FILED
                                                         United States Court of Appeals
                                                                 Tenth Circuit

                                                                March 9, 2015
                                   PUBLISH                   Elisabeth A. Shumaker
                                                                 Clerk of Court
                   UNITED STATES COURT OF APPEALS

                                TENTH CIRCUIT


 SANDRA ELLIS, on behalf of herself
 and others similarly situated,

       Plaintiff - Appellant,

 v.                                                    No. 13-1346

 J.R.’S COUNTRY STORES, INC., a
 Colorado corporation,

       Defendant - Appellee.


                 Appeal from the United States District Court
                         for the District of Colorado
                   (D.C. No. 1:12-CV-01916-CMA-KLM)


Donna E. Dell’Olio (Bradley J. Sherman, Cornish & Dell’Olio, P.C., Colorado
Springs, Colorado, with her on the briefs), Cornish & Dell’Olio, P.C., Colorado
Springs, Colorado, for Plaintiff-Appellant.

Christian D. Hammond (Lawrence D. Stone, Dufford & Brown, P.C., Denver,
Colorado, with him on the brief), Dufford & Brown, P.C., Denver, Colorado, for
Defendant-Appellee.


Before HOLMES, McKAY, and BACHARACH, Circuit Judges.


HOLMES, Circuit Judge.
      Sandra Ellis appeals from the district court’s grant of summary judgment in

favor of J.R.’s Country Stores, Inc., her former employer, on her claim for

violation of the Fair Labor Standards Act of 1938 (“FLSA”), 29 U.S.C. § 207.

Our jurisdiction arises under 28 U.S.C. § 1291, and we affirm.

                                         I

      We offer at the outset an overview of the applicable statutory framework.

We then provide the factual background and procedural history of this case,

reciting all summary-judgment evidence in the light most favorable to Ms. Ellis

as the nonmovant. See Salazar v. Butterball, LLC, 

, 1136 (10th Cir.

2011); Gwinn v. Awmiller, 

, 1215 (10th Cir. 2004).

                                        A

      Congress enacted the FLSA in order to “protect all covered workers from

substandard wages and oppressive working hours, ‘labor conditions [that are]

detrimental to the maintenance of the minimum standard of living necessary for

health, efficiency and general well-being of workers.’” Barrentine v.

Arkansas-Best Freight Sys., Inc., 

, 739 (1981) (alteration in original)

(quoting 29 U.S.C. § 202(a)); see Chavez v. City of Albuquerque, 

,

1304 (10th Cir. 2011) (“The purpose of FLSA overtime is ‘to compensate those

who labored in excess of the statutory maximum number of hours for the wear

and tear of extra work and to spread employment through inducing employers to

shorten hours because of the pressure of extra cost.’” (quoting Bay Ridge

                                         2
Operating Co. v. Aaron, 

, 460 (1948))). The FLSA effectuates this

purpose by “generally requir[ing] an employer to pay its employees at a rate of

one and one-half times their regular rate of pay for any time worked in excess of

forty hours in a given workweek.” Archuleta v. Wal-Mart Stores, Inc., 

, 1228 (10th Cir. 2008); accord Albers v. Bd. of Cnty. Comm’rs, 

, 699 (10th Cir. 2014). In other words, the statute provides for overtime pay

in specified circumstances. See 29 U.S.C. § 207(a)(2). If an employee satisfies

her burden of “prov[ing] that the employer is violating the FLSA,” 

, she may be entitled to recoup unpaid overtime compensation and

liquidated damages, see 29 U.S.C. § 216(b).

      But despite “the remedial nature of [the] FLSA,” Pacheco v. Whiting

Farms, Inc., 

, 1206 (10th Cir. 2004), “not all workers require[ ] the

same kind of protection” under the statute, Ackerman v. Coca-Cola Enters., Inc.,

, 1263 (10th Cir. 1999). Notably, as is relevant here, “employee[s]

employed in a bona fide executive, administrative, or professional capacity . . . as

such terms are defined . . . by regulations of the Secretary [of Labor],” 29 U.S.C.

§ 213(a)(1), are exempt from the FLSA’s requirements and are thus ineligible for

overtime compensation. These regulations “are entitled to judicial deference and

are the primary source of guidance for determining the scope of exemptions to the

FLSA.” 

(internal quotation marks omitted).




                                          3
      Exercising its delegated authority on behalf of the U.S. Department of

Labor (“DOL”), the Secretary of Labor (“Secretary”) has promulgated several

implementing regulations for the FLSA, one of which defines the term “employee

employed in a bona fide executive capacity” as an employee who is:

             (1) Compensated on a salary basis at a rate of not less than $455
             per week . . . ;
             (2) Whose primary duty is management of the enterprise in which
             the employee is employed or of a customarily recognized
             department or subdivision thereof;
             (3) Who customarily and regularly directs the work of two or
             more other employees; and
             (4) Who has the authority to hire or fire other employees or
             whose suggestions and recommendations as to the hiring, firing,
             advancement, promotion or any other change of status of other
             employees are given particular weight.

29 C.F.R. § 541.100(a); accord Maestas v. Day & Zimmerman, LLC, 

, 827 (10th Cir. 2012).

      Pursuant to the FLSA’s regulatory framework, an employee must be paid

on a salary basis to satisfy the requirements of an “executive” exempt from the

statute’s overtime requirements. See Auer v. Robbins, 

, 455–56

(1997); In re Wal-Mart Stores, Inc., 

, 1180 (10th Cir. 2005). “An

employee will be considered to be paid on a ‘salary basis’” under the Secretary’s

regulations if she “regularly receives each pay period on a weekly, or less

frequent basis, a predetermined amount constituting all or part of [her]

compensation, which amount is not subject to reduction because of variations in




                                         4
the quality or quantity of the work performed.” 29 C.F.R. § 541.602(a). 1 As we

have explained, “[s]ince exempt employees are not paid by the hour, the FLSA’s

implementing regulations prohibit employers from docking their pay for working

less than a full eight-hour day.” McBride v. Peak Wellness Ctr., Inc., 

, 705 (10th Cir. 2012). We are obliged to interpret this “executive” exemption

narrowly against employers. See Chessin v. Keystone Resort Mgmt., Inc., 

, 1192 (10th Cir. 1999) (“In a case involving the FLSA, an employer

bears the burden of proving . . . the applicability of an FLSA exemption; we must

construe the exemption narrowly against the employer.”); Aaron v. City of

Wichita, 

, 657 (10th Cir. 1995) (“Exemptions to the FLSA are to be

narrowly construed . . . . An employer must prove that the employee is exempt by

‘clear and affirmative’ evidence.” (citation omitted) (quoting Donovan v. United

Video, Inc., 

, 581 (10th Cir. 1984))).

      An employer may lose the right to treat otherwise eligible employees as

exempt “executives” in certain situations. For instance, “[a]n employer who

makes improper deductions from salary shall lose the exemption if the facts

demonstrate that the employer did not intend to pay employees on a salary basis.”

29 C.F.R. § 541.603(a) [hereinafter “subsection (a)”]. Whether an employer


      1
             “Because the salary-basis test is a creature of the Secretary’s own
regulations, his interpretation of it is . . . controlling unless plainly erroneous or
inconsistent with the regulation.” 

(internal quotation marks
omitted).

                                          5
making such improper deductions should be deemed not to have intended to pay

on a salary basis turns on whether the employer has an “actual practice of making

improper deductions,” 

to a “theoretical possibility of such

deductions,” 

. The Secretary has provided guidance in

that regard, namely by detailing five applicable factors for courts to consider:

             The factors to consider when determining whether an employer
             has an actual practice of making improper deductions include,
             but are not limited to: the number of improper deductions,
             particularly as compared to the number of employee infractions
             warranting discipline; the time period during which the employer
             made improper deductions; the number and geographic location
             of employees whose salary was improperly reduced; the number
             and geographic location of managers responsible for taking the
             improper deductions; and whether the employer has a clearly
             communicated policy permitting or prohibiting improper
             deductions.

29 C.F.R. § 541.603(a).

      On the face of subsection (a), it is evident that one variable which may

inform courts as to the existence vel non of an “actual practice of making

improper deductions” is whether the employer has promulgated a “clear[ ]” policy

concerning improper deductions. 

provides helpful explication

regarding this final salary-basis-test factor in another subsection of the same

regulation, noting that a “clearly communicated policy” is one “that prohibits the

improper pay deductions specified in [subsection] (a) and includes a complaint

mechanism.” 

[hereinafter “subsection (d)”]. Further, according

to subsection (d):

                                          6
             [t]he best evidence of a clearly communicated policy is a written
             policy that was distributed to employees prior to the improper
             pay deductions by, for example, providing a copy of the policy
             to employees at the time of hire, publishing the policy in an
             employee handbook or publishing the policy on the employer’s
             Intranet.

under the rubric of subsection (a)—as informed by subsection

(d)—“if the facts demonstrate that” the employer maintained “[a]n actual practice

of making improper deductions,” the default conclusion is that the employer

“shall lose the exemption.” 

(emphasis added). The employer is

stripped of the exemption because its “actual practice” of improperly deducting

pay vitiates the intent to pay a salary under the salary-basis test. See 

the facts so indicate, “the exemption is lost during the time

period in which the improper deductions were made.” 

accord

Baden-Winterwood v. Life Time Fitness, Inc., 

, 628 (6th Cir. 2009).

