CITATION: Heller v. Uber Technologies Inc., 2026 ONSC 4102

COURT FILE NO.: CV-17-00567946-00CP

DATE: 20260714

SUPERIOR COURT OF JUSTICE - ONTARIO

BETWEEN: DAVID HELLER and FELICIA GARCIA, Plaintiffs

AND:

UBER TECHNOLOGIES INC., UBER CANADA INC., UBER B.V., RASIER OPERATIONS B.V. and UBER PORTIER B.V. (TORONTO), Defendants

BEFORE: Justice Glustein

COUNSEL: Michael D. Wright, Nora Parker, Danielle Stampley, and Youssef Kodsy, for the plaintiffs

Sarah Whitmore, Alexander Bogach, and Colette Koopman, for the defendants

HEARD: June 22, 2026

TABLE OF CONTENTS

NATURE OF THE MOTIONS AND OVERVIEW... 1

FACTS. 2

The parties. 2

The nature of the certified action. 2

Discovery of the representative plaintiffs. 6

Class Members sought to be examined. 7

ISSUES AND THE LAW... 8

The Uber refusals motion. 8

The plaintiffs’ refusals motion. 9

Questions regarding how and why Uber developed features. 9

(i) The nature of the questions. 9

(ii) The questions are not relevant to the common issues. 10

Questions seeking all pricing presentations. 13

Question on tipping function research. 14

Question seeking production of all hyperlinked documents in a document describing how the matching algorithm worked. 14

Question for production of batching plans. 15

Question about Uber Pool 15

Questions related to Uber’s use of behavioural scientists and other professions. 16

Question on production of documents related to surveys. 17

Questions concerning Uber participation with academics, think tanks or consultants to publish reports 17

Questions related to offer card testing. 17

Questions relating to key performance indicators. 18

Conclusion on the plaintiffs’ refusals motion. 19

Uber’s motion for leave to examine ten additional Class Members, pursuant to s. 15(2) 19

General principles. 19

The stage of the proceeding and the issues to be determined at that stage – s. 15(3)(a) 20

(i) The stage of the proceeding. 20

(ii) The issues to be determined at this stage. 21

(a) The evidence is relevant to the court’s understanding of the licensing agreements, Community Guidelines and app functionality. 22

(b) The evidence is relevant to how each Earner’s individual choices impact the classification test, such that a class-wide determination is not possible. 23

The presence of certified subclasses – s. 15(3)(b) 23

The necessity of discovery – s. 15(3)(c) 24

(i) The issues on which the proposed Class Members are to be examined are in dispute ……………………………………………………………………………………24

(ii) Uber does not have the information it is seeking. 26

(a) Uber’s records do not provide the relevant information for the common issues trial….. 26

(b) The certification affidavits and informal conversations do not constitute evidence at a common issues trial 27

The approximate monetary value of individual claims – s. 15(3)(d) 28

Whether discovery would result in oppression or in undue annoyance, burden or expense for the class members sought to be discovered – s. 15(3)(e) 28

Other factors – s. 15(3)(f) 29

Conclusion on granting leave. 29

Process for additional discovery – the plaintiffs’ alternative submission. 30

(i) The applicable law.. 30

(ii) Process issue 1: Should Uber select the additional Class Members?. 31

(iii) Process issue 2: Should the number of proposed additional Class Members be reduced? 31

(iv) Process issue 3: Should the examinations for discovery be conducted orally?. 32

(v) Process issue 4: The length of the examinations for discovery. 32

(vi) Other process issues. 33

ORDER AND COSTS. 33

REASONS FOR DECISION

NATURE OF THE MOTIONS AND OVERVIEW

1There are three motions before the court:

(i) The defendants, Uber Technologies Inc., Uber Canada Inc., Uber B.V., Rasier Operations B.V. and Uber Portier B.V. (Toronto) (collectively, “Uber”) bring a motion to compel the plaintiffs to answer questions refused during the examination for discovery of one of the representative plaintiffs, Felicia Garcia (“Garcia”).

(ii) The plaintiffs bring a motion to compel Uber to answer questions refused during the examinations for discovery of Uber’s representatives, J.M. Nickels (“Nickels”) and Niti Bashambu (“Bashambu”).

(iii) Uber brings a motion for leave to examine ten additional class members, pursuant to s. 15(2) of the Class Proceedings Act, 1992, S.O. 1992, c. 6 (the “CPA”).

2Uber’s answers to (i) questions 363-65, 550-52, 562-64, and 573 from Nickels’ examination for discovery and (ii) questions 253, 334, and 360 from Bashambu’s examination for discovery remain outstanding. The parties have agreed that Uber shall provide the answers to those questions by July 31, 2026. If Uber cannot provide the answers to any of those questions, Uber shall advise the plaintiffs of the steps Uber took to attempt to obtain the answers. This will allow the plaintiffs to determine whether they should take further steps to compel Uber to obtain the answers.

3For the reasons that follow, I dismiss the refusals motions brought by both Uber and the plaintiffs. The questions refused were either irrelevant (which was the case for the vast majority of the refusals) or were overbroad or lacked foundation.

