA) 
    2016 U.S. Dist. LEXIS 33409
     at pp. *22–23; Kamar v. Radio Shack Corp. (C.D. Cal. 2008) 
    254 F.R.D. 387
    , 405; see Aleman, supra, 209 Cal.App.4th at pp. 574–575 [interpreting split shift
    Wage Order as a legal matter].) Finally, whether ABM fails to properly reimburse its
    employees for work-related travel, despite its asserted policy to do so, is also subject to
    common proof. (See Brewer v. General Nutrition Corp. (N.D. Cal. Nov. 12, 2014,
    No. 11-CV-3587 YGR) 
    2014 U.S. Dist. LEXIS 159380
     at pp. *27–30 [predominance of
    common questions on a travel reimbursement claim supported by evidence of
    35
    “exceedingly small percentage of employees who sought reimbursement”; evidence of
    mileage incurred could be determined on a class-wide basis where all relevant locations
    known].) Under these circumstances, fear that the determination of individual damages
    might prove overly complex should not have provided a basis for denial of class
    certification.
    Indeed, although we do not reach the issue, the trial court’s concern regarding the
    need for numerous individualized damage inquiries in this case may turn out to be over-
    exaggerated, given existing precedent indicating that the burden of proof shifts to
    employers “in the wage and hour context when an employer’s compensation records are
    so incomplete or inaccurate that an employee cannot prove his or her damages.” (Amaral
    v. Cintas Corp. No. 2 (2008) 
    163 Cal.App.4th 1157
    , 1189; see Cicairos v. Summit
    Logistics, Inc. (2005) 
    133 Cal.App.4th 949
    , 961 (Cicairos) [“ ‘[W]here the employer has
    failed to keep records required by statute, the consequences for such failure should fall on
    the employer, not the employee. In such a situation, imprecise evidence by the employee
    can provide a sufficient basis for damages.’ ”], overruled on another ground as stated in
    York v. Starbucks Corp. (C.D. Cal. Sept. 12, 2012, No. CV 08-07919 GAF (PJWx))
    
    2012 U.S. Dist. LEXIS 190086
    .) Thus, for example, since employers have a duty to
    record their employees’ meal periods, “[i]f an employer’s records show no meal period
    for a given shift over five hours, a rebuttable presumption arises that the employee was
    not relieved of duty and no meal period was provided.” (Brinker, 
    supra,
     53 Cal.4th at
    p. 1053 [Werdegar, J., conc.]; see Cal. Code Regs., tit. 8, § 11050, subd. (7)(A)(3).)
    Under such circumstances, a court may award damages, even if they are only
    approximate and based on statistical sampling. (Bell v. Farmers Ins. Exchange, supra,
    115 Cal.App.4th at pp. 746–751.)
    In summary, given that the classes proposed by plaintiffs in this case were
    ascertainable and plaintiffs’ allegations presented predominantly common questions, the
    trial court’s determinations to the contrary cannot stand. Rather, we conclude that the
    trial court’s denial of class certification—including its decision regarding the
    admissibility of the Woolfson materials—rested on improper criteria and erroneous legal
    36
    assumptions, amounting to an abuse of discretion. Plaintiffs have made a showing
    sufficient to allow them to take the next step in attempting to prove the merits of their
    contentions on a classwide basis.15
    III.       DISPOSITION
    The trial court’s order denying class certification is reversed and the matter
    remanded for certification of classes as set forth in this opinion. Plaintiffs are entitled to
    their costs on appeal.
    15
    In making this determination, we are cognizant of the trial management
    concerns raised by counsel for ABM at oral argument in this case, issues which may
    make the ultimate resolution of all or parts of this matter on a classwide basis
    problematic. However, as detailed above, our review following a denial of class
    certification is limited. (See Alberts, supra, 241 Cal.App.4th at p. 399.) Because the trial
    court’s order was based on improper criteria and erroneous legal assumptions, we
    reverse. Moreover, based on the record before us and as we have detailed at length
    above, it appears that plaintiffs have identified a number of common questions suitable
    for classwide resolution. Should plaintiffs’ trial plan subsequently prove unworkable,
    however, ABM may address any such issues to the trial court.
    37
    _________________________
    REARDON, J.
    We concur:
    _________________________
    RUVOLO, P. J.
    _________________________
    RIVERA, J.
    38
    Filed 1/10/18
    CERTIFIED FOR PUBLICATION
    IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA
    FIRST APPELLATE DISTRICT
    DIVISION FOUR
    ABM INDUSTRIES OVERTIME CASES                       JCCP No. 4502
    A132387, A133077 & A133695
    (City & County of San Francisco
    Super. Ct. No. CJC-07-004502)
    ORDER MODIFYING OPINION
    AND CERTIFYING OPINION FOR
    PUBLICATION
    (NO CHANGE IN JUDGMENT)
    THE COURT:
    It is ordered that the opinion filed December 11, 2017, be modified as follows:
    The citation to Cicairos v. Summit Logistics, Inc. (2005) 
    133 Cal.App.4th 949
    located on page 36 of the opinion shall be modified to delete the reference to subsequent
    history such that the case citation shall read in full: “see also Cicairos v. Summit
    Logistics, Inc. (2005) 
    133 Cal.App.4th 949
    , 961 (Cicairos) [“ ‘Where the employer has
    failed to keep records required by statute, the consequences for such failure should fall on
    the employer, not the employee. In such a situation, imprecise evidence by the employee
    can provide a sufficient basis for damages’ ”].)
    There is no change in the judgment.
    1
    In addition, the opinion in the above matter was not certified for publication in the
    Official Reports when filed on December 11, 2017. For good cause it now appears that
    the opinion, as modified herein, should be published in the Official Reports and it is so
    ordered.
    Dated: _______________________                   _________________________
    RUVOLO, P. J.
    2
    Trial Court:               City & County of San Francisco Superior Court
    Trial Judge:               Hon. Richard A. Kramer
    Counsel for Appellants:    Weinberg, Roger & Rosenfeld, David A.
    Rosenfeld, Christian L. Raisner, Emily P. Rich,
    Roberta D. Perkins; Mallison & Martinez, Stan
    S. Mallison, Marco A. Palau, Joseph D. Sutton;
    Rastegar & Matern, Matthew J. Matern
    Counsel for Respondents:   Littler Mendelson, Keith A. Jacoby, Dominic J.
    Messiha, Lauren E. Robinson
    3