      However, the Secretary has carved out a savings provision for “[i]mproper

deductions that are either isolated or inadvertent,” which allows the exemption to

stand “if the employer reimburses the employees for such improper deductions.”

29 C.F.R. § 541.603(c) [hereinafter “subsection (c)”]; see 

(discussing an employer’s attempt to “take advantage of” this savings

provision “by reimbursing its pharmacists for the times it paid them less than

their guaranteed minimum salary”). As a practical matter, “[t]he ‘window of


                                           7
correction’ provided by [subsection (c)] allows employers to treat otherwise

eligible employees as salaried, regardless of the employer’s one-time or

unintentional failure to adhere to [the] requirements” of the salary-basis test set

forth in subsection (a). Spradling v. City of Tulsa, 

, 1503 (10th Cir.

1996).

                                           B

         Having explained the legal standards material to the disposition of this

appeal, we proceed to the salient background facts, viewed in the light most

favorable to Ms. Ellis.

         J.R.’s Country Stores, Inc. (“J.R.’s” or “the Company”) is a chain of

convenience markets located throughout Colorado. Its individual shops, which

sell gas, groceries, and sundries, are operated by store managers. As set forth in

the Company’s Store Manager Base Pay Plan (“the Pay Plan”), each manager is

considered a “salaried employee” and is expected to work “a minimum of 50

hours per week and a minimum 5 day work week.” Aplt. App. at 378 (2011 Pay

Plan, filed Jan. 11, 2013). The Company also requires every store manager to

“work whatever number of hours per week is necessary to effectively run [her]

store.” 

2007, Ms. Ellis began working as the manager of a J.R.’s

store in Holly, Colorado. The Company classified her as an exempt, salaried

employee and paid her $600 on a weekly basis (an amount that increased to $625

                                           8
in February 2011), as well as a monthly bonus commensurate with her store’s

performance. While Ms. Ellis was employed in this capacity, she generally 2

worked fifty-hour (and sometimes closer to sixty-hour) weeks, which was

consistent with the terms of the Pay Plan.

      On April 3, 2012, Ms. Ellis received a paycheck in the amount of $593.80,

instead of the $625 she customarily earned in a pay period. The Company had

deducted $31.20 from that paycheck because for the work week of March 16 to

March 22, 2012, Ms. Ellis reported having worked only 40.91 hours of the

requisite fifty. More specifically, Ms. Ellis “missed less than a full day of work

[that week].” Aplt. Opening Br. at 5. 3 This was the only instance during Ms.

Ellis’s tenure with the Company that she received less than her typical,

predetermined pay.




      2
             The Company submitted time sheets indicating that for certain
isolated weeks during her employment, Ms. Ellis reported working fewer than
fifty hours.
      3
             As the district court observed, the “parties maintain different, though
overlapping, explanations as to why Ellis’s pay was reduced.” Aplt. App. at 448
n.1 (Order Granting Mot. for Summ. J., filed July 12, 2013). Compare 

n.3 (Mot. for Summ. J., filed Oct. 11, 2012) (“Plaintiff . . . informed J.R.’s that
she wanted to use a sick day instead of a vacation day.”), with 

(Aff. of
Sandra Ellis, dated Jan. 9, 2013) (“I . . . lost a vacation day because I was sick.”).
For our purposes, the principal points are (1) that Ms. Ellis failed to work the
mandatory fifty-hour business week during that pay period, and (2) that “the
parties agree that Ellis’s pay was reduced as a result of her inability to” do so. 

n.1.

                                          9
      Ms. Ellis resigned on April 6, 2012. Soon afterward, she sent the Company

a letter “claim[ing] that she [was] owed $42,187.50 in unpaid overtime wages.”

Aplt. App. at 284 (Letter, dated May 18, 2012). In lodging that request, Ms. Ellis

opined that she had “lost her exempt status under the [FLSA] . . . when J.R.’s

made a one-time deduction from [her] pay for not working fifty hours during a

workweek,” which, in turn, entitled her to three years’ worth of retroactive

overtime pay. 

the correct result, according to Ms. Ellis, because the

paycheck in question demonstrated a “willful” “practice of making improper

deductions” by the Company. 

marks omitted).

      The Company initially denied both that this singular deduction was

improper and that it constituted an actual practice of improper deductions under

the applicable regulatory provisions. Nonetheless, the Company informed Ms.

Ellis by letter that it would “take the high road . . . and pay [her] for the overtime

she worked during the time period in which the deduction occurred.” 

(internal quotation marks omitted). To explain the amount of the accompanying

check, $332.88—which was more than ten times the amount of the deduction—the

Company stated:

             During the workweek in which the deduction was made, [you]
             worked a total of 56.42 hours. . . . [Your] hourly rate was
             $12.50, meaning [your] overtime rate is $18.75. [Your] $625
             weekly salary was based on a fixed 50 hour workweek. By
             reimbursing [you] the $150 that was deducted . . . , J.R.’s has
             now made [you] whole for that workweek and has therefore paid
             [you] for the first 50 hours for purposes of the overtime

                                      10
             calculation. What remains is (1) [your] regular pay for the
             additional 6.42 hours, and (2) the “and a half” for the 16.42 hours
             of overtime. Therefore, the calculation is as follows:

             (1) 6.42 hours x $12.50 (regular time) = $80.25

             (2) 16.42 hours x $6.25 (half time) = $102.63

             Total = $182.88.

             Therefore, on behalf of J.R.’s, enclosed please find a
             check . . . in the amount of $332.88 ($150 + $182.88).

(footnote omitted). The Company further noted that it had

“use[d] the hours worked during the March 23, 2012 workweek for purposes of

the overtime calculation” for two reasons: (1) to comport with the regulatory

language “during the time period in which the [deduction was] made,” 29 C.F.R.

§ 541.603(b); and (2) to produce “a more favorable calculation for” Ms. Ellis.

Aplt. App. at 285 n.2. It did so even though the deduction at issue “was made

based on the number of hours [Ms. Ellis] worked in the previous

workweek”—40.91 hours. 

      On July 23, 2012, Ms. Ellis filed a lawsuit against her former employer in

the United States District Court for the District of Colorado, bringing one claim

under the FLSA. Alleging that her “compensation was reduced when she did not

work at least 50 hours” per week, Ms. Ellis urged that she “was not an exempt

employee [under the FLSA] because she was not paid on a true salary basis.” 

(Compl., filed July 23, 2012). Thus, reasoned Ms. Ellis, the Company’s


                                         11
failure to classify her as an hourly employee entitled her to recoup retroactive

overtime pay “for all hours worked in excess of forty hours in a single work

week” during her term of employment. 

Ms. Ellis also signaled her

intent to certify a class of similarly situated store managers.

      The Company filed a motion for summary judgment, attaching as exhibits

all time sheets and payroll records pertaining to Ms. Ellis’s employment. Instead

of responding to the Company’s motion, Ms. Ellis filed a “Motion to Certify

Conditionally a Collective Action Under the [FLSA] and for Court Assisted

Notice,” 

(Mot. to Certify, filed Oct. 18, 2012) (capitalization altered),

and, shortly thereafter, also a motion to deny or defer judgment under Federal

Rule of Civil Procedure 56(d) so that she could “reasonably engage in discovery

sufficient to rebut a motion for summary judgment,” 

(Mot. to Den. or

Defer J., filed Oct. 31, 2012) (capitalization altered).

      For its part, the Company then sought a temporary stay of class

certification, pending the district court’s summary-judgment determination. The

district court, in an order dated December 11, 2012, denied Ms. Ellis’s Rule 56(d)

motion and granted the Company’s motion to stay a class-certification ruling until

the court’s decision on the motion for summary judgment. Consequently, on

January 11, 2013, Ms. Ellis filed her response to the Company’s summary-

judgment motion.




                                          12
      Then, on July 12, 2013, the district court granted summary judgment in

favor of the Company. The substance of the court’s ruling was twofold. First,

the court determined that the undisputed material facts demonstrated the

Company’s intent to pay Ms. Ellis a salary and that the one-time pay deduction

did not change that intent. Stated otherwise, the district court ruled that the

Company’s deduction did not constitute a violation of the salary-basis test.

Second, the court reasoned, even if Ms. Ellis could demonstrate that the Company

did not intend to pay her a salary, her FLSA claim would still founder because the

Company was entitled to rely on the protection of the window-of-correction

defense detailed in the FLSA’s implementing regulations.

      In light of the foregoing findings, the district court awarded summary

judgment to the Company and denied Ms. Ellis’s motion for class certification as

moot. The court entered judgment on July 15, 2013, dismissing the lawsuit with

prejudice. Ms. Ellis timely appealed from the district court’s judgment.

                                          II

      On appeal, Ms. Ellis challenges the district court’s summary-judgment

order on several fronts. “We review the district court’s summary judgment

decision de novo, applying the same standard as the district court.” Brammer-

Hoelter v. Twin Peaks Charter Acad., 

, 1184 (10th Cir. 2010);

accord Colo. Cross-Disability Coal. v. Abercrombie & Fitch Co., 

,

1218 (10th Cir. 2014). Summary judgment is appropriate “if the movant shows

                                          13
that there is no genuine dispute as to any material fact and the movant is entitled

to judgment as a matter of law.” Fed. R. Civ. P. 56(a); accord Felkins v. City of

Lakewood, 

, 650 (10th Cir. 2014). As noted above, we “examine the

record and all reasonable inferences that might be drawn from it in the light most

favorable to the non-moving party”—that is, Ms. Ellis. Merrifield v. Bd. of Cnty.