4I grant Uber’s motion for leave to examine ten additional class members, pursuant to s. 15(2).¹ The evidence of those class members (i) is highly relevant to the common issues to be determined at the common issues trial; (ii) is sought at an early stage of the class proceeding; (iii) would not result in undue annoyance, burden or expense for the class members sought to be discovered; and (iv) is necessary given the distinctions in how class members interacted with Uber, particularly when that evidence is not available from the representative plaintiff or determinable solely on a document review.

FACTS

The parties

5Uber is comprised of a group of technology companies that develop software applications (collectively, the “Uber Apps”). This action relates to two of their marketplaces: (i) the Rides platform, which through an Uber App, allows a person seeking transportation (the “Rider”) to connect to a person offering transportation services (the “Driver”); and (ii) the Eats platform, which through an Uber App, connects a person ordering food or other items (the “Eater”) with the restaurant or person supplying the goods (the “Merchant”) and the person providing the delivery services (the “Delivery Person”). I use the term “Earner” (a term used by Uber) to describe a person who is a Driver or Delivery Person.

6Uber controls the design of its Apps and determines which features are deployed in which regions, although they seek input from App users. At the time of certification, Uber operated in 69 countries around the world. Uber’s operations in different regions are led by local operations teams who respond to local needs.

7Uber earns more when its customers use its products. Uber strives to encourage increased engagement by all users—not just by Earners, but also by Eaters, Riders and Merchants.

8There are two representative plaintiffs. David Heller used the Uber App to provide delivery services using his car in the greater Toronto area from February 24, 2016 until April 12, 2018. He also used DoorDash—one of Uber’s competitor applications—but did not use the two apps to complete offers simultaneously.

9Garcia began using the Uber App in October 2015 to provide ridesharing services and in 2017 to provide delivery services in her car. She continues to use the App today. Garcia did not try other platforms during the class period.

The nature of the certified action

10In his decision certifying the present class action, Heller v. Uber Technologies Inc., 2021 ONSC 5518, Justice Perell reviewed the relationship between Uber and its Earners. He held that the relationship is governed by several sources (Heller, at paras. 79-111, 115-17):

(i) Earners sign a standard form Service Agreement, which an Earner must accept to provide services (including agreeing to any amendments).

(ii) The Uber Apps provide Drivers with access to available Riders and Delivery People with access to Merchants for food deliveries to Eaters.

For the Rider applications, the Rider downloads one version of the Uber Ridesharing App from an Internet site. The Driver, who provides the transportation, downloads a complementing and corresponding version of the software.

For the “Uber Eats” application, there are three complementing and corresponding versions of the software. The Eater, who orders the food or other items, downloads one version. The Merchant, which is the restaurant or person supplying the goods, downloads an associated version. The Delivery Person, who provides the carriage of the food/items (either by driving, walking or cycling), downloads a complementing version of the Uber Eats App.

(iii) Uber offers a variety of incentives to increase the use of the Uber Apps including:

(a) Surge pricing, which is an incentive through which Uber encourages Drivers and Delivery People to work at certain high demand locations, or during high demand times, in exchange for additional compensation.

(b) “Quests”, which are bonus pay incentives for Drivers and Delivery People who complete a high number of trips in a set time period as determined by Uber.

(c) The “Uber Pro” and the “Uber Eats Pro” programs offer perks to Drivers and Delivery People who drive more frequently, drive during Uber’s preferred times, and maintain high ratings and low cancellation rates. Through Uber Pro, Drivers earn points for providing services at certain times or in certain places as determined by Uber. As they accumulate points, they can increase their “status” in the program and earn benefits, including: (1) “Airport Priority Rematch”, which gives Drivers a better chance for a quick pickup at an airport terminal after dropping off a Rider at the airport; (2) “Trip Duration and Direction”, which allows Drivers to view the estimated duration and direction of all Ride requests before accepting; (3) discounted car and bicycle maintenance; (4) discounted tax preparation; and (5) Moneygram discounts.

(d) Other examples of financial incentives that Uber offers or has offered to Drivers and Delivery People are: (1) providing referral codes that give Drivers a financial incentive to recruit other Drivers; (2) providing additional compensation for completing a certain number of rides in a given week; and (3) promotions for completing trips overnight in rural areas.

(iv) Under Uber’s “Community Guidelines”, which were formerly entitled as Uber’s “Code of Conduct”, Uber may discontinue access to the Uber App if the App user violates the Community Guidelines.²

Under the Community Guidelines, Class Members were required to keep all documentation with Uber up to date; maintain the minimum average rating set by Uber; not contact Riders or Eaters after a trip or delivery, except to return a lost item; not share their account with anyone; not accept street hails; not solicit payment of fares outside the Uber App; and not discriminate against someone based on their destination.

(v) Earners are also governed by some external rules and regulations imposed by municipalities on the users of the Uber Apps.

