The English language contains many examples of homonyms-"words that have the same sound and often the same spelling but differ in meaning ...." The American Heritage Dictionary of the English Language 843 (5th ed. 2011). The words "letter" (a symbol in the alphabet or a note) and "bark" (a dog's cry or the outside covering of a tree trunk), for example, both fit the bill (as does "bill," for that matter).

But the language of the law has its share of homonyms, too, and in this case we confront a couple of subtle ones. Specifically, this case turns on the difference in meaning between the term "employer" under the Fair Labor Standards Act, 29 U.S.C. § 203(d) ("FLSA"), and that same term under the general common law. Both definitions require us to ask how much "control" Defendant-Appellant Citrus Consolidated Limited Partnership ("Consolidated Citrus" or "the company") exerted over a group of farm workers who performed labor on Consolidated Citrus's groves. But the answer to that question depends, in turn, on the meaning of "control," which is also a legal homonym. Like "employer," it also has different meanings under the FLSA and the common law.

Plaintiffs-Appellees are migrant workers in the United States under the federal government's H-2A visa program. Ruiz Harvesting, Inc. ("Ruiz Harvesting")-a farm-labor contractor and a separate entity from Defendant-Appellant Consolidated Citrus-hired Plaintiffs to pick fruit at Consolidated Citrus's groves. Then, apparently without Consolidated Citrus's knowledge, Ruiz Harvesting forced Plaintiffs to *1116kick back a portion of their paychecks under threat of deportation.

Based on these circumstances, Plaintiffs sued Ruiz Harvesting, Basiliso Ruiz (the owner of Ruiz Harvesting), and Consolidated Citrus for violations of the FLSA and for breach of contract. Both Ruiz Harvesting and Ruiz settled with Plaintiffs and ceased to be parties to this lawsuit. As for Consolidated Citrus, the district court held a bench trial and found it liable for both causes of action.

Then this case made its first appearance before us. Garcia-Celestino v. Ruiz Harvesting, Inc. , 843 F.3d 1276 (11th Cir. 2016) (" Garcia-Celestino I "). We upheld Consolidated Citrus's liability on the FLSA claim, but we remanded the matter to the district court on the breach-of-contract claim. We explained that the district court had mistakenly applied the definition of "employer" from the FLSA in determining whether Consolidated Citrus was a "joint employer" for purposes of resolving the breach-of-contract claim. See id. at 1284. Instead, we noted, that claim depends on the definition of "employer" under general common-law principles. See id. at 1289-90. So we remanded the case to the district court to determine whether Consolidated Citrus was an "employer" under the common-law definition of the term. Id. at 1293.

On remand, the district court again concluded that Consolidated Citrus was an "employer" for purposes of the breach-of-contract claim. Consolidated Citrus challenges that determination.

Our review of this case reveals that some confusion appears to exist concerning the practical ways in which the definitions of "employer" under the FLSA and of that same term under general common-law principles differ. So we take this opportunity to clarify that area of the law. And once we apply the common-law definition here, we conclude that Consolidated Citrus is not a joint employer for purposes of Plaintiffs' breach-of-contract claim since the company is not an "employer" under the common-law definition of that term. We therefore vacate the judgment of the district court.

I. BACKGROUND

We start with the relevant factual background, which we take from the district court's factual findings entered after a bench trial.

Between 2007 and 2009, Plaintiffs worked as manual laborers picking fruit at Consolidated Citrus's Florida groves, though, as we have noted, Consolidated Citrus did not hire Plaintiffs.1 Rather, Ruiz Harvesting did.

We pause to explain how that situation arose. As Mexican nationals, Plaintiffs received clearance to work in the United States through the federal government's H-2A visa program, which allows employers to hire foreign agricultural workers on a temporary basis. Under the program, employers must submit to the Department of Labor an application commonly referred to as a "clearance order" detailing the terms and conditions of their prospective workers' employment. By federal regulation, the clearance order becomes the employees' work contract by default if the employer does not draw up a separate contract for them. See 20 C.F.R. § 655.122(q) (2016).2

*1117Although Consolidated Citrus hired some of its laborers directly, it also engaged contractors to hire others. Ruiz Harvesting was one such contractor. Ruiz Harvesting recruited Plaintiffs, submitted clearance orders to the Department of Labor on their behalves, and ultimately hired them for work in Consolidated Citrus's groves. For work contracts, Ruiz Harvesting and Plaintiffs relied on only their clearance orders for each year at issue.

As for Consolidated Citrus, it had no role in deciding how much Ruiz Harvesting's workers would be paid. Rather, Consolidated Citrus simply paid Ruiz Harvesting for its total fruit production, and Ruiz Harvesting then determined payments to Plaintiffs.

But because Consolidated Citrus required all workers to be hired through the H-2A program, Ruiz Harvesting had to comply with a number of federal regulations governing the minimum pay its workers would receive. As relevant here, even though Ruiz Harvesting chose to pay its workers on a "piece-rate" basis, meaning a fixed rate for every container of fruit they picked, federal regulations still required each worker to receive a minimum amount each pay period. So if a worker's piece-rate earnings fell below the federally mandated minimum, Ruiz Harvesting had to pay that minimum amount, anyway.