Comm’rs, 

, 1077 (10th Cir. 2011) (internal quotation marks

omitted). Moreover, “we can affirm on any ground supported by the record, so

long as the appellant has had a fair opportunity to address that ground.” Alpine

Bank v. Hubbell, 

, 1108 (10th Cir. 2009) (internal quotation marks

omitted).

      Ms. Ellis also seeks review of the district court’s denial of deferment for

expanded discovery pursuant to Rule 56(d), which we review for an abuse of

discretion. See FDIC v. Arciero, 

, 1116 (10th Cir. 2013). We will

find an abuse of discretion “when the district court bases its ruling on an

erroneous conclusion of law or relies on clearly erroneous fact findings.” Walters

v. Wal-Mart Stores, Inc., 

, 1172 (10th Cir. 2013) (internal quotation

marks omitted); accord Trentadue v. FBI, 

, 806 (10th Cir. 2009). “A

finding of fact is clearly erroneous if it is without factual support in the record or

if, after reviewing all of the evidence, we are left with the definite and firm

conviction that a mistake has been made.” In re Vaughn, 

, 1180

(10th Cir. 2014) (internal quotation marks omitted). In the Rule 56(d) context,

                                           14
the party requesting deferral of judgment—here, Ms. Ellis—shoulders the burden

of demonstrating an abuse of discretion. See 

.

                                           III

      Ms. Ellis presents two overarching questions for our review of the district

court’s summary-judgment ruling: (1) whether the court misapplied the salary-

basis test associated with her FLSA exemption (i.e., subsection (a)), and (2)

whether the court incorrectly permitted the Company to avail itself of the

window-of-correction defense (i.e., subsection (c)) to rectify any improper

deduction to her pay. Not surprisingly, Ms. Ellis answers both questions in the

affirmative and urges us to find that the district court committed error in awarding

summary judgment to the Company. We are not persuaded by any of Ms. Ellis’s

appellate arguments and, in point of fact, find ourselves in substantial agreement

with the district court in all respects.

      Before commencing our substantive discussion, however, we note the

Company’s suggestion that we cabin our affirmance to the issue of whether the

window-of-correction defense was satisfied. The Company reasons that such an

approach “would forestall the need to address most of [Ms.] Ellis’s arguments,”

Aplee. Br. at 6—most notably, her contention that the district court’s reading of

subsections (a) and (c) contravenes settled principles of statutory interpretation.

Although we recognize that, under certain circumstances, the task of reconciling




                                           15
subsections (a) and (c) could potentially be difficult, 4 on this record, Ms. Ellis’s

statutory-interpretation argument does not give us cause to shrink from our

jurisprudentially beneficial plan of comprehensively resolving this dispute. That

is because (as more fully explicated below), on this record, the Company can

clearly rely on subsections (a) and (c) and secure a successful outcome by doing

so. More specifically, we conclude that there is no genuine issue of material fact

that the Company had no practice or policy of making improper deductions for

salaried employees (which effectively resolves the subsection (a) question in the

Company’s favor). And we also determine that the documented improper

deduction that the Company made from Ms. Ellis’s paycheck was truly isolated

(which effectively gives the Company victory on the subsection (c) question).

Accordingly, for the reasons stated herein, we are satisfied that the district court

properly awarded summary judgment to the Company.

                                           A

      We first address Ms. Ellis’s contention that the district court incorrectly

found no genuine issues of material fact concerning the Company’s intent to pay

her a salary; instead, reasons Ms. Ellis, the court should have considered her “an

      4
             For example, Ms. Ellis contends that if the Company is stripped of its
exemption under subsection (a) by virtue of a practice or policy of making
improper deductions, then, as a matter of law, such conduct cannot be “isolated”
under subsection (c). If we were obliged to reach this issue, we certainly would
consider Ms. Ellis’s position worthy of serious consideration and would not be
inclined to summarily reject it out of hand. However, because the two provisions
harmonize for purposes of this appeal, we need not venture down this road.

                                          16
hourly employee in every sense.” Aplt. Opening Br. at 20. According to Ms.

Ellis, the terms of the Pay Plan ineluctably yield this conclusion. The Pay Plan,

she says, is “unlike any other to be examined by a federal court for conformance

to the salary basis test,” 5 

a “combination of unique factors”: a “hard

floor” of fifty hours per week, a requirement to log the number of hours worked,

and an alleged “stated policy of deducting pay for less than a day’s absence,”

Aplt. Reply Br. at 18. For this reason, among others that we address below, Ms.

      5
              While we would not describe the Pay Plan as one “unlike any other,”
Aplt. Opening Br. at 20, we note that existing caselaw at least suggests that the
Pay Plan is unusual. But the few isolated cases we have located—involving
compensation plans that obligate employees to work in the vicinity of fifty hours
per week—are not favorable to Ms. Ellis’s position. For instance, in an
unpublished opinion, a panel of the Third Circuit considered the salary-basis test
fulfilled when, under the operative plan, “[the plaintiff] was entitled to a
guaranteed base pay of $60,000, and she was expected to work 45 hours per
week.” Sander v. Light Action, Inc., 525 F. App’x 147, 148 (3d Cir. 2013). Some
district courts have likewise found the test satisfied on similar facts. See, e.g.,
Ottaviano v. Home Depot, Inc., 

, 1008–09 (N.D. Ill. 2010)
(deeming employees exempt when Home Depot required a fifty-five-hour work
week); Clougher v. Home Depot U.S.A., Inc., 

, 295 (E.D.N.Y.
2010) (finding that “the mere fact that Home Depot required even its salaried
employees to work no less than 55-hours per week or face discipline . . . [did] not
undermine [the plaintiff’s] proper classification as a salaried employee”);
Palazzolo-Robinson v. Sharis Mgmt. Corp., 

, 1192 (W.D.
Wash. 1999) (finding an employee who “was told [when hired] . . . that all
managers worked 50 hours a week, . . . and that compensation would be on a
salary basis” to be exempt (internal quotation marks omitted)). Other cases
involving mandatory “forty-hour-plus” work weeks also lend Ms. Ellis no succor
because they arise in the context of the “fluctuating” work week, described at 29
C.F.R. § 778.114, which is inapposite here.

       Ultimately, under our controlling precedent, we cannot agree that the terms
of the Pay Plan militate in favor of considering Ms. Ellis non-exempt and
stripping the Company of its exemption under subsection (a).

                                          17
Ellis argues that the district court’s ruling constitutes reversible error. Though

Ms. Ellis strives mightily on multiple fronts to attack the salary-basis aspect of

the district court’s decision, we reject each of her arguments. We conclude that

the Company did not lose this exemption as regards Ms. Ellis.

                                          1

      Under subsection (a), an “actual practice of making improper deductions

demonstrates that the employer did not intend to pay employees on a salary basis”

and results in the employer’s loss of the executive exemption for those

employees. 29 C.F.R. § 541.603(a). As noted, the Secretary has offered several

guideposts for courts confronting this issue:

             The factors to consider when determining whether an employer
             has an actual practice of making improper deductions include,
             but are not limited to: [1] the number of improper deductions,
             particularly as compared to the number of employee infractions
             warranting discipline; [2] the time period during which the
             employer made improper deductions; [3] the number and
             geographic location of employees whose salary was improperly
             reduced; [4] the number and geographic location of managers
             responsible for taking the improper deductions; and [5] whether
             the employer has a clearly communicated policy permitting or
             prohibiting improper deductions.

argues that the district court rendered an erroneous decision by

focusing exclusively on the first factor of the salary-basis test—the number of

improper deductions. We disagree. On the contrary, we believe the district

court’s ruling can be fairly read to address all of the salary-basis factors in some

                                          18
appreciable manner. First, the district court noted that the Company “reduced

[Ms. Ellis’s] pay on only one occasion,” Aplt. App. at 452, even though—as the

time sheets reflected—“[Ms.] Ellis worked less than 50 hours on 13 separate

occasions, not counting holidays or days when [she] was sick or on vacation,” 

Next, with regard to the time period during which the Company made the

improper deduction (i.e., the second factor), the court observed that Ms. Ellis’s

pay was improperly reduced on April 3, 2012, which impacted one week-long pay

cycle. As to the third and fourth factors, the court stated:

             [W]ith respect to the number and geographic location of
             additional employees whose salary was improperly reduced, the
             Court stayed proceedings on that issue because, having
             experienced only one allegedly improper deduction, Ellis
             appeared to lack a viable claim, which could not be salvaged by
             recourse to evidence relating to other individuals. . . . [F]or
             substantially the same reason, the number and geographic
             location of managers responsible for taking the improper
             deductions does not support Ellis’s position.         No such
             “deductions” (plural), occurred—only the singular deduction that
             Ellis acknowledges.

(internal quotation marks omitted). And, finally, in assessing “whether

the [Company] has a clearly communicated policy permitting or prohibiting

improper deductions,” 29 C.F.R. § 541.603(a)—that is, the fifth factor—the

district court rejected Ms. Ellis’s assertion, which she made in an affidavit, that

the Company’s policy was “not clear,” Aplt. App. at 454 n.4 (internal quotation

marks omitted), and expressly referenced the Company’s policy as set forth in its

employee handbook, see 

(concluding that “[the Company’s] employee

                                         19
handbook contains a written policy” that “clearly communicates that improper

deductions are prohibited,” as contemplated by subsection (d)).