11The class period spans from January 1, 2012 to August 12, 2021 (the date of certification) and includes over 360,000 individual Earners³ who used Uber Apps in Ontario to provide rideshare or delivery services (the “Class Members”).

12The core of the claim is that Uber improperly characterized the Class Members as independent contractors when they should have been classified as employees. The class alleges that as such, Uber breached the Employment Standards Act, 2000, S.O. 2000, c. 41 (the “ESA”) and its contracts with the Class Members, failing to provide, inter alia, the minimum requirements for wages, overtime pay, vacation pay, public holiday and premium pay, EI and CPP contributions, and their out-of-pocket expenses.

13The plaintiffs seek $200 million in damages and an additional $200 million for punitive, aggravated and exemplary damages.

14Justice Perell found that there was a basis in fact for a common determination to be made about the relationship between Uber with the Class Members. However, he did not find that the common issues judge would necessarily be able to make such a common determination. Justice Perell applied the “some basis in fact” test, which provides for a “low bar”⁴ on certification: at para. 185.

15Justice Perell recognized that upon full evidence on the common issues at trial, a court might be able to find a common answer to the issue of whether the Class Members were employees or independent contractors, but noted that the common issues judge might find that a common resolution of the employment status of Class Members (i.e. as an employee or independent contractor) might not be possible and instead would require a case-by-case assessment. He held, at para. 192:

In this regard, based on the voluminous evidentiary record that I have reviewed and considered there is some basis in fact for any of the following answers to the common issue questions, including several answers that would be favourable to the putative Class Members and some that would be favourable to Uber; visualize there is some basis in fact for concluding:

a. In some or all Uber Service Agreements, there was no employment or independent contractor relationships between the Uber App users and Uber and the relationship between Uber users and Uber was of a customer and a service provider.

b. In some or all of the Uber Service Agreements, the relationship between Uber App users and Uber is that of independent contractor and employer.

c. In some or all of the Uber Service Agreements, the relationship between Uber App users and Uber was or is that of employee and employer.

d. In all Uber Service Agreements, it will take a case-by-case analysis to determine whether there was an employment or independent (or dependent) contractor relationship but either relationship is possible depending upon the circumstances of the particular case.

16Justice Perell’s conclusion was based, in part, by his review of affidavit evidence from six other Earners, which was filed by Uber to establish, among other things, distinctions between how different Earners used the Uber Apps and the reasons for the choices they made: at paras. 61, 64-68, 71.

17Garcia’s evidence at certification (which she adopted at her examination for discovery) also identified different categories of Earners: (i) employees, like she asserts that she is, who use the App with a “consistent pattern”; (ii) self-employed workers, who want “flexible hours” and the benefit of “writing off everything”; and (iii) “sole proprietor[s]” who want to “operate as their own business”, use multiple applications and “be a taxi service on their own.”

18A parallel Alberta class action was certified in 2023 by Justice Neufeld. Like Justice Perell, Justice Neufeld concluded that it is possible that each class member’s relationship with Uber “can only be determined through assessment of each Driver’s relationship with Uber on a case-by-case basis”: Virani v. Uber Portier Canada Inc., 2023 ABKB 240, at paras. 52-56.

Discovery of the representative plaintiffs

19Discovery of the representative plaintiffs occurred in July 2025 and answers to undertakings were delivered by March 2026. During the examinations, class counsel refused to permit the affidavits from Class Members filed during certification to be filed as exhibits.

20Although Garcia and Heller were able to respond to questions about their own reasons and methods for using the Uber Apps, they could not give evidence of (i) the experience or methods of other Class Members, and (ii) any choices those Class Members made or the reasons for such methods or choices. Neither representative plaintiff spoke to other Class Members to prepare for their examinations for discovery.

21Consequently, there is no evidence from the examinations for discovery of Heller or Garcia on other Class Members’ reasons for using the Uber Apps and choices made while using the Uber Apps, including:

(i) Reasons for using Uber’s technology: Both Heller and Garcia stated that they were economically dependent on Uber. But Garcia admitted that there are many different reasons why people use the Uber Apps, and (consistent with the evidence she gave at her cross-examination on the certification motion) she did not know them all.

(ii) Dual-Apping: Neither Garcia nor Heller had experience using multiple apps to provide ridesharing or delivery services simultaneously. The practice of using two (or more) apps at the same time is commonly referred to as “dual-apping”. The process can occur by a Class Member (a) using competing apps (such as Door Dash) at different times or (b) using competing apps at the same time using the Uber App.

(iii) Strategies: At certification, various Class Members gave evidence on different strategies they used to increase their profitability. Garcia was asked about some of these strategies and stated that she had either not used them or not heard of them. Heller’s strategies focused on what time of day he used the Uber Eats App.

(iv) Reasons for declining or cancelling trips: Both Garcia and Heller accepted that Class Members may have different reasons for declining or cancelling trips and they were unable to provide those reasons.

(v) Tools: Heller did not know what kinds of tools Class Members may use or why. For instance, he did not know if other Class Members purchased cars or bikes exclusively for rideshare or food delivery services or what kinds of other equipment Class Members may have purchased.