In 2010, Plaintiffs brought suit alleging, among other things, violations of the FLSA and breach of contract. For starters, Plaintiffs sued Ruiz Harvesting and Ruiz, asserting that they forced the workers to pay them illegal kickbacks that impermissibly reduced the workers' take-home pay.3 More specifically, Plaintiffs averred, whenever a worker's piece-rate earnings fell below the federal minimum, Ruiz Harvesting paid the worker in full but then demanded repayment of the portion it had supplemented. To extract the cash kickback payments, Plaintiffs alleged, Ruiz Harvesting officials often threatened the workers with deportation.

This occurred despite the fact that Consolidated Citrus established a thorough auditing process to monitor Ruiz Harvesting's finances.

Based on the theory that Consolidated Citrus and Ruiz Harvesting were "joint employers" under the law, Plaintiffs also named Consolidated Citrus as a defendant in their lawsuit, contending the company was equally liable for Ruiz Harvesting's kickback scheme. Plaintiffs eventually settled with both Ruiz Harvesting and Ruiz.

Then they proceeded to trial against only Consolidated Citrus. The district court issued findings of fact and conclusions of law following a six-day bench trial. Ultimately, the court determined that Consolidated Citrus was a joint employer for purposes of both the breach-of-contract and FLSA claims. Based on these conclusions, the court found Consolidated Citrus liable for both claims.

Consolidated Citrus appealed, and a panel of this court affirmed in part and reversed in part. Garcia-Celestino I , 843 F.3d at 1295. We affirmed the district court's conclusion that Consolidated Citrus was a joint employer under the FLSA and therefore upheld Consolidated Citrus's liability under that statute. Id. at 1294-95. But we concluded that the district court used the wrong legal standard to determine whether Consolidated Citrus was a joint employer for purposes of the breach-of-contract claim. Rather than the FLSA's "economic dependency" test, we explained that the district court should have applied the definition of "employer" found in the common law of agency. Id. at 1295.

On remand, the district court analyzed its prior factual findings under the common-law definition of "employer" and once again determined that Consolidated Citrus was a joint employer for purposes of the breach-of-contract *1118claim. Consolidated Citrus now appeals.

II. STANDARD OF REVIEW

On review after a bench trial, we accept all of the district court's factual findings unless they are clearly erroneous, but we consider legal issues de novo . Id. at 1284 n.4 (citing Tartell v. S. Fla. Sinus & Allergy Ctr., Inc. , 790 F.3d 1253, 1257 (11th Cir. 2015) ). Whether a company is a joint employer raises a question of law. Id. (citing Aimable v. Long & Scott Farms , 20 F.3d 434, 440 (11th Cir. 1994) ).

III. DISCUSSION

As we have noted, the contracts at the center of Plaintiffs' breach-of-contract claims are Plaintiffs' clearance orders issued under the H-2A visa program, which, in turn, require compliance with the H-2A statutory and regulatory framework. That framework uses the term "employer." So we begin by reviewing the meaning of that term under the Immigration and Nationality Act ("INA"), as amended by the Immigration Reform and Control Act of 1986 ("IRCA"), which governs the H-2A visa program. See 8 U.S.C. § 1188.

Notably, although the INA and several federal regulations set out requirements for employers who take on H-2A workers, neither the statute nor any relevant regulation expressly defines the term "employer."

But the word "employer" does have a particular meaning in the common law. And as we explained in Garcia-Celestino I , where a federal statute contains a term with settled meaning under the common law, courts must presume Congress meant to import that meaning unless the statute says otherwise. 843 F.3d at 1289-90 (citing NLRB v. Amax Coal Co. , 453 U.S. 322, 329, 101 S.Ct. 2789, 69 L.Ed.2d 672 (1981) ). Since the INA does not define "employer," we concluded that Congress intended the statute to carry the definition of that term from the common law of agency. Id. Consequently, we reasoned, whether Plaintiffs' work contract makes Consolidated Citrus a "joint employer" under the relevant portions of the INA depends on the definition of "employer" taken from the general common law of agency. Id. at 1290.4

For that definition, we looked chiefly to Nationwide Mutual Insurance Company v. Darden , in which the Supreme Court articulated several factors relevant to determining whether an employer-employee relationship exists at common law. See 503 U.S. 318, 323-24, 112 S.Ct. 1344, 117 L.Ed.2d 581 (1992) (citing Cmty. for Creative Non-Violence v. Reid , 490 U.S. 730, 751-52, 109 S.Ct. 2166, 104 L.Ed.2d 811 (1989) ). Foremost among those factors, we observed, is "the hiring entity's 'right to control the manner and means by which *1119the product is accomplished.' " Garcia-Celestino I , 843 F.3d at 1292-93 (quoting Darden , 503 U.S. at 323, 112 S.Ct. 1344 ). See also Restatement (Second) of Agency § 220(1) (1958) (defining "servant" as someone "employed to perform services in the affairs of another and who with respect to the physical conduct in the performance of the services is subject to the other's control or right to control"); Clackamas Gastroenterology Assoc., P.C. v. Wells , 538 U.S. 440, 448, 123 S.Ct. 1673, 155 L.Ed.2d 615 (2003) (designating "the common-law element of control" as "the principal guidepost that should be followed" in determining joint-employer status); Crew One Prod., Inc. v. N.L.R.B. , 811 F.3d 1305, 1311 (11th Cir. 2016) (citing N.L.R.B. v. Associated Diamond Cabs, Inc. , 702 F.2d 912, 919 (11th Cir. 1983) ) (observing that among the common-law factors, control over employees should receive "special attention" in determining employer status).