      Because even a brief review of the challenged district court order belies

Ms. Ellis’s assertion that only one component of the salary-basis test—the number

of improper deductions—received consideration, this argument does not avail her.

We are satisfied by the district court’s cogent reasons offered in applying the

test—and, although it disposed of the third and fourth factors summarily, we

discern no error in that choice. The court had already scrutinized those factors in

its December 11, 2012, ruling on Ms. Ellis’s Rule 56(d) motion and, absent

significant intervening developments, was under no legal obligation to revisit the

issues. In any event, as the Company has noted, the fact that “other courts have

examined as few as one of the five factors,” Aplee. Br. at 19, strongly militates

against a conclusion that the court’s analytical approach here constituted error.

See, e.g., 

–05 (arguably only considering the existence

vel non of a deduction); Cash v. Cycle Craft Co., 

, 683–84 (1st Cir.

2007) (assessing only the first factor); Mathews v. Bronger Masonry, Inc., 772 F.

Supp. 2d 1004, 1012–14 (S.D. Ind. 2011) (assessing the first and second factors).

All told, the district court’s straightforward application of the salary-basis test

affords us no legally cognizable basis to reverse.




                                          20
                                         2

      More broadly, Ms. Ellis disputes the district court’s conclusion that she

failed to raise genuine issues of material fact concerning the existence of an

actual Company practice or policy of making improper deductions. For the

reasons discussed, we find her position unconvincing.

                                         a

      First, Ms. Ellis argues that the district court incorrectly found no genuine

issue of material fact concerning the existence of an actual practice of improper

deductions by the Company—and, therefore, no genuine issue of material fact

concerning the Company’s intent to pay her on a salary basis. We disagree.

      After setting out the components of the salary-basis test, the district court

determined that the factors to consider when determining whether an employer

has an actual practice of making improper deductions, on the record presented,

did not prompt a finding that the Company had such a practice. The court

appropriately invoked the Supreme Court’s Auer decision to support its view that

the Company’s one-time improper deduction from Ms. Ellis’s pay, “taken under

unusual circumstances[, would] not defeat [Ms. Ellis’s] salaried status.” Aplt.

App. at 452 (citing 

–61). This is hardly an unusual

conclusion in FLSA jurisprudence, and it is the correct result here. See, e.g.,

Carpenter v. City & Cnty. of Denver, 

, 767 (10th Cir. 1997) (noting

with approval that “although there were two cases of alleged deductions, the

                                         21
[district] [c]ourt specifically recognized that such one time deductions under

unusual circumstances will not oust exempt status and may be remedied”); see

also Kennedy v. Commonwealth Edison Co., 

, 372 (7th Cir. 2005)

(discussing subsection (a) and its predecessor regulation, and concluding that

“[i]dentifying a few random, isolated, and negligible deductions is not enough to

show an actual practice or policy of treating as hourly the theoretically

salaried. . . . This means that these isolated instances of deductions do not create

a genuine issue of fact about the proper characterization of the plaintiffs’

positions.”); Block v. City of L.A., 

, 415 (9th Cir. 2001) (“In some

cases, the number of [employer] suspensions alone may be sufficient indicia of

the employer’s intent to resolve the ‘actual practice’ determination. For example,

the number may be so small that there would be no way to say that the employer

meant to treat an entire class of employees as hourly rather than salaried by virtue

of one or two isolated suspensions.”); Aiken v. City of Memphis, 

,

762 (6th Cir. 1999) (“The evidence shows that only one captain ever faced a

deduction in pay. Thus, plaintiffs cannot demonstrate an actual practice of

applying such deductions to captains.”).

      In affirming the district court on this basis, we note that the regulatory

language—which “[is] entitled to judicial deference [as] the primary source of

guidance for determining the scope of exemptions to the FLSA,” 

(internal quotation marks omitted)—cannot be overlooked. We find

                                           22
especially telling the Secretary’s use of the term deductions (rather than

deduction) multiple times in the text of the FLSA’s implementing regulations.

See generally 29 C.F.R. §§ 541.603(a)–(c). The Supreme Court has instructed

that such word choices should be considered deliberate. See, e.g., Hertz Corp. v.

Friend, 

, 93 (2010) (finding the fact that “[t]he word ‘place’ is in the

singular, not the plural” critical in interpreting the relevant statute); Metro.

Stevedore Co. v. Rambo, 

, 296 (1995) (“The use of ‘conditions,’ a

word in the plural, suggests that Congress did not intend to limit the bases for

modifying awards to a single condition.”). 6

      Our growing body of FLSA caselaw reflects this understanding of the

Secretary’s regulatory focus: we have held that “one change in base hours

[producing a deduction] . . . is not the kind of frequent change necessary to create

a factual dispute.” 

; see also In re 

(in crafting rule regarding prospective deductions, expressly

requiring proof of “frequent” “salary changes” and “adjustments” (emphases

added)). Along those lines, we find it implausible that subsection (a)’s

requirement of a “practice” of such deductions contemplates isolated conduct.

See 29 U.S.C. § 203(o) (treating “custom” and “practice” as synonymous); see

      6
             The Court relied on a leading treatise on statutory construction which
provides that, “[o]rdinarily[,] the legislature by use of a plural term intends a
reference to more than one matter or thing,” 2A N. Singer, Sutherland on
Statutory Construction § 47.34, at 274 (5th ed. 1992), in reaching this conclusion.
See Metro. Stevedore 

.

                                          23
also Webster’s Third New International Dictionary 1780 (2002) [hereinafter

“Webster’s”] (defining “practice,” inter alia, as “performance or application

habitually engaged [in]” and “repeated or customary action” (emphases added)).

      In sum, we agree that Ms. Ellis did not meet her summary-judgment burden

regarding her first argument for reversal. We reach this conclusion because, as a

matter of law, Ms. Ellis’s failure to identify more than the April 3, 2012,

deduction precluded a finding of an “actual practice,” 29 C.F.R. § 541.603(a), by

the Company of making improper salary deductions.

                                          b

      Ms. Ellis’s second salary-basis argument does not give us pause in reaching

this conclusion. Specifically, she directs her fire at what she perceives to be the

Company’s impermissible deductions policy and contends that “[she] raised a

genuine issue of material fact concerning the existence of an employer policy

requiring deduction from her pay when she was absent from work for a brief

period of time.” Aplt. Opening Br. at 21. This argument implicates both

subsections (a) and (d): subsection (a) expressly counsels courts in reaching an

“actual practice” determination to inquire regarding the existence vel non of a

“clearly communicated [employer] policy permitting or prohibiting improper

deductions,” and subsection (d)’s provisions inform that specific inquiry.

However, we agree with the district court’s determination that the Company had

no “policy” requiring improper deductions. And, therefore, we do not fault the

                                         24
court’s explicit repudiation of Ms. Ellis’s “assert[ion] that the $31.20 deduction

on April 3, 2012, . . . was part of a policy requiring a deduction maintained by

J.R.’s.” Aplt. App. at 453 (internal quotation marks omitted).

      As a preliminary matter, we note that the approach taken by the DOL with

respect to an employer’s “policy” regarding deductions in the context of the

salary-basis test is not hypertechnical. That is, the Secretary

             rejects a wooden requirement of actual deductions, but in their
             absence . . . requires a clear and particularized policy—one
             which effectively communicates that deductions will be made in
             specified circumstances. This avoids the imposition of massive
             and unanticipated overtime liability . . . in situations in which a
             vague or broadly worded policy is nominally applicable to a
             whole range of personnel but is not significantly likely to be
             invoked against salaried employees.

(emphasis added) (citation omitted) (internal quotation

marks omitted).

      Relevant here, the Company’s “policy,” as set forth in the version of the

employee handbook in effect during Ms. Ellis’s employment, declares that “[t]he

Company prohibits deductions from an exempt, salaried employee’s pay except

under the circumstances set forth in the [FLSA] and state law. If you believe that

improper [deductions have occurred], it will promptly reimburse the employee

and ensure the mistake will be corrected in the future.” Aplt. App. at 384 (Emp.

Handbook, filed Jan. 11, 2013). We would be hard-pressed to conclude that this

language evinces a policy that mandates improper deductions, as Ms. Ellis


                                          25
intimates on appeal. Indeed, essentially the obverse is true. This is so because

the foregoing handbook language tracks the text of the Secretary’s guidance

regarding the salary-basis test, which specifies that an employer will retain the

exemption as long as it maintains “a clearly communicated policy that prohibits

the improper pay deductions specified in [subsection] (a) and includes a

complaint mechanism, reimburses employees for any improper deductions and

makes a good faith commitment to comply in the future.” 29 C.F.R.

§ 541.603(d).

      Nevertheless, in Ms. Ellis’s view, the Company’s FLSA-related handbook

policy flouts the regulatory scheme for two reasons. First, she claims, “[a] policy

that requires the employee to research the law to determine the employer’s policy

does not clearly communicate which deductions are not allowed.” Aplt. Opening

Br. at 32. We understand Ms. Ellis to argue that the law requires employers not

to task their exempt employees with “learning that deductions from an exempt

employee’s pay for absences from work of less than a day’s duration are

prohibited by the FLSA.” Aplt. App. at 360 (Resp. to Mot. for Summ. J., filed

Jan. 11, 2013). This contention did not avail her in the district court because she

failed to marshal any authority “indicating that a policy fails to be clear merely

because it incorporates by reference, but does not reiterate in full, the governing

law.” 

n.4. Nor does this contention help Ms. Ellis here; for, like the




                                          26
district court did, we find her hypothesis unsupported. 7 Given Ms. Ellis’s

obligation to prove the Company violated the FLSA, we cannot conclude that she

has succeeded in light of the Company handbook’s obvious discussion of

deductions in the context of “the circumstances set forth in the federal Fair Labor

Standards Act (FLSA) and state law.” 