(vi) Other topics: There were a variety of other topics which the representative plaintiffs did not have experience with, including providing services through a corporation, participating in Uber hosted focus groups, negotiating tips, and resolving complaints with Riders and Eaters directly instead of involving Uber.

Class Members sought to be examined

22Uber provided detailed evidence on the information available about the Class Members they seek to examine. While there is some overlap between these Class Members’ dealings with Uber, each also has unique evidence about their reasons, methods, and choices when using the Uber Apps. I adopt Uber’s summary of the evidence about each of these Class Members as follows:

(i) Earla Phillips: Based on publicly available information, Ms. Phillips is the vice president of the Rideshare Drivers Association of Ontario. She has provided rides using various platforms including Uber, Lyft and FaceDrive. There is evidence that she completed the Taxi 100 program at Centennial College as a form of driver training. According to Uber’s records, Phillips completed over 13,000 offers using the Uber App during the class period.

(ii) Jennifer Lynn Scott: Based on publicly available information, Ms. Scott was a bike, e-bike and walking delivery person and the President of Gig Workers United CUPW. In interviews, Ms. Scott stated that she had work clothes and safety equipment which she used when making bike deliveries. Ms. Scott also stated that she logged into two or three apps at the same time and used Foodora, SkipTheDishes and DoorDash’s platforms. Uber’s records indicate that she rejected over 50% of the offers made to her through the Uber App during the class period.

(iii) Raymond Paquette: According to publicly available information, Mr. Paquette is a retiree who used the Uber App in Kingston to take advantage of the flexibility. Mr. Paquette provided an interview where he said he was hopeful that Uber’s competitor, Lyft, would also enter the Kingston market.

(iv) Olivia Henry: Ms. Henry is a YouTuber who posted videos about her use of the Uber App. In her videos, Ms. Henry described different strategies she employs and how she tracks her expenses to file taxes.

(v) Ejaz Rasul Butt: According to publicly available information, Mr. Butt is a retired veteran Hamilton cab driver who started using the Uber App. He stated that he uses the App “here and there.” Based on Uber’s records, Mr. Butt provided rideshare services in Hamilton, Toronto, Kitchener-Waterloo and Niagara.

(vi) Craig West: Mr. West was invited to an Uber roundtable because he used multiple platforms. Mr. West advised Uber that, in addition to using other platforms, he had a full-time job as a software engineer while using the Uber App.

(vii) Ryan Ruokonen: There is evidence that Mr. Ruokonen’s primary source of income was derived through using various platforms, including Uber, SkipTheDishes, DoorDash, Amazon Flex and Foodora. According to Uber’s records, Mr. Ruokonen completed only 14.6% of the offers made to him during the class period.

(viii) Hector Goldby: Mr. Goldby also participated in an Uber roundtable for Drivers and Delivery People who used multiple platforms. According to Uber’s records, Mr. Goldby rejected over 50% of the offers made to him during the class period.

(ix) Donald McMullen: Mr. McMullen provided wheelchair accessible transportation services using the Uber App and used a product called UberHOP, which enabled Drivers with large vehicles to transport multiple Riders on busy routes that did not have public transit options.

(x) Ashfaq-Ashraf Kiyani: Mr. Kiyani is the director and a shareholder of 11975188 Canada Inc. According to Uber’s records, Mr. Kiyani used the registered business name 11975188 Canada Inc. for tax purposes.

ISSUES AND THE LAW

23I first address the refusals motions brought by Uber and the plaintiffs, respectively. I then address the motion brought by Uber for leave to examine ten additional class members, pursuant to s. 15(2).

The Uber refusals motion

24Uber only seeks answers to two refused questions on Garcia’s examination for discovery (QQ. 924 and 931).

25At Q. 924, Uber asked Garcia “whether there are any provisions of the Community Guidelines that [she] disagrees with being bound to, and if so, to advise why?”

26At Q. 931, Uber asked Garcia “to advise whether there are any Uber platform rules or policies that [she] has come across that she would remove if she could. If so, to advise which ones and why.”

27Both of the refusals are proper. Neither of these questions are relevant to the issues before the court at the common issues trial.

28At the common issues trial, the court will be required to determine whether it can find a common employee or independent contractor relationship arising from the structure under which the Class Members worked, as summarized above at para. 10. That issue will require the court to understand how (and the extent to which) Class Members made choices regarding when, where, and how to work as Earners.

29Consequently, the common issues court will have to consider the experience of a Class Member using the Uber Apps and being subject to aspects such as the standard form agreement, Community Guidelines, incentives, rules, and policies.

30However, the Class Members do not allege that they were not bound by any of the Service Agreements or by the Community Guidelines, nor that the provisions ought to have been different.

31Uber does not allege that the Class Members could change the Service Agreements or the Community Guidelines. Uber acknowledged that (i) it has “ultimate discretion over enforcing the Community Guidelines”; (ii) “[a]n Earner can decide not to be bound by the Community Guidelines by terminating the licensing agreement they have with an Uber entity”; and (iii) Class Members were required to abide by the Community Guidelines or face removal from the Uber App.