Yet while the right to control is indispensable to our analysis and bears more weight than any other single factor, that consideration alone "is not dispositive." Reid , 490 U.S. at 752, 109 S.Ct. 2166. Rather, we must also account for other aspects of the relationship between the putative employer and the worker. Among those, we noted in Garcia-Celestino I , the Supreme Court has identified for possible consideration the following: (1) "the skill required [for the work]"; (2) "the source of the instrumentalities and tools"; (3) "the location of the work"; (4) "the duration of the relationship between the parties"; (5) "whether the hiring party has the right to assign additional projects to the hired party"; (6) "the extent of the hired party's discretion over when and how long to work"; (7) "the method of payment"; (8) "the hired party's role in hiring and paying assistants"; (9) "whether the work is part of the regular business of the hiring party"; (10) "whether the hiring party is in business"; (11) "the provision of employee benefits"; and (12) "the tax treatment of the hired party." Garcia-Celestino I , 843 F.3d at 1293 (internal quotation marks omitted) (quoting Darden , 503 U.S. at 323-34, 112 S.Ct. 1344 ).

Nevertheless, we emphasized that "[t]hough these factors may be instructive, 'there is no shorthand formula or magic phrase that can be applied to find the answer' [to whether a party is an "employer"] under the common law approach." Id. (quoting NLRB v. United Ins. Co. of Am. , 390 U.S. 254, 258, 88 S.Ct. 988, 19 L.Ed.2d 1083 (1968) ). Rather, courts must assess what is relevant in a given case. And because Darden involved the question of whether the plaintiff there was an independent contractor or an employee (as did Reid , from which the Darden Court adopted its analytical framework), the Darden factors do not always apply easily to cases concerning other work relationships. Sometimes some-or even most-of the usual factors will not shed light on a particular set of facts. In those cases, courts have focused on other considerations more relevant to the specific facts before them.

For instance, in Clackamas , the Supreme Court addressed whether four physician shareholders who jointly owned a practice and comprised its board of directors also counted as "employees" of the practice under the common law. 538 U.S. at 442, 123 S.Ct. 1673. The Supreme Court observed that the entity at issue, a "professional corporation," had "no exact precedent in the common law" and found the Darden factors unhelpful to answering the question of whether the physicians were "employees." Id. at 445-47, 123 S.Ct. 1673. So the Court set about identifying relevant factors for the lower courts to use to analyze whether the professional corporation *1120was the physicians' "employer" under the common law.

The Court began by reaffirming that "the common-law element of control is the principal guidepost" for any analysis. Id. at 447-48, 123 S.Ct. 1673. But the factors it held to be relevant, which it drew from an Equal Employment Opportunity Commission compliance manual, focused on the very specific question of "whether a shareholder-director is an employee." Id. at 449, 123 S.Ct. 1673. Those factors included "[w]hether the organization can hire or fire the individual or set the rules and regulations of the individual's work"; "[w]hether and, if so, to what extent the organization supervises the individual's work," "[w]hether the individual reports to someone higher in the organization"; "[w]hether and, if so, to what extent the individual is able to influence the organization"; "[w]hether the parties intended that the individual be an employee, as expressed in written agreements or contracts"; and "[w]hether the individual shares in the profits, losses, and liabilities of the organization." Id. at 449-50, 123 S.Ct. 1673 (citation omitted).

As was true in Clackamas , the relationship between the parties here has "no exact precedent in the common law" of which we are aware. See Clackamas , 538 U.S. at 447, 123 S.Ct. 1673. Though Plaintiffs performed work on Consolidated Citrus's groves, Consolidated Citrus was not itself the "hiring party," and the question we must answer here is whether, in addition to Ruiz Harvesting, Consolidated Citrus was also Plaintiffs' employer. So unsurprisingly, not all of the typical common-law factors identified in Darden are relevant to our analysis. For that reason, we must identify and balance the factors that are actually relevant here, keeping in mind that "all of the incidents of the relationship must be assessed and weighed with no one factor being decisive." Darden , 503 U.S. at 324, 112 S.Ct. 1344 (internal quotation marks omitted) (quoting NLRB v. United Ins. Co. of America , 390 U.S. 254, 258, 88 S.Ct. 988, 19 L.Ed.2d 1083 (1968) ); see also Cilecek v. Inova Health Sys. Servs. , 115 F.3d 256, 260-61 (4th Cir. 1997) (setting out tailored factors relevant to whether a doctor is an employee of an entity providing medical services).