The “best evidence of a clearly

communicated policy” is one “distributed to employees . . . by, for example,

providing a copy of the policy to employees at the time of hire, [or] publishing

the policy in an employee handbook.” 29 C.F.R. § 541.603(d). 8 The Company

clearly satisfied the criterion of adequate communication in this case by

furnishing its handbook.

      Ms. Ellis’s second argument on this score is even more tenuous: she claims

that “[the Company’s] policy did not contain a complaint mechanism” and,



      7
             In fact, though decided in contexts different than this one, several
cases expressly discussing incorporation by reference undermine Ms. Ellis’s
argument. See, e.g., Doe v. Harris, 535 F. App’x 630, 631 (9th Cir. 2013)
(declaring that California’s “general rule” is that plea agreements “will be deemed
to incorporate . . . the existing law” (internal quotation marks omitted)), cert.
denied, --- U.S. ----, 

(2014); Fortier v. Principal Life Ins. Co.,

, 238 (4th Cir. 2012) (finding incorporation of the Internal Revenue
Code appropriate when the insurance policy stated that expenses “[were]
deductible for Federal Income Tax purposes” (internal quotation marks omitted)).
      8
             While we are not suggesting that the Company’s employee handbook
is a model for FLSA compliance, the salient point is that the regulatory scheme
does not require that a policy be flawless. As long as the applicable policy
complies with governing federal law, as the Company’s deductions policy does,
the employer is entitled to retain the exemption. See generally 29 C.F.R.
§ 541.603(d).

                                         27
therefore, the Company could not retain the exemption. Aplt. Opening Br. at 32.

This contention is patently incorrect. Our review of the employee handbook

makes clear that it contemplates at least three species of employee complaints.

An employee who experiences or observes harassment “must promptly notify the

supervisor”—or, if she believes a violation of the law has occurred, she “is

directed to advise the [Company] President.” Aplt. App. at 383. If an employee

suspects her pay has been reduced because of a “cash shortage or gas skip,” she

“should notify [her] supervisor or the Company President immediately.” 

       Finally—and most germane to this case—the handbook has a section

discussing “open door communications” that provides, “If you have a problem,

complaint, or need help with anything related to your job, . . . [y]ou may . . .

[t]alk to your supervisor; or . . . if you are not satisfied with the results of such a

discussion, please contact the President.” 

(capitalization altered)

(emphases added). In light of these provisions, we reject Ms. Ellis’s bald

assertion regarding the absence of a complaint mechanism and reiterate our view

that the Company’s deductions policy permits it to retain the challenged

exemption.

       For these reasons, we hold that the district court correctly found no

evidence of an “actual practice” of making improper deductions or of a “policy”

of making improper deductions by the Company. Therefore, the district court

                                            28
properly declined Ms. Ellis’s invitation to strip the Company, under the authority

of subsection (a), of its exemption as to Ms. Ellis.

                                          3

      Next, Ms. Ellis entreats us to reverse the district court based upon its

purported “evasion of the spirit of the FLSA,” Aplt. Opening Br. at 19—that is,

its failure to consider her employment conditions sufficiently oppressive. She

complains, “Reading the District Court’s order one could conclude that an

employer can adopt a hard floor of fifty hours and assign to the employee a

virtually unlimited amount of work,” 

and that such an employer could

still evade the FLSA’s overtime requirements. The crux of her argument,

however, is that the district court relied upon improper precedent in making its

determinations. Having carefully reviewed the district court’s ruling, we reach a

different conclusion.

      Ms. Ellis challenges the district court’s reliance upon various DOL

interpretations of the FLSA’s implementing regulations, claiming that the court

impermissibly cited DOL statements and opinion letters in order to avoid

addressing her allegedly “extraordinary” working conditions. 

In

lodging this argument, though, she overlooks the deference that we and other

courts have shown to the DOL’s interpretation of its own regulations. Indeed,

seventy-one years ago, we held that




                                          29
             Congress, in effect, provided that employees should be exempt
             who fell within certain general classifications as rationally and
             reasonably defined and made certain by the [DOL’s Wage and
             Hour] Administrator. The general standard was laid down by
             Congress[ ]—bona fide executive, administrative, or professional
             capacity. The policy was made manifest by Congress[ ]—that of
             precise definition and delimitation by a reasonable and rational
             regulation defining and delimiting the general terms. The
             delegation of such a power to define and delimit has been
             sustained by the decisions of the Supreme Court[.] Congress has
             laid down a general standard and manifest a policy and within the
             framework thereof has delegated to the Administrator the duty to
             supply the details.

Walling v. Yeakley, 

, 832 (10th Cir. 1944) (footnote omitted). In

looking to the DOL’s Wage and Hour Administrator for such interstitial

particulars, we have subsequently recognized the permissibility of deferring to the

DOL’s interpretations of the FLSA scheme by way of opinion letters. See In re

(“The DOL considers these opinion letters to be

rulings . . . . [O]pinion letters are entitled to great weight when they interpret the

DOL’s own (ambiguous) regulations.” (citations omitted) (internal quotation

marks omitted)); see also Rodriguez v. Whiting Farms, Inc., 

, 1189

(10th Cir. 2004) (“[W]e may look to such opinions for guidance. Such

interpretations are entitled to respect.” (internal quotation marks omitted)).

      And, in this same vein, the Supreme Court has established the propriety of

seeking interpretive guidance from the DOL’s preambles to its regulations. See

Christopher v. SmithKline Beecham Corp., --- U.S. ----, 

, 2162–63

(2012) (interpreting the term “employee employed in the capacity of outside

                                          30
salesman” in an FLSA exemption by looking to the DOL’s regulations, and also

finding that “[a]dditional guidance concerning the scope of the outside salesman

exemption can be gleaned from . . . the preamble to the 2004 regulations”

(internal quotation marks omitted)); see also Ramos v. Baldor Specialty Foods,

Inc., 

, 560–61 (2d Cir. 2012) (interpreting the term “a customarily

recognized department or subdivision” in FLSA’s executive exemption, and

looking to the preamble to the 2004 regulation as an interpretive guidepost).

      In light of the foregoing authority, it is clear that the district court was

entitled to consult DOL interpretations when ruling on the potentially ambiguous

concept of intent to pay a salary. In particular, the court did not err in reasonably

relying on the preamble to the regulations or a 2006 DOL opinion letter. See In

re 

(“[W]e need not take a jaundiced view of the

DOL’s own interpretation of ‘salary.’”). Both agency interpretations provide

strong justification for the district court’s conclusion that Ms. Ellis’s purportedly

unduly burdensome job requirements did not run afoul of the Company’s duties

under the FLSA.

      We are not unsympathetic to the hours of diligent work Ms. Ellis provided

to the Company as a store manager. Nonetheless, it is pellucid that she was aware

of this requirement throughout her employment—indeed, she admits that the

Company’s leadership regularly apprised store managers of the Pay Plan’s

position on work hours. Critically, Ms. Ellis has not met her burden of proving

                                          31
that the fifty-hour-week minimum or the directive to track her hours expressly

violates the FLSA. The DOL made clear, in unveiling amended salary-basis

regulations in 2004, that “employers, without affecting their employees’ exempt

status, may take deductions from accrued leave accounts; may require exempt

employees to record and track hours; may require exempt employees to work a

specified schedule; and may implement across-the-board changes in schedule

under certain circumstances.” Defining and Delimiting the Exemptions for

Executive, Administrative, Professional, Outside Sales and Computer Employees,

69 Fed. Reg. 22,122, 22,178 (Apr. 23, 2004) (emphasis added). 9 It is difficult to

harmonize that language with Ms. Ellis’s suggestion that the DOL’s interpretation

lacked the “power to persuade” the district court (or, us, for that matter). Aplt.

Opening Br. at 16 (quoting Christensen v. Harris Cnty., 

, 587

(2000)) (internal quotation marks omitted). 10 In our view, the district court’s

choice to credit the above-cited statements of the DOL was entirely reasonable.

      Similarly, we are not troubled by the district court’s citation to a 2006 DOL

opinion letter in its assessment. This particular letter was issued in response to an



      9
           See also U.S. Dep’t of Labor, Wage & Hour Div., Op. Letter No.
FLSA2005-5 (Jan. 7, 2005), http://www.dol.gov/whd/opinion/flsa.htm (follow
“Timekeeping system and 29 CFR 541.602” hyperlink).
      10
             Christensen is distinguishable from Ms. Ellis’s case in that the
Supreme Court was concerned about “permit[ting] the agency, under the guise of
interpreting a regulation, to create de facto a new 

.
We harbor no such concern here.

                                         32
employer’s “wish[ ] to require exempt employees to work either 45 or 50 hours a

week” and “to require . . . that exempt employees make up work time lost due to

personal absences of less than a day.” Aplt. App. at 33 (DOL Op. Letter No.