32The Class Members allege that because they were bound to this structure, a common determination can be made that they were employees. Uber alleges that within this structure, Class Members enjoyed flexibility as to how, when, and where to work, such that a common issues judge could either (i) make a common determination that the Class Members were independent contractors, or (ii) find that a case-by-case analysis is required to determine the status of each Class Member.

33Consequently, it is not relevant whether a Class Member did not agree that they should be bound to a particular provision of the Community Guidelines, or whether a Class Member would change or remove any rules or policies under the Uber platform. The issue before the common issues court is not whether the structure was defective. The issue is whether the structure, as it existed, created a common employment or independent contractor relationship, or whether a case-by-case analysis is required given the choices made by Class Members within that structure.

34Consequently, I find that the refusals of the plaintiffs to QQ. 924 and 931 of Garcia’s examination for discovery were proper.

The plaintiffs’ refusals motion

35The plaintiffs brought a refusals motion to address approximately 30 refusals taken by Uber during the examinations for discovery of Nickels and Bashambu. I categorize and address these refusals below.

Questions regarding how and why Uber developed features

(i) The nature of the questions

36The refused questions are questions 68/69, 479, 717, 802, 807, 808, 813, 822, 998, 1120, 1121 from the Nickels examination for discovery and questions 203, 248/251, 285, 292 and 301 from the Bashambu examination for discovery.

37The plaintiffs asked a series of questions as to how Uber designed, marketed, developed, researched and tested the Uber Apps. In their factum for this motion, the plaintiffs submitted that such information was necessary since “the Plaintiffs are entitled to understand Uber’s business” under the “whose business is it” test.⁵

38The plaintiffs summarized the nature of these questions as follows:

(i) how Uber set about to increase the engagement of Class Members (para. 2)⁶;

(ii) how Uber “designed, marketed, developed, researched and tested the Uber App” (para. 2);

(iii) “how the services provided by the Class Members were factored into Uber’s business and its business decisions” (para. 20);

(iv) “why [Uber] decided to implement certain features over others on the Uber App to increase engagement” (para. 20);

(v) how “Uber developed its pricing structures, and how it chose which pricing structures to implement (and which it rejected and why)” (para. 29);

(vi) “how [Uber] developed, planned, and advertised [Uber Pool]” (para. 39); and

(vii) “how the offer card was designed and what kinds of individuals contributed to its design” (para. 51).

(ii) The questions are not relevant to the common issues

39I do not agree that these questions, which focus on “how” and “why” Uber included certain features on their Apps, are relevant to the common issues trial.

40The plaintiffs will ask the court at the common issues trial to find that, on a common basis, the Class Members’ work was directed and controlled by Uber, and as such, under the “whose business is it” test, the Class Members are employees of Uber.

41Uber will ask the court at the common issues trial to find that, on a common basis, the Class Members directed and controlled their own work and as such, the Class Members are independent contractors under the test.

42When applying the test, the common issues court will take a holistic approach in considering factors relating to control and direction. In the traditional workplace, control might be established by evidence of a company directing an employee on when, where, and how to do certain tasks (through direct contact with a supervisor or through corporate manuals). Evidence of the company providing the worker with necessary tools to perform the job could also be relevant.

43In the new technology-based gig economy, the plaintiffs submit that there is no boss, foreperson, or supervisor interacting with an employee. Instead, the plaintiffs’ theory is that through the Uber Apps, standard form service agreements, Community Guidelines, and Uber’s rules and practices, the “business” belongs to Uber with Class Members being directed as to when, where, and how to work.

44At the certification stage, Justice Perell held that based on the Uber structure, the plaintiffs met the “low bar” needed to establish a common issue that the Class Members were employees. Similarly, Uber led evidence to establish that individual Earners directed and controlled their own work. Consequently, Uber also met the “low bar” to establish a common issue that the Class Members were independent contractors, or that a case-by-case analysis was required to determine whether each individual Class Member was an employee or independent contractor.

45In order for a question to be relevant, it must seek to address whether a common finding can be made (through the application of the “whose business is it” test) on whether a Class Member is an employee, an independent contractor, or if a case-by-case analysis is needed.

46Information on how specific App features operated during the class period may assist the plaintiffs in arguing that Class Members were directed on how, when, and where to provide services, such that they were employees of Uber. Consequently, this evidence may assist the common issues court in determining (i) whether the Class Members were engaged in “activity … [as] part of the business organization”: Braiden, at para. 33; and (ii) whether that issue can be determined in common for all Class Members.

47However, questions as to how and why Uber developed features on the Uber Apps do not relate to these common issues.

48Uber has answered all questions and produced relevant documents addressing the structural elements of the Uber system which related to the work of the Class Members during the class period, including the operation of Uber Apps features, and any other structural components such as the Service Agreements, Community Guidelines, incentives, and policies.