A. Relevant Factors for Determining Whether Consolidated Citrus Was a Common-law "Employer"

We begin by pinpointing which factors do, in fact, bear on our analysis. As in Darden and Clackamas , first and foremost, we consider control, which we have emphasized is "the proper focus" of our inquiry. See Garcia-Celestino I , 843 F.3d at 1292-93 ; Crew One , 811 F.3d at 1311. We also find three of the other common-law factors relevant as traditionally formulated: "the source of the instrumentalities and tools," "the location of the work," and "the provision of employee benefits." See Garcia-Celestino I , 843 F.3d at 1293.

Some of the remaining factors are also of value, once we customize them to address the circumstances of the relationship at issue here. As we have noted, in Darden , the Supreme Court detailed factors from the traditional common-law framework for identifying an "employer." There, the putative employer had a direct relationship with the plaintiff, and the question concerned whether the plaintiff was an employee or an independent contractor. See Darden , 503 U.S. at 321, 112 S.Ct. 1344. So the traditional common-law framework made sense to apply, since the guiding factors refer to "the hiring party" and serve to distinguish employees from independent contractors.

Here, however, Ruiz Harvesting-not Consolidated Citrus-was the hiring party.

*1121As a result, we must tweak some of the remaining Darden factors to account for this difference between the circumstances in Darden and those here. See Clackamas , 538 U.S. at 449-50, 123 S.Ct. 1673 ; Cilecek , 115 F.3d at 260-61.

We therefore consider the following additional factors: whether Consolidated Citrus had the right to directly assign Plaintiffs additional work, cf. Darden , 503 U.S. at 323, 112 S.Ct. 1344 ("whether the hiring party has the right to assign additional projects to the hired party"); the extent to which Consolidated Citrus had discretion over when and how long Plaintiffs could work, cf. id. ("the extent of the hired party's discretion over when and how long to work"); and whether the work Plaintiffs did is part of Consolidated Citrus's regular business, cf. id. at 324, 112 S.Ct. 1344 ("whether the work is part of the regular business of the hiring party"). See also Garcia-Celestino I , 843 F.3d at 1293.

Together, these comprise all of the factors relevant to this case.5 We now explain why, viewed on the whole, they show that Consolidated Citrus was not Plaintiffs' employer under the common law.

B. Factors Favoring the Determination that Consolidated Citrus Was Not a Common-law "Employer"

We first discuss those factors that weigh in favor of the conclusion that Consolidated Citrus was not an "employer" under common-law principles: (1) whether and to what extent Consolidated Citrus had "the right to control the manner and means by which the product is accomplished," Garcia-Celestino I , 843 F.3d at 1292-93 (quoting Darden , 503 U.S. at 323, 112 S.Ct. 1344 ) (internal quotation marks omitted); (2) "the source of the instrumentalities and tools," id. at 1293 (quoting Darden , 503 U.S. at 323, 112 S.Ct. 1344 ); (3) "the provision of employee benefits," id. (quoting Darden , 503 U.S. at 324, 112 S.Ct. 1344 ); and (4) whether Consolidated Citrus had discretion over when and how long Plaintiffs could work, cf. id. ("the extent of the hired party's discretion over when and how long to work") (quoting Darden , 503 U.S. at 323, 112 S.Ct. 1344 ). Because, as we have noted, control is the most important of these factors, we start with it.

1. Consolidated Citrus did not have control over the manner and means of Plaintiffs' work under the general common law.

At a general level, the common-law control test "takes into account the degree of supervision, the entrepreneurial interests of the agent and any other relevant factors." Associated Diamond Cabs , 702 F.2d at 919-20. We emphasize that "it is the right to control, not the actual exercise of control, that is significant." Id. (emphasis in the original) (citation omitted).

We examined this factor in Crew One , where we held that a freelance-worker-referral agency was not the common-law employer of the workers it referred. The entity in question, called Crew One, ran a business contracting out stagehands to staff large events fully operated by third-party producers. Crew One, 811 F.3d at 1308. Each stagehand signed an "Independent Contractor Agreement" with Crew One and, after being entered into the company's database, received work offers on a first-come, first-served basis with full freedom to accept or decline without consequence. Id. at 1309. Crew One required its stagehands to abide by numerous policies, *1122such as wearing proper attire and limiting interactions with event attendees, but it provided no equipment to the stagehands other than a reflective vest marked "Crew One" to be worn while staffing events. Id. at 1308. Crew One also required its stagehands to check in with a Crew One project coordinator, who confirmed attendance and assigned workers to a particular department, such as rigging or carpentry. Id. at 1309.

But after the stagehands went to their designated departments, they reported exclusively to the third-party producers, except that they had to sign out with Crew One to record their times of departure. Id. And though Crew One did not withhold taxes or offer its stagehands any benefits, it did directly pay wages to its stagehands after receiving payment from the third-party producers. Id. at 1309.

Based on these facts, we concluded that Crew One did not have the common-law right to control the stagehands. Id. at 1311. We noted that the third-party event producers solely exercised all control over the stagehands' minute-to-minute work-that is, "the means of the work" the stagehands performed. Id. Nor did it matter to our analysis that the stagehands checked in and out with Crew One. Id. As we explained, that requirement "evince[d] control over the ends of the job, not the means of it." Id. And under the common law of agency, we must focus on an entity's "control over the manner and means of the agent's performance and the details of the work" and ignore "mere economic control or control over the end result of the performance." See id. (internal quotation marks omitted) (citing Associated Diamond Cabs , 702 F.2d at 919 ).