FLSA2006-6, dated Mar. 10, 2006). The employer’s concern was not whether it

could effect any salary deductions for employees who violated these proposed

rules, but whether “consistent failure to observe the proposed requirements

[c]ould result in discipline.” 

marks omitted). The DOL

opined that the employer could implement the requirements with impunity—i.e.,

under those hypothetical facts, the employer would not lose the exemption. More

to the point, the DOL advised that

             [t]he number of hours worked by an employee . . . is a matter to
             be determined between the employer and the employee. An
             employer may require an exempt employee to make up work time
             lost due to personal absences of less than a day without loss of
             the exemption . . . . As the preamble to the final rule explains,
             an employer may require an exempt employee to do things such
             as to record and track hours and to work a specified schedule
             without affecting the employee’s exempt status.

v. Ingalls Shipbuilding, Inc., 

(5th Cir. 2000)).

These principles find support in both binding and persuasive caselaw. See, e.g.,

(“[T]he fact that the [employees’] paystubs indicated the

number of hours covered is . . . not inconsistent with salaried status. Such an

accounting of hours is necessary . . . .”); see also 

(noting that “the facts pertaining to [the] plaintiffs’ pay are reviewed in light of


                                          33
Aaron”); accord Guerrero v. J.W. Hutton, Inc., 

, 835–36 (8th Cir.

2006); Renfro v. Ind. Mich. Power Co., 

, 516 (6th Cir. 2004).

      Having failed to advance a compelling—let alone plausible—justification

for us to discount the district court’s consultation of DOL interpretations of the

FLSA, Ms. Ellis cannot prevail on this score. This particular strand of reasoning

only strengthens our view that affirmance is the correct result.

                                          4

      Ms. Ellis also argues that the district court mishandled her attempt to

demonstrate, via her own affidavit, a violation of the salary-basis test. The

affidavit detailed Ms. Ellis’s view that the Company maintained an impermissible

policy of making improper deductions—which, if true, would support her

conclusion that the salary-basis test was violated. Ms. Ellis relied heavily on this

affidavit to support the assertion that the Company could not have intended to pay

her a salary when the deduction at issue was intentional and made pursuant to its

policy. In granting the Company’s summary-judgment motion, the district court

noted that Ms. Ellis had “support[ed] her assertion only with [this] affidavit

which, in turn, merely recite[d] alleged statements of another [Company] manager

that arguably substantiate[d] Ellis’s description of [a] ‘policy’ of docking salaried

employees’ pay.” Aplt. App. at 453 (emphasis added). The court found Ms.

Ellis’s affidavit conclusory and self-serving. And we conclude that it was entitled

to do so.

                                         34
      As a threshold matter, Ms. Ellis is mistaken in her view that “the Court

reviews this issue de novo.” Aplt. Opening Br. at 32. It is well-settled that a

district court’s assessments of the admissibility and probative value of affidavits

at the summary-judgment phase involve, at bottom, evidentiary determinations; as

such, they are reviewed for an abuse of discretion. See Sports Racing Servs., Inc.

v. Sports Car Club of Am., Inc., 

, 894 (10th Cir. 1997) (“Like other

evidentiary rulings, we review a district court’s decision to exclude evidence at

the summary judgment stage for abuse of discretion. . . . We cannot say that the

district court abused its discretion in finding [the plaintiff’s] affidavit a sham and

excluding it.” (citations omitted)); accord Mitchael v. Intracorp, Inc., 

, 854–55 (10th Cir. 1999).

      Affidavits must contain certain indicia of reliability. “Unsubstantiated

allegations carry no probative weight in summary judgment proceedings[; they]

must be based on more than mere speculation, conjecture, or surmise.” Bones v.

Honeywell Int’l, Inc., 

, 875 (10th Cir. 2004) (citation omitted).

Information presented in the nonmovant’s affidavit must be “based on personal

knowledge and [must set] forth facts that would be admissible in evidence. We

do not consider conclusory and self-serving affidavits.” Garrett v. Hewlett-

Packard Co., 

, 1213 (10th Cir. 2002) (citation omitted) (quoting

Murray v. City of Sapulpa, 

, 1422 (10th Cir. 1995)) (internal

quotation marks omitted).

                                          35
      Because the statements in Ms. Ellis’s affidavit constitute inadmissible

hearsay, they do not pass muster. In her affidavit, Ms. Ellis alleged that when she

presented her time sheet for the week of her relevant absence (reflecting 40.91

hours worked), the Company’s payroll clerk asked whether she “wanted to take

sick pay or vacation.” Aplt. App. at 371. Ms. Ellis claimed that she replied,

“[D]o whatever you need to do.” 

marks omitted). She then

described the following alleged conversation she had with the Company’s office

manager after she realized she did not receive her full salary:

             I asked Kathy “Do we have an HR Department?” She said that
             would be her or Dan. I explained that I had lost a vacation day
             because I was sick. I was upset and asked whether that was what
             I got for all the 60 and 70 hour[ ] weeks I worked. . . . She said
             that is their “policy” and that is what she had to do. She said “I
             can get that straightened out. [The payroll clerk] should have
             done sick pay.”

Ellis claimed that other Company leaders “were extremely

upset that [she] had questioned the policy.” 

challenges the district

court’s conclusion that these particular statements would be inadmissible at trial.

      Basic hearsay jurisprudence makes clear that Ms. Ellis could not have

introduced these statements at trial: she offered statements pertaining to her

conversation with the office manager in order “to prove the truth of the matter

asserted in the statement[s].” Fed. R. Evid. 801(c)(2). Indeed, as Ms. Ellis notes

in her opening brief, she presented the alleged statement “[s]he said that is their

‘policy’ and that is what she had to do,” Aplt. App. at 371, in order to prove that

                                          36
“[the Company’s] policy required salaried employees to use a full vacation day or

a full sick day when they did not work fifty hours,” Aplt. Opening Br. at 23.

Stated otherwise, Ms. Ellis offered the manager’s alleged assertion that the

Company had a policy—that would be legally impermissible—to prove that the

Company in fact had such a policy. This is classic hearsay; it consequently

“cannot support [Ms. Ellis’s] opposition to summary judgment.” Llewellyn v.

Allstate Home Loans, Inc., 

, 1180 (10th Cir. 2013); see Jaramillo v.

Colo. Judicial Dep’t, 

, 1314 (10th Cir. 2005) (“Hearsay testimony

that would not be admissible at trial is not sufficient to defeat a motion for

summary judgment.”).

      Nevertheless, Ms. Ellis contends that the office manager’s statements are

outside the ambit of Federal Rule of Evidence 802 because they are defined by

the federal rules as non-hearsay. See Fed. R. Evid. 801(d). She urges us to

consider those statements admissions of a party opponent under Rule

801(d)(2)(D). But that would be error; “[u]nder our controlling precedent, an

employee’s statements are not attributable to her employer as a party-opponent

admission in an employment dispute unless the employee was ‘involved in the

decisionmaking process affecting the employment action’ at issue.” Johnson v.

Weld Cnty., 

, 1208–09 (10th Cir. 2010) (quoting 

). Ms. Ellis’s contention that “[t]he office manager spoke as [the

Company’s] representative on the subject of the pay deduction,” Aplt. Opening

                                          37
Br. at 23, finds no support in the record. And reciting the office manager’s

alleged statement that the Company’s human resources department “would be

her,” Aplt. App. at 371, to prove that the office manager was involved in “the

employment action at issue,” 

(internal quotation marks

omitted), is precisely what the hearsay rule forbids. Nor is this problem cured by

Ms. Ellis’s assertion that “[the Company] . . . does not offer any evidence” in

rebuttal. Aplt. Reply Br. at 12. The Company had no need to do so, as the record

is silent on the office manager’s involvement in the decisionmaking process for

Ms. Ellis’s pay deduction—except for the inadmissible “evidence” offered in Ms.

Ellis’s affidavit.

       Ms. Ellis’s remaining argument regarding her affidavit—that we should

reject the “more stringent standard” of Johnson and Jaramillo, Aplt. Reply Br. at

12—is also unconvincing. In her view, the better case upon which to rely is

Aliotta v. National Railroad Passenger Corp., 

(7th Cir. 2003). She

seizes upon the Seventh Circuit’s description there of “general employment

cases,” Aplt. Reply Br. at 12, as “[having] law . . . [that] is somewhat muddled

because the great bulk of cases interpreting what is within an employee’s ‘scope

of employment’ deals with employment discrimination.” 

–62. This hardly seems relevant, because even if Aliotta were an overtime-

pay lawsuit (which it was not), or binding upon us (which it is not), the court in

that case clearly determined that “the subject matter of the admission [must]

                                         38
match the subject matter of the employee’s job description.” 

With no

record evidence detailing the office manager’s job description other than Ms.

Ellis’s inadmissible hearsay statements, we could not make a similar matching

determination here.

      We likewise agree in material respects with the district court’s resolution of

Ms. Ellis’s arguments regarding the remaining affidavit statements—i.e., her

averments that she worked more than her time sheets indicated. More

specifically, we conclude that the district court’s reasoning was on target insofar

as it found that Ms. Ellis’s unsupported assertions were at best “conclusory,”

Aplt. App. at 454, and consequently of marginal probative value. See, e.g.,

. Indeed, a huge dollop of conjecture would be necessary

to divine from these averments a genuine issue of material fact. Cf. Aplt. App. at

373 (generally stating that “[The Company] has submitted documents . . . as

evidence that I didn’t work 50 hours or more during [a certain time period]. I did

work more than 50 hours during [that time period]. I know I did . . . .”). Perhaps,

as the district court noted, Ms. Ellis could have improved her cause by submitting

affidavits of other managers or co-workers. Her choice not to do so sounds the

death knell for this argument.