49The refused questions, however, do not relate to the interaction between Uber and the Class Members during the class period. Instead, the questions ask Uber to disclose why they chose to include the features, including all internal documents from economists, behavioural scientists, psychologists, sociologists, political scientists, cognitive scientists or neuroscientists, whose insights contributed to the development of the features. Those questions are not relevant to the common issues.

50By way of example, Uber may have considered various methods to provide incentives to Class Members to encourage their use of the Uber Apps. However, the reasons for considering or rejecting various types of incentives have no connection to Uber’s employment relationship with Class Members, which depends on the effect of the actual incentives offered by Uber.

51Hypothetically, Uber may well have considered numerous types of incentives and obtained internal reports from economists (as to profitability) or behavioural scientists (as to likelihood of acceptance by Earners). However, under the “whose business is it” test, the only relevant issue is the effect of the actual incentives on Class Members, and whether that issue can be decided in common.

52Another example is the design of the “offer card”, through which Class Members would provide services as a Driver or Delivery Person. The plaintiffs can rely on the actual design of the offer card to suggest that Uber was using tools to attempt to direct how, when, and where services would be provided, and that such an issue can be decided in common. In contrast, Uber can rely on the offer card to support its position that the Class Members were making their own decisions as to how, when, and where to work. Thus, evidence on how offer cards worked during the class period might be relevant to determining whether Class Members were employees or independent contractors.

53However, evidence as to the process by which an offer card was designed or alternatives which were considered and rejected, is not relevant to the employment relationship, which depends solely on the actual interaction between Uber and the Class Members.

54The plaintiffs submit that they will be “in the dark” on cross-examination if an expert such as a behavioural scientist retained by Uber provides evidence that the features of the Uber Apps do not have the effect of directing how, when, or where a Class Member provides services. The plaintiffs submit that if Uber has studies that relate to how and why Uber selected a particular feature or rejected various options, then those studies are required to effectively challenge Uber’s expert evidence. I do not agree.

55Evidence as to how or why a feature was selected (or was rejected) is not relevant to whether a particular feature resulted in a Class Member being controlled or directed when completing their work. Any expert who provides evidence on the effect of a feature can only be challenged based on the actual feature, not on studies relating to which feature is most effective at increasing engagement.

56Internal documents relating to App features will only be relevant if they pertain to the employment relationship. For example, if a document states that a certain feature (which was actually in effect during the class period) is more likely to give Uber control over where, when, and how an Earner performs their work, then such a document would be relevant under the “whose business is it” test.

57However, such documents (if they exist and are not privileged) are not at issue on this motion. Uber advised the court that it produced such documents (so long as they were not privileged). The plaintiffs advised the court that they did not recall receiving any such documents. At the hearing, I advised the parties that if the parties cannot resolve this issue, then I would address the production of this limited set of documents (if they exist and are not privileged) at a subsequent case conference.

58In the present case, the plaintiffs seek any documents dealing with any development, marketing, design, and testing of any feature of the Uber Apps, regardless of whether the feature was ever used by Class Members during the class period. The plaintiffs seek these documents, even if they do not relate to the classification issues before the court, and instead only speak to the effectiveness of features Uber contemplated including in their apps. Internal documents relating to increasing engagement of Earners do not assist in determining whether the increased engagement arises as employees or independent contractors.

59The common issues judge will decide whether a common finding can be made on whether the structure in place created an employment or independent contractor relationship, based on all of the evidence, the non-exhaustive Sagaz factors, and the “whose business is it” test. Documents pertaining to how and why features on the Uber Apps were designed, marketed, developed and researched would result in a massive “fishing expedition” for irrelevant documents.

60Consequently, I find that these refusals were proper.

Questions seeking all pricing presentations

61The refused questions are questions 68-69 from the Nickels examination for discovery.

62The plaintiffs request production of all presentations related to the design of pricing structures in which Nickels was involved.

63As I have discussed above in relation to the questions on Uber App features, the reasons why Uber chose a particular pricing policy over other options are not relevant to the relationship between Uber and the Class Members. What is relevant is the actual pricing model that was in place during the class period. Even if a behavioural scientist or economist considered the effect of different pricing structures, such information is not relevant to the common issues judge who must consider whether the actual pricing structure is a factor supporting a common determination of employment or independent contractor status, or the need for a case-by-case analysis.

64By way of example, whether a worker is paid by the hour or by the piece might be relevant to the issue of whose business it is. However, what is relevant is the final payment structure, and the impact it had on the relationship between Uber and the Class Members.⁷ There is no reason for the common issues judge to consider pricing presentations which may (or may not have) resulted in the actual pricing structure.

65In addition to these questions being irrelevant, they are also overbroad; they seek any pricing presentation performed by anyone within Uber, regardless of the individual’s role at the company or the context surrounding the presentation. Consequently, even if I found that the questions sought relevant information (which I do not), I would dismiss the questions as overbroad.

Question on tipping function research

66The refused question is question 206 from the Nickels examination for discovery.

67The plaintiffs request production of research that was conducted by Uber, and which led Uber to include the tipping function in the Uber App.