The common-law definition of "control" explored in Crew One stands in contrast to the markedly different "control" analysis relevant to defining the term "employer" under the FLSA. As we have observed, the FLSA defines "employ" as "suffer or permit to work." Garcia-Celestino I , 843 F.3d at 1287 ; 29 U.S.C. § 203(g). "Control" under that rubric does not focus on the actual work itself. See Crew One , 811 F.3d at 1311. Rather, under the FLSA, we ask "whether, as a matter of 'economic reality,' the hired individual is 'economically dependent' upon the hiring entity." Garcia-Celestino I , 843 F.3d at 1294. See Aimable v. Long & Scott Farms , 20 F.3d 434, 440-41 (11th Cir. 1994) (proper focus under FLSA definition of "control" is on "direct employment decisions such as whom and how many employees to hire, whom to assign to specific tasks, and how to design the employees' management structure").

The common-law and FLSA definitions of "employer" diverge from one another in other important ways as well, most notably in their breadth. The FLSA's "suffer or permit to work" standard "was developed to assign responsibility to businesses that did not directly supervise putative employees." Antenor v. D & S Farms , 88 F.3d 925, 933 (11th Cir. 1996). We have called this definition "one of the broadest possible delineations of the employer-employee relationship." Garcia-Celestino I , 843 F.3d at 1287.

The common-law test, on the other hand, may be reduced to identifying who has the right to control workers' "physical conduct in the performance of" their work. See Restatement (Second) of Agency § 220(1). That results in a much narrower analytical approach. Under the common law, we must look at only who controls "the manner and means" and "the details of the work," giving no consideration to "mere economic control or control over the end result of the performance." See Crew One , 811 F.3d at 1311.

*1123When we apply that test here, we must conclude that Consolidated Citrus did not exhibit significant control over Plaintiffs.

Focusing on the right to control the manner and means of Plaintiffs' work, see Crew One , 811 F.3d at 1311, it is clear that Ruiz Harvesting-and not Consolidated Citrus-enjoyed that right. The district court's findings show that, for instance, Ruiz Harvesting decided what implements Plaintiffs would use: ladders, sacks for gathering fruit, and tubs in which to deposit it. And Consolidated Citrus played no role in repairing or replacing Plaintiffs' equipment when it became damaged. By contrast, though, Consolidated Citrus often troubleshot for those workers it had hired directly.

In addition, Ruiz Harvesting crew leaders alone communicated with Plaintiffs as they worked, providing direction, clarification, or correction to them as needed. Consolidated Citrus, on the other hand, specifically "[did] not interfere with the crew assignments for [Ruiz Harvesting] harvesters," such as, for instance, when conflicts between workers would arise.

Similarly, while Consolidated Citrus sometimes reminded its own workers to wear their safety goggles, the company's field supervisors did not give any reminders to Plaintiffs.6 In short, Consolidated Citrus neither had nor exercised the right to direct the specifics of Plaintiffs' work.

The only semblance of control Consolidated Citrus retained was the right to halt work for any reason, a right the district court found Consolidated Citrus did periodically exercise. While this demonstrates at least some control over the workers, it merits only minimal weight. Most landowners who contract for on-site laborers retain the capability to halt work on their own property, at least under certain conditions. Consolidated Citrus's right to do so, though broad in scope, does not differ from this in kind. To give it too much weight would risk converting all landowners into joint employers over anyone working on their land.

That Consolidated Citrus provided neither tools nor instructions, and otherwise had virtually no right to control the details of Plaintiffs' physical work, drives the analysis here. See Crew One , 811 F.3d at 1311 ; Restatement (Second) of Agency § 220(1). Since control over the manner and means of Plaintiffs' performance-that is, control over the details of Plaintiffs' physical work-forms the crux of the common-law definition, we must conclude that Consolidated Citrus exhibited barely any control over Plaintiffs. So this factor weighs in favor of finding that Consolidated Citrus was not Plaintiffs' joint employer.

The district court reached the opposite conclusion concerning control. We explain why we must disagree. First, the district court emphasized what it described as the "high degree of supervision" Consolidated Citrus exercised over Plaintiffs. In doing so, though, the court relied on several practices not relevant to common-law control. In particular, the court pointed to the company's role in clocking workers in and out, the company's general requirement that workers start work "at some time in the morning and ... fill a particular number of trailers with citrus each day," the company's determination of which parts of the grove the workers *1124would harvest, and the company's supervisors' on-site presence during portions of the workday.

But practices such as assigning worksites and establishing production goals serve to regulate only the ends rather than the manner and means of work, having no effect on workers' moment-to-moment tasks. See Crew One , 811 F.3d at 1311. These kinds of considerations may well bear on the FLSA's definition of "control," but they do not factor into the operative definition here. Similarly, clocking Plaintiffs in and out does not bear on Plaintiffs' actual work, either. Instead, that practice merely "ensure[s] that the [workers] are present ...." See id. As a result, these practices do not support a showing of common-law control, regardless of whether we consider them individually or together.