      For all of the reasons detailed above, we affirm the judgment of the district

court insofar as it based its grant of summary judgment on a finding that the

Company did not violate the salary-basis test, as explicated in subsection (a) and

                                         39
informed by subsection (d). We conclude that the Company intended to pay Ms.

Ellis a salary, which means that she is exempt and may not receive overtime pay

under the FLSA.

                                         B

      Next, Ms. Ellis insists that the district court’s ruling cannot stand because

its interpretation of the window-of-correction defense explicated in subsection (c)

“conflicts with the principles of statutory interpretation and the narrow

interpretation of FLSA exemptions.” Aplt. Opening Br. at 29. We disagree and

hold that the district court did not err in concluding that the Company was

entitled to rely on this defense.

      Subsection (c) provides that “[i]mproper deductions that are either isolated

or inadvertent will not result in loss of the exemption for any employees subject

to such improper deductions, if the employer reimburses the employees for such

improper deductions.” 29 C.F.R. § 541.603(c) (emphases added). Here, three key

facts prompted the district court to rely upon this provision. First, Ms. Ellis

identified one improper paycheck deduction—$31.20 taken on April 3, 2012.

Second, the Company reimbursed Ms. Ellis in the amount of $332.88 shortly after

making that improper deduction. Third, Ms. Ellis never disputed that this was an

isolated deduction or that she was reimbursed in a timely manner. Consequently,

given these undisputed facts, and with an eye on the “either . . . or” language of

subsection (c), the district court reasoned that “Ellis’s argument is untethered to

                                         40
the language of the exception . . . . To the extent Ellis means to equate

‘unintentional’ with ‘inadvertent,’ . . . the defense can equally be satisfied when

deductions are ‘isolated.’” Aplt. App. at 459 (citation omitted).

      Resisting this conclusion, Ms. Ellis argues that subsection (c) should not be

“read in isolation from the remainder of the regulation,” Aplt. Opening Br. at 30,

but instead, if read in context with subsections (a) and (d), “should . . . apply only

when there is no other evidence of the employer’s intent, such as a policy

allowing deductions,” 

Otherwise, she claims, subsections (a) and (d)

would be nullified because an employer would have no need for a clear policy

prohibiting improper deductions; instead, it could “intentionally dock an exempt

employee’s pay and avoid all liability for overtime simply by reimbursing the

employee.” 

she argues that subsection (c) is only triggered where

the improper deduction is both isolated and unintentional.

      We cannot accept Ms. Ellis’s statutory-construction argument. Beginning

with the text of subsection (c), as we must, see United States v. Lamirand, 

, 1094 (10th Cir. 2012), it is apparent that this language renders the

window-of-correction defense available to an employer who has made

“[i]mproper deductions that are either isolated or inadvertent,” but has

“reimburse[d] the employees for such improper deductions.” 29 C.F.R.

§ 541.603(c) (emphasis added). “[G]iven the clarity of that provision’s language,

[we] could end [our analysis] there as well.” Tapia v. United States, --- U.S. ----,

                                          41


, 2388 (2011). We find subsection (c) particularly clear insofar as

it employs disjunctive terminology. The word “either” may be “used as a

function word before two or more coordinate words . . . joined usu[ally] by or to

indicate that . . . two or more alternatives . . . are equally applicable.” 

. Its partner, “or,” is a “choice between alternative things.” 

Thus, it is patent that the district court could choose between “isolated”

and “inadvertent” deductions and that both alternatives could satisfy the statute.

The court’s conclusion that “[t]he one-time deduction . . . was certainly

‘isolated,’” Aplt. App. at 458–59, means that it chose the “isolated” alternative as

the way the Company prevailed on the window-of-correction defense. This was

proper, for “isolated” means “solitary” or “occurring alone or once.” 

(capitalization omitted).

      The district court thus did not err in concluding that the lone improper

deduction affecting Ms. Ellis was “isolated”—even without addressing whether

that improper deduction was also “inadvertent.” Generally speaking, we

understand statutes containing disjunctive language to require that only one of the

listed requirements must be satisfied. See, e.g., Graham v. Hartford Life &

Accident Ins. Co., 

, 1353 (10th Cir. 2009) (finding that a definition

stating that “a plan need only be established or maintained” would be met if the

party either established the plan or maintained the plan (internal quotation marks

omitted)). Additionally, we have said that “effect must be given, if possible, to

                                           42
every word, clause and sentence of a statute . . . [which means we] therefore

cannot ignore the use of the ‘or’ in” a statute and its accompanying regulations.

Knutzen v. Eben Ezer Lutheran Hous. Ctr., 

, 1348–49 (10th Cir.

1987) (citation omitted) (quoting 2A N. Singer, Sutherland on Statutory

Construction § 46.06, at 104 (4th ed. 1984)) (internal quotation marks omitted).

      Ms. Ellis correctly observes that courts should consider statutory and

regulatory text as a whole, see Bethesda Hosp. Ass’n v. Bowen, 

, 405

(1988), but she does not do so herself. Notably, her argument would contravene

another subpart of the regulation—29 C.F.R. § 541.603(e)—which states, “This

section shall not be construed in an unduly technical manner so as to defeat the

exemption.” Cf. First Nat’l Bank of Durango v. Woods (In re Woods), 

, 702 (10th Cir. 2014) (discouraging interpretation that would effectively

mean “that the exception would almost never apply” and concluding that

Congress could not have intended such a “limiting effect” that “would almost

entirely eviscerate the exception”). In that regard, a leading authority on

statutory construction instructs us that:

             [i]t is not a proper use of the [“whole-text”] canon to say that
             since the overall purpose of the statute is to achieve x, any
             interpretation of the text that limits the achieving of x must be
             disfavored. . . . [L]imitations on a statute’s reach are as much a
             part of the statutory purpose as specifications of what is to be
             done.




                                       43
Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal

Texts 168 (2012). This concept is particularly apt here because Ms. Ellis

repeatedly argues that the FLSA is a remedial statute designed to be resolved “in

favor of employee coverage.” Aplt. Opening Br. at 14; accord 

The fact that one purpose of the FLSA is to ensure overtime pay for

non-exempt employees does not preclude the possibility that an employer may

“intentionally dock an exempt employee’s pay and avoid all liability for overtime

simply by reimbursing the employee.” 

Such a situation does not

necessarily abuse the window-of-correction defense or eviscerate the employee’s

exempt status—provided, of course, that the intentional (i.e., not inadvertent)

deduction was “isolated.”

      While we note that some of our sister circuits have intimated that this

defense can only apply when an employer has made an innocent mistake, 11 we

remind Ms. Ellis that these decisions are not binding upon us. Our own precedent

makes clear that “[t]he ‘window of correction’ provided by [the predecessor

version of subsection (c)] allows employers to treat otherwise eligible employees

as salaried, regardless of the employer’s one-time or unintentional failure to

adhere to” the statute’s prohibition on improperly deducting salaried employees’

pay based on quantity of work performed. 

(emphasis

      11
            See, e.g., Klein v. Rush-Presbyterian-St. Luke’s Med. Ctr., 

, 287–88 (7th Cir. 1993); Martin v. Malcolm Pirnie, Inc., 

, 616
(2d Cir. 1991).

                                         44
added). And our holding in Spradling leads ineluctably to the conclusion that the

district court properly allowed the Company to rely upon the window-of-

correction defense regarding its lone improper deduction to Ms. Ellis’s pay.

                                          C

      In conclusion, we hold that subsections (a) and (c) can be read

harmoniously. We further hold that the Company can rely on both provisions

here: as to the former, there was no practice or policy of making inappropriate

pay deductions for salaried employees; and, as to the latter, the one time the

Company acted improperly as regards Ms. Ellis, that action was isolated. On

these grounds, we affirm.

                                          IV

      Finally, we address the district court’s denial of Ms. Ellis’s motion to defer

summary judgment for the gathering of additional discovery under Federal Rule

of Civil Procedure 56(d). 12 In doing so, we are mindful of the Supreme Court’s

directive that “summary judgment [should] be refused where the nonmoving party

has not had the opportunity to discover information that is essential to his

opposition.” Anderson v. Liberty Lobby, Inc., 

, 250 n.5 (1986). We

likewise note our obligation, under the abuse-of-discretion rubric, to “defer to the

district court’s judgment so long as it falls within the realm of [the] rationally

      12
            Until December 2010, the substance of Rule 56(d) was embodied in
Rule 56(f). See Cohlmia v. St. John Med. Ctr., 

, 1286 n.10 (10th
Cir. 2012).

                                          45
available choices.” Shook v. Bd. of Cnty. Comm’rs, 

, 603 (10th Cir.

2008) (internal quotation marks omitted); accord Barber ex rel. Barber v. Colo.

Dep’t of Revenue, 

, 1228 (10th Cir. 2009). Here, as we explain

below, we lack any cogent basis to disturb the reasoned judgment of the district

court, and we conclude that it properly applied the governing rule.

      Rule 56(d) allows the nonmovant to “show[ ] by affidavit or declaration

that, for specified reasons, [she] cannot present facts essential to justify . . .

opposition [to a motion for summary judgment].” Fed. R. Civ. P. 56(d). The

district court may then: “(1) defer considering the motion or deny it; (2) allow

time to obtain affidavits or declarations or to take discovery; or (3) issue any

other appropriate order.” Id.; accord 

. The party

seeking to enlarge the discovery period under this rule “must present an affidavit

that identifies ‘the probable facts not available and what steps have been taken to

obtain these facts. The [party] must also explain how additional time will enable

[her] to rebut the movant’s allegations of no genuine issue of material fact.’”