68I rely again on my analysis of the questions relating to the Uber Apps features. What is relevant to the common issue is whether the tipping structure supports an employment or independent contractor relationship, and whether that issue can be determined in common. The reasons why Uber included the tipping function in the App (instead of allowing tipping to be done outside the App) are not relevant.⁸

Question seeking production of all hyperlinked documents in a document describing how the matching algorithm worked

69The refused question is question 479 from the Nickels examination for discovery.

70Uber produced a document describing how the matching algorithm worked. That document contained over 30 hyperlinks. Uber produced linked documents that explain features that were in use during the class period. The plaintiffs seek production of all other hyperlinked documents.

71I agree with Uber’s submission that only hyperlinked documents which relate to the relevant issue before the court ought to be produced. In the present case, the matching algorithm may be considered by the common issues judge when considering whether a determination can be made that the Class Members are, in common, employees or independent contractors or whether such a determination must be done on a case-by-case basis.

72Consequently, hyperlinked documents that do not relate to the relationship between Uber and the Class Members during the class period cannot assist the common issues judge.

73I note that even the hyperlinked documents that were produced must be treated with caution as the author of the Uber report on the matching algorithm has no control over whether any of the hyperlinked documents have been changed by the authors of those documents: Crookes v. Newton, 2011 SCC 47, [2011] 3 S.C.R. 269, at para. 27.

74Finally, the plaintiffs seek the code for the matching algorithm hyperlinked to the document. However, that code is not relevant to the common issue of the employment relationship. Code is a programming language used to tell a computer what to do: Eros - Équipe de Recherche Opérationnelle en Santé Inc. v. Conseillers en Gestion & Informatique C.G.I. Inc., 2004 FC 178, 258 F.T.R. 172, at para. 89. I agree with Uber’s submission that the code is not necessary to understand what, if any, control is exerted by the Uber Apps through the matching algorithm.

75The outputs of the code have already been produced through descriptions of the functionality of the matching mechanism on the Uber App and provide the relevant information.

Question for production of batching plans

76The refused question is question 101 from the Bashambu examination for discovery.

77The plaintiffs request production of “batching plans.” However, there is no foundation for this question.

78Bashambu understood that orders might be “batched” by an Earner who received a set of orders that would be picked up in succession. However, Bashambu had no knowledge of any document described by class counsel as a “batching plan”. Class counsel did not attempt to elicit from Bashambu what such a document might have been called or whether it even existed.

79The statement by counsel that a batching plan “is a plan that would have set out how batching works in Ontario during the class period” does not establish the requisite foundation.

80Consequently, the question is vague, unclear, and speculative, and as such the refusal is proper: Doucet v. The Royal Winnipeg Ballet (The Royal Winnipeg Ballet School), 2019 ONSC 6982, at para. 29(a).

Question about Uber Pool

81The refused question is question 717 from the Nickels examination for discovery.

82The plaintiffs request production of all documents relating to why Uber introduced Uber Pool, how it made its decision, and what testing it did. Uber Pool is a product which allows a Rider to request a ride with other Riders going in a similar direction.

83Again, as I discuss in relation to the Uber Apps features, what is relevant is how Uber Pool operated in Ontario during the class period. All other documents leading to its implementation are not relevant. Uber has produced documents related to the operation of Uber Pool which outline things such as how fares were calculated, how matching occurred and how Rider complaints were resolved.

84The internal Uber Pool planning documents sought by the plaintiffs are irrelevant. There is no basis to find that economic or behavioural science views on the feature relate to the common issue of the employment relationship during the class period.⁹ For Class Members that used the Uber Pool product, what is relevant is how that product worked, whether it supports direction and control, and how it factors into the “whose business is it” test. Uber’s choices as to how it (i) designed the Uber Pool product, or (ii) decided when and where to launch it in Ontario, are not relevant to the common issue.

Questions related to Uber’s use of behavioural scientists and other professions

85The refused questions are questions 802, 807, 808, 813, 998, 1120, 1121 from the Nickels examination for discovery and question 203 from the Bashambu examination for discovery.

86The plaintiffs ask a variety of questions about Uber’s use of behavioural scientists and other professionals in designing Uber Apps features. As I set out above, questions as to who was engaged in this process are irrelevant to the common issue of whether the features support a finding that each Class Member was an employee or independent contractor (or whether such a question must be answered on a case-by-case basis).

87By way of example, “surge pricing” is a feature on Uber Apps which increases fares when there is greater demand from Riders than there are Drivers. The plaintiffs rely on surge pricing as a factor to establish that Class Members were directed when and where to work. Uber submits that Class Members were not required to drive during surge pricing, and as such were independent contractors or that the relationship can only be determined on a case-by-case basis.

88Evidence as to the use of behavioural scientists, psychologists, or economists to determine the best way to implement surge pricing does not assist the common issues judge; this evidence will not shed light on the nature of the relationship between Uber and the Class Members during the class period.¹⁰

Question on production of documents related to surveys

89This refused question is question 822 from the Nickels examination for discovery.

90The plaintiffs request production of all documents related to surveys conducted of Class Members. Uber has agreed to produce the results of these surveys but refuses to produce all of the underlying documents.