Monitoring the workers in the field may, on the other hand, appear more suggestive of control in the abstract. But the district court's specific factual findings here tell a different story. Consolidated Citrus's harvesting supervisors checked on each crew of workers for a few ten-to-fifteen-minute increments throughout each workday, during which they "confirmed that the crew was picking in the right place," "tested the fruit," "scanned the block to see if any fruit had been mistakenly left on the trees," "and checked for garbage, debris ..., and obvious safety hazards."7 Company officials did not interact with the workers directly; rather, they notified Ruiz Harvesting supervisors when any issues required attention. And Ruiz Harvesting then ascertained how best to fix the situation. In contrast, Consolidated Citrus sometimes spoke personally with those workers it had hired, directing their individual tasks.

Here, Consolidated Citrus's periodic field presence primarily affected the ends-not the means-of Plaintiffs' work. True, close monitoring may support a finding of common-law control in some cases, especially where the evidence shows that the monitoring translated into concrete changes to the workers' behavior or to the direct expectations placed upon the workers. But again, our key inquiry must focus on whether Consolidated Citrus exhibited control over "the manner and means of the agent's performance and the details of the work." See Crew One, 811 F.3d at 1311. And the district court's findings do not show a connection of that kind here.

For example, the district court found that whenever a Consolidated Citrus supervisor noticed unpicked fruit or garbage left on the ground, the supervisor "sp[oke] to [Basiliso] Ruiz and/or the crew leader and ask[ed] [Ruiz Harvesting] to rectify the situation." But the district court's findings end there, never revealing what results, if any, this brought about. So we do not know whether the Ruiz Harvesting crew leaders (or Ruiz himself) instructed the workers to correct each problem or whether instead, the crew leaders simply *1125addressed the problem on their own.8 And we are left with no basis to conclude Consolidated Citrus's supervision in fact translated into a right to control Plaintiffs' performance of their work.

Plaintiffs bore the burden of proving that Consolidated Citrus was their "employer" for purposes of establishing their breach-of-contract claims. See Malvino v. Delluniversita , 840 F.3d 223, 231 (5th Cir. 2016) ("[W]hen a case does go to trial, the burden is on the plaintiff to prove every element."); Rivera-Flores v. Puerto Rico Tel. Co. , 64 F.3d 742, 747 (1st Cir. 1995) (observing the plaintiff possesses the "burden of introducing evidence at trial on every element essential to her claim"). So Plaintiffs' failure to show that Consolidated Citrus's in-grove presence actually affected their moment-to-moment tasks prevents us from upholding the district court's conclusion that the company's field supervision demonstrated common-law control.

Besides these circumstances, the district court also pointed to Consolidated Citrus's mandatory decontamination procedures, which the company implemented in an effort to prevent the spread of citrus canker disease, as a basis for concluding Consolidated Citrus had a right to control Plaintiffs. We again disagree that this shows evidence of control.

To explain why, we take a moment to describe what citrus canker is and how it can affect a citrus grove. At trial, one of Consolidated Citrus's owners testified that citrus canker is an airborne bacterial disease causing citrus trees both to lose their leaves and to drop their fruit prematurely.9 Trial Tr. (Feb. 12, 2014) at 18-19. Because the disease spreads so easily, government officials have periodically inspected orange groves and ordered any infected trees-and all those nearby-to be removed. Id. at 19; Trial Tr. (Feb. 19, 2014) at 49. At least one trial witness said that for each infected tree, this could mean clearing between 250 and 300 surrounding acres of trees. Trial Tr. (Feb. 12) at 19-20. Consolidated Citrus lost a significant number of trees as a result, possibly as much as 3,000 acres' worth. Trial Tr. (Feb. 12) at 20.

Against this backdrop, Consolidated Citrus required Plaintiffs to walk through an antibacterial mist and dip their picking sacks in decontaminant solution.10 But under the circumstances, this process does not reflect the control necessary to help evidence an employer-employee relationship under the common law.

*1126Rather, anti-canker decontamination procedures are merely a species of what we have labeled "agricultural decisions" in the FLSA context: necessary parts of agricultural administration such as choosing which fields to pick on which days or dictating what planting specifications should be used. See Aimable , 20 F.3d at 441 ; Martinez-Mendoza v. Champion Int'l Corp. , 340 F.3d 1200, 1210-11 (11th Cir. 2003). We have said such decisions do not show "control" under the FLSA definition even though they might indirectly affect how many workers need to be hired, a factor usually relevant to the inquiry under that statute. See id. Rather, they simply reflect decisions that the science of successful agriculture mandates. Requiring anti-canker decontamination procedures in Florida falls into this category because without such procedures, the grower risks significant depletion of its groves.

Just as purely agricultural decisions do not demonstrate "control" under the FLSA definition, they do not show "control" as defined under the common law, either. Such decisions do not affect the manner and means of the work because they involve no real intervention over how the workers go about the details of performing their moment-to-moment labor-in this case, picking the fruit. See Crew One , 811 F.3d at 1311. These types of decisions also effectively represent necessary preconditions to the existence of agricultural businesses. For these reasons, we cannot conclude that the decontamination procedures show common-law "control." See also Restatement (Second) of Agency § 220 cmt. l ("If ... rules are made only for the general policing of the premises ..., mere conformity to such regulations does not indicate that the workmen are servants of the person making the rules.").