(quoting Trask v. Franco, 

, 1042 (10th

Cir. 2006)); see also Been v. O.K. Indus., Inc., 

, 1235 (10th Cir.

2007) (“[A] party at a minimum must ask the court to refrain from acting on the

summary judgment request until additional discovery can be conducted.” (internal

quotation marks omitted)). We expect Rule 56(d) motions to be robust, and we

have observed that “[an] affidavit’s lack of specificity” counsels against a finding

                                            46
that the district court abused its discretion in denying a request for additional

discovery under the rule. 

.

      Ms. Ellis attempted to satisfy the foregoing directives in the district court

by submitting the affidavit of Attorney Bradley Sherman. According to that

affidavit, the probable facts unavailable to Ms. Ellis included information in “pay

records, deposition testimony, and other documents” that would “show a wide

ranging policy and practice of making improper deductions” and allow Ms. Ellis

“to determine if any deductions were in fact ‘isolated’ or not.” Aplt. App. at 319

(Aff. of Bradley Sherman, dated Oct. 31, 2012). The affidavit further indicated

that Ms. Ellis had participated in the discovery conference required by Federal

Rule of Civil Procedure 26(f), which we construe as an averment that Ms. Ellis

had taken a step to obtain the allegedly unavailable facts. Finally, the affidavit

explained that additional time was essential to the litigation not only because of

its early procedural posture (i.e., that Ms. Ellis simply had not had adequate time

to investigate these facts), but also because, in Ms. Ellis’s view, “the analysis

under 29 C.F.R. § 541.603 extends beyond a single employee and instead

encompasses the employer’s pay practices regarding a broad range of employees.”

      But the district court was not persuaded that such a wide swath of

discovery was necessary. In fact, it found that Attorney Sherman’s affidavit

failed to demonstrate the need for any further discovery beyond Ms. Ellis’s pay

                                          47
and time records. Reasoning that this affidavit only raised a speculative hope of

unearthing evidence sufficient to prevail at summary judgment, the court denied

Ms. Ellis’s motion. We are satisfied—notwithstanding Ms. Ellis’s arguments on

appeal—that this ruling involved no abuse of discretion.

      We first address Ms. Ellis’s argument that she was unable to rebut the

Company’s “isolated-deduction” allegations without a thorough “analysis of [the

Company’s] pay practices regarding other employees.” 

Ms. Ellis

apparently believes that subsection (a) always contemplates an employer-wide

assessment—and, to be sure, subsection (a) lists among its nonexhaustive factors

“the number and geographic location of employees whose salary was improperly

reduced.” 29 C.F.R. § 541.603(a). However, the inclusion of that particular

factor among many does not render it mandatory. 13 Thus, the district court was

not legally required under all circumstances to consider any improper deductions

involving other employees. And, on these facts, it did not abuse its discretion in

declining to grant Ms. Ellis additional time to conduct discovery relating to any

improper deductions made from other employees’ salaries. In this regard, we note

that, at the precise moment the Company filed its motion for summary judgment,

the scope of the lawsuit was explicitly limited to one named plaintiff: Ms. Ellis.

      13
              Indeed, we believe the pertinent regulatory text—“[t]he factors to
consider . . . include, but are not limited to,” 29 C.F.R. § 541.603(a) (emphases
added)—does not suggest “[b]y its plain language” that subsection (a) “mandates
consideration of the employer’s pay practices towards all employees,” as Ms.
Ellis contends, Aplt. Reply Br. at 19.

                                         48
This remained true throughout all proceedings germane to this appeal, as

evidenced by the district court’s contemporaneous rulings on the Company’s

summary-judgment motion (which it granted) and Ms. Ellis’s class-certification

motion (which it denied). Thus, the district court did not commit legal error or

otherwise abuse its discretion in declining to allow additional time to conduct

other-employee discovery.

      Next—and relatedly—we reject Ms. Ellis’s contention that the procedural

posture of her case justified relief under Rule 56(d). More to the point, one of the

reasons she filed a Rule 56(d) motion was that she was convinced that the

Company’s then-pending summary-judgment motion was directly aimed at

thwarting an anticipated class action. The timeline of this litigation does not

conclusively foreclose the possibility that the Company actually was motivated in

this manner, but it does cast substantial doubt on the notion: the Company moved

for summary judgment on October 11, 2012; Ms. Ellis moved to certify a class on

October 18, 2012; and Ms. Ellis moved for relief under Rule 56(d) on October 31,

2012. Specifically, the problem for Ms. Ellis is her failure to attach any legal

significance to the fact that the Company’s motion for summary judgment

predates her motion for class certification—viz., she has presented no convincing

argument that the district court was somehow constrained to rule on the class-

certification motion before the summary-judgment motion. We submit that she

cannot make this showing because no such obligation existed. See Crowley v.

                                         49
Montgomery Ward & Co., 

, 879 (10th Cir. 1978) (affirming the

district court’s decision to dispose of case on summary judgment even though “no

ruling was made on the certification of the class”); cf. 7AA Charles Alan Wright

et al., Federal Practice and Procedure § 1785.3, at 457 (3d ed. 2005) (“[I]t

generally seems agreed that certification at the judgment stage is not

automatically prohibited.”). Consequently, given the generous nature of the

abuse-of-discretion standard, this argument lends Ms. Ellis no succor.

      On this record, we would be hard-pressed to find fault with the district

court’s conclusion that Ms. Ellis required no further discovery to respond to the

Company’s summary-judgment motion. By that point in the case, the Company

had provided time sheets and pay records for the relevant time period set forth in

Ms. Ellis’s complaint, which was all Ms. Ellis could reasonably have required to

rebut the Company’s allegations as they related to her, the sole named plaintiff.

Those materials unmistakably revealed that the Company made only one

deduction to Ms. Ellis’s pay in the amount of $31.20—a fact Ms. Ellis has never

disputed. Moreover, while the summary-judgment and class-certification motions

were pending, the Company amended its answers to Ms. Ellis’s interrogatories to

reflect that it had previously made pay deductions for two other store managers.

This amendment did not benefit Ms. Ellis, though, because both cited subtractions

stemmed from the employees’ use of “a full sick day,” Aplt. App. at 423 (Resp. to




                                         50
Pl.’s Mot. to Supplement R., Ex. A, filed Feb. 22, 2013), which is permissible

under 29 C.F.R. § 541.602(b)(2).

      In light of Ms. Ellis’s obvious failure to marshal evidence of improper

deductions, as the FLSA regulations contemplate, it was hardly irrational for the

district court to have found her Rule 56(d) motion speculative. We have held that

a litigant’s unsupported averments cannot justify deferring a ruling on—let alone

denying altogether—a motion for summary judgment. See Meyer v. Dans un

Jardin, S.A., 

, 537 (10th Cir. 1987) (“The nonmoving party has the

burden of showing . . . how additional time will enable [her] to rebut the movant’s

allegations.” (emphasis added)); accord Culver v. Town of Torrington, 

, 1458–59 (10th Cir. 1991). 14 And, in a similar procedural context in Lewis

v. City of Ft. Collins, we reached a result that further undermines Ms. Ellis’s

position:

             In her brief in response to the motion for summary judgment,
             plaintiff asserts that there has been “insufficient time for
             discovery” and that “[t]he parties have not begun the discovery
             process.” Thus, plaintiff argued that appellants’ motion for
             summary judgment was “premature” and “improper.” Within the
             affidavit itself, plaintiff repeatedly asserts that she “will be able
             to show” certain facts relative to her claims . . . . These
             inspecific references are simply insufficient to meet [Rule 56(d)]

      14
             In the same sense, persuasive authority from our sister circuits
suggests that if Ms. Ellis had no viable FLSA claim, she was not an appropriate
representative in any event for a hypothetical class of similarly situated
employees. See Camesi v. Univ. of Pittsburgh Med. Ctr., 

, 248 (3d
Cir. 2013); White v. Baptist Mem’l Health Care Corp., 

, 878 (6th
Cir. 2012); In re Family Dollar FLSA Litig., 

, 519 (4th Cir. 2011).

                                              51
             muster . . . . Plaintiff has not attempted to particularize her
             request for discovery. She has not explained how any specific
             documents or depositions will aid in rebutting defendants’
             showing of objective reasonableness. [Rule 56(d)] is not a
             license for a fishing expedition . . . .

, 758–59 (10th Cir. 1990) (first alteration in original) (emphasis

added). Guided by these cases, we believe denying Ms. Ellis’s Rule 56(d) motion

was entirely appropriate.

      Reduced to its essence, then, Ms. Ellis’s Rule 56(d) challenge on appeal

comprises her view that because the district court could have granted the motion,

it should have done so. That is patently not the law; “[u]nder an abuse of

discretion standard, a trial court’s decision will not be disturbed unless the

appellate court has a definite and firm conviction that the lower court made a

clear error of judgment or exceeded the bounds of permissible choice in the

circumstances.” Wright ex rel. Trust Co. of Kan. v. Abbott Labs., Inc., 

, 1235 (10th Cir. 2001) (internal quotation marks omitted). No such error or

impermissible choice is evident from this record. Accordingly, we affirm the

district court’s denial of Ms. Ellis’s Rule 56(d) motion.

                                          V

      For the reasons explicated above, we AFFIRM.




                                          52