We find Plaintiffs' arguments that Consolidated Citrus had the right to control their work similarly unavailing. Plaintiffs raise two such arguments on appeal. They first note that under the common law, the proper inquiry is not whether Consolidated Citrus in fact exercised control over them but whether it simply retained the right to do so. We agree that this correctly states the law. See Associated Diamond Cabs , 702 F.2d at 920 ("[T]he courts have noted that it is the right to control, not the actual exercise of control, that is significant.") (emphasis in the original). But the distinction Plaintiffs draw has no effect on our analysis for a simple reason: they do not identify any such control rights Consolidated Citrus purportedly had.

And in fact, the record strongly indicates that Consolidated Citrus retained no additional rights, at least on paper, beyond those it exercised. The company's written agreements with Ruiz Harvesting for each growing season specify that Consolidated Citrus "will not direct employees of [Ruiz Harvesting] in any fashion, but will communicate with [Ruiz Harvesting] regarding timing and quality control of harvest operations on [Consolidated Citrus's] groves."11 They say that all workers Ruiz Harvesting hires "shall be subject to the exclusive control and direction of [Ruiz Harvesting]," and Ruiz Harvesting will manage its workers "without interference from [Consolidated Citrus]." And most significantly of all, each agreement states that Consolidated Citrus "shall not exert actual control over, nor possess the right to control , the actions of any employees of [Ruiz Harvesting] in performing duties under this Agreement" (emphasis added). The most natural reading of these provisions indicates that the *1127parties intended for the right of control over the workers to belong solely to Ruiz Harvesting. And Plaintiffs do not present evidence that this agreement was some type of sham.12 So Plaintiffs' emphasis on the right to control does not does not aid their cause.

Plaintiffs next argue that control can be "entirely indirect" and "exercised through a contractor as an intermediary." They cite to Hodgson v. Griffin and Brand of McAllen, Inc. , 471 F.2d 235, 238 (5th Cir. 1973), in which the former Fifth Circuit concluded that work instructions passed on through an intermediary supervisor nonetheless evinced control over the workers.13 See id. ("The fact that appellant effected the supervision by speaking to the crew leaders, who in turn spoke to the harvest workers, rather than speaking directly to the harvest workers does not negate a degree of apparent on-the-job control over the harvest workers.").

Though Hodgson arose under the FLSA, its discussion of "on-the-job control" indeed pertains equally to the common-law standard of control over the "manner and means" or "physical conduct in the performance" of work. See Crew One , 811 F.3d at 1311 ; Restatement (Second) of Agency § 220(1). But Plaintiffs again fail to identify any actions Consolidated Citrus carried out indirectly that would make this standard applicable here. As we have noted, the district court's factual findings do not show that even Consolidated Citrus's onsite supervision and direct interactions with Ruiz Harvesting crew leaders translated into control over the workers. Nor does the record offer any indication of any other indirect actions Consolidated Citrus might have taken to control Plaintiffs' work through Ruiz Harvesting.

Overall, then, Consolidated Citrus exhibited little to no control over Plaintiffs in ways relevant to the common-law "control" analysis. So this factor strongly indicates that Consolidated Citrus was not a joint employer under the common law.

2. Other Factors Indicate that Consolidated Citrus Was Not an Employer Under the Common Law

Other factors we have mentioned also support the conclusion that Consolidated Citrus was not a joint employer under the common law. First, Ruiz Harvesting was the sole source of the workers' instrumentalities and tools. Though we noted this fact in our discussion on control, the source of the tools represents an independent factor we separately consider under general common-law principles of agency. See *1128Garcia-Celestino I , 843 F.3d at 1293. And here, Consolidated Citrus provided Plaintiffs with none of the materials used to perform their jobs. Nor, based on its agreements with Ruiz Harvesting, would it have had the right to do so, either.

Second, as the district court pointed out, the provision of employee benefits was solely the province Ruiz Harvesting. Plaintiffs' only benefit noted by the district court was workers' compensation insurance, which Ruiz Harvesting provided in full. Though Consolidated Citrus required Ruiz Harvesting to maintain coverage for its workers, Consolidated Citrus played no role in choosing a provider or paying the insurance premiums, both of which were left solely to Ruiz Harvesting.

Third, Ruiz Harvesting retained the great bulk of the discretion over when and how long Plaintiffs could work. Consolidated Citrus generally expected Plaintiffs to begin work "at some time in the morning," but it was Ruiz Harvesting that chose their precise start time. And while Consolidated Citrus designated how much total fruit was to be picked every day, Plaintiffs' end time was up to Ruiz Harvesting. On some occasions, Plaintiffs would continue picking even after meeting their required quotas. Likewise, during the workday, Ruiz Harvesting determined when Plaintiffs could take breaks and for how long.

These factors, taken together with control, all weigh strongly in favor of finding that Consolidated Citrus was not Plaintiffs' joint employer.

C. Factors Indicating that Consolidated Citrus Was Not a Joint Employer Under the Common Law

The remaining relevant factors militate in the other direction, but they do not outweigh the factors we have so far described. True, Plaintiffs performed their work at Consolidated Citrus's own groves, and picking citrus fruit is at the heart of Consolidated Citrus's business. It is also true that while Ruiz Harvesting directed Plaintiffs' work throughout each day, Consolidated Citrus could assign them additional work in the future by increasing their daily production targets. These three factors do provide some weight suggesting that Consolidated Citrus was a joint employer under the common law.

But on balance, they cannot outweigh control and the other factors favoring a finding that Consolidated Citrus was not Plaintiffs' joint employer under the common-law standard. As we have explained, the company lacked the right to control the manner and means of Plaintiffs' work, the weightiest of all considerations germane to our analysis. And Ruiz Harvesting was the sole source of Plaintiffs' tools, benefits, and work schedules.

D. Factors Irrelevant to Determining Whether Consolidated Citrus Was a Joint Employer Under the Common Law

Finally, we think it worthwhile to explain why several other common-law factors noted in Darden do not bear on our analysis.

First, because some of the common-law factors were conceived for the purpose of differentiating between an employee and an independent contractor-a matter that is not at issue here-they cannot provide insight into this case, even if modified.

For instance, "the method of payment" inquiry focuses on whether the workers were paid "by the time or by the job"; an hourly-pay arrangement usually suggests workers are employees, while a by-the-job arrangement tends to indicate they are independent contractors. See Crew One , 811 F.3d at 1311 (quoting *1129Restatement (Second) of Agency § 220(2)(g) ). But Consolidated Citrus had no say in which method Ruiz Harvesting used to pay Plaintiffs.14 So regardless of which payment method Ruiz Harvesting chose (by-the-job, as it happens), this factor gives us no insight into whether Consolidated Citrus was Plaintiffs' employer.

The same is true for "[Plaintiffs'] role in hiring and paying assistants." See Garcia-Celestino I , 843 F.3d at 1293. Because we are not trying to discern whether Plaintiffs were independent contractors, the presence or absence of hired assistants cannot shed light on Consolidated Citrus's "employer" status.

"[T]he duration of the relationship between the parties" likewise is unhelpful for our purposes. This consideration gives insight into whether the workers had an ongoing relationship with the putative employer or were simply hired for a one-off job-another concern relevant to determining whether a worker is an employee or an independent contractor. See Marie v. Am. Red Cross , 771 F.3d 344, 358 (6th Cir. 2014) (explaining that "in evaluating this factor, the court 'is not concerned with the length of the relationship, but rather, when hired, whether the relationship was one of a long-term at-will employee or one to complete a particular task in a specified time-frame' "). It sheds no light on whether one of multiple entities counts as an "employer."

Second, some factors do not bear on our analysis here because the record before us does not provide the necessary insight to tell which way they cut. For instance, neither party identifies relevant evidence about Plaintiffs' tax treatment. Perhaps certain facts of this nature-such as whether Consolidated Citrus issued Plaintiffs relevant tax forms, or how state and federal authorities regarded the relationship between the company and Plaintiffs for tax purposes-may have borne on our analysis. Cf. Rev. Rul. 87-41, 1987-1 C.B. 296 (setting forth Internal Revenue Service guidance on who qualifies as an "employee" under federal taxation legal framework). But neither party has identified any record evidence to this effect, and we have not found anything especially relevant in the record, either.15

The same is also true for the skill required. Even assuming picking fruit constitutes unskilled labor, the Restatement gives conflicting advice on the significance of that circumstance. On the one hand, it states that "[u]nskilled labor is usually performed by those customarily regarded as servants." But on the other, where an entity "furnishes unskilled workmen to do work for another, it is not abnormal to find that the workmen remain the servants of the one supplying them." Restatement (Second) of Agency § 220 cmt. i. So standing *1130alone, the fact that Plaintiffs engage in unskilled labor tells us little. Instead, the Restatement urges that "[t]he custom of the community as to the control ordinarily exercised ..., together with the skill which is required in the occupation, is often of almost conclusive weight." Id. cmt. l.

But the parties have not identified anything in the record showing how much control a grower customarily exercises over laborers picking its crops. The district court below cited to precedent of ours noting that "the grower is not expected to look over the shoulder of each farmworker every hour of every day," and that a grower's supervision may render it an employer "whether orders are communicated directly to the laborer or indirectly through the contractor." Dist. Ct. Slip Op. (May 11, 2017) at 8 (citing Antenor , 88 F.3d at 935 ). But we have already explained why, on this record, those considerations do not assist us. The record before us includes no evidence about the customary relationship between growers and unskilled workers in Plaintiffs' situation.

So when we consider only those common-law considerations relevant here, we are left with the conclusion that on balance, Consolidated Citrus was not Plaintiffs' joint employer.

Nevertheless, we emphasize once more that the common-law determination of who is an employer does not reduce to any "shorthand formula or magic phrase." See Garcia-Celestino , 843 F.3d at 1293. Under other facts, our conclusion might be different.

IV. CONCLUSION

For the reasons we have explained above, we conclude Consolidated Citrus was not Plaintiffs' joint employer under the common law. The district court's judgment is vacated, and the case is remanded for entry of judgment for Consolidated Citrus on the breach-of-contract claim.

VACATED AND REMANDED.