91Again, the plaintiffs’ focus on how and why the documents were created does not address the common issue of the employment relationship between the parties. While the plaintiffs can seek to rely on the survey questions and results as evidence of an employer-employee relationship, such a submission would be based on the actual surveys provided to the Class Members, not on questions which were considered but not asked, or the reasons for the survey questions.¹¹

92Consequently, the refusal is proper.

Questions concerning Uber participation with academics, think tanks or consultants to publish reports

93The refused questions are questions 815 and 817 from the Nickels examination for discovery.

94The parties have resolved these refusals on the basis that in advance of trial or a motion for summary judgment, Uber will advise whether any of the academic research it intends to rely on was commissioned or sponsored by Uber.

Questions related to offer card testing

95The refused questions are questions 248-251 from the Bashambu examination for discovery.

96The plaintiffs request that Uber advise if they tested different ways to have Drivers accept offers. As with my analysis of the Uber Apps features, whether Uber tested different methods of how Drivers might accept offers is not relevant. What is relevant are the offer cards which Uber implemented during the class period. Uber has responded to undertakings regarding the information displayed on offer cards during the class period and have responded to questions about the goals they are seeking to achieve when designing offer cards.

97Again, it is open to the plaintiffs to rely on the offer cards when asking the common issues court to make a common finding that the Class Members were employees, just as Uber can rely on the offer cards when asking the common issues court to make a common finding that the Class Members were independent contractors, or that a case-by-case analysis is required. However, the how and why of the testing of the offer cards is not relevant.¹²

Questions relating to key performance indicators

98The refused questions are questions 285, 292 and 301 from the Bashambu examination for discovery.

99A key performance indicator (“KPI”) is a general business term used to describe a measure of something. These KPIs were for Uber’s employees (not for Earners).

100The plaintiffs seek production of the KPIs for all of the groups in Uber’s entire organization.

101These questions are not relevant for the following reasons:

(i) The common issue is the employment classification. Whether an Uber employee had a KPI relating to the number of Earners does not assist the common issues judge in determining whether the relationship that existed, based on the Uber Apps, Service Agreements, Community Guidelines, and Uber’s procedures and practices, created a common employee or independent contractor relationship.

(ii) Further, even if the KPIs were relevant to the employment classification issue (which I do not find), the evidence sought would not assist the court in understanding whether any KPI was met or the reason why it was/was not achieved. Only an analysis of the KPI (for example, acceptance rates) could determine whether the changes made to the Apps resulted in increased acceptance.

(iii) In the alternative, the plaintiffs seek production of KPIs related only to those people involved in the development of the Uber Apps. The request remains irrelevant (see subparagraph (i) above), and is also overbroad as it encompasses KPIs from Uber’s business not related to its dealings with the Class Members. Merchants, Eaters, and Riders who use the Uber Apps are not at issue in this class action, so any KPIs related to those individuals would be overbroad.

Conclusion on the plaintiffs’ refusals motion

102For the above reasons, I dismiss the plaintiffs’ refusals motion in its entirety.

Uber’s motion for leave to examine ten additional Class Members, pursuant to s. 15(2)

103Under s. 15(2), a party may move for discovery of other class members following the discovery of the representative parties. Section 15(3) sets out a list of factors the court shall consider when deciding to grant leave.

104Uber seeks an order under s. 15(2) for leave to examine ten additional class members. The plaintiffs oppose the request, and in the alternative, propose a process by which (i) the plaintiffs would select the additional Class Members; (ii) the number of Class Members would be reduced (although the plaintiffs did not provide a specific number); (iii) the additional examinations for discovery would take place in writing; and (iv) if the additional examinations for discovery proceed orally, their duration should be for less than what Uber has proposed (one full day per each Class Member).

105For the reasons that follow, I grant the request for leave under s. 15(2). I also generally accept the process proposed by Uber. That is, (i) permitting Uber to select the Class Members; (ii) maintaining the number of additional Class Members to be examined (ten); and (iii) conducting oral examination for discovery. The only modification I order to the proposed process is to reduce the time required for examination for discovery to a half day for each additional Class Member.

106I first consider the general principles governing examination for discovery in a class action and the test for leave under s. 15(2). I then review each of the factors under s. 15(3). Lastly, I consider the process to be applied for the examinations for discovery of the additional Class Members.

General principles

107When considering a motion for leave under s. 15(2), I rely on the following general principles:

(i) Discoveries allow all parties to: (a) know the case they must meet, (b) procure admissions to build their case, (c) procure admissions to undermine the other side’s case, and (d) narrow the issues: Besner v. Her Majesty the Queen in Right of Ontario, 2011 ONSC 7335, at para. 20.

(ii) Examinations for discovery play an essential role in facilitating settlement: Besner, at para. 21.

(iii) At trial, transcripts from the examinations for discovery of other class members are treated as those of a party: Rules of Civil Procedure, R.R.O. 1990, Reg. 194, r. 12.03(1). Consequently, they can be read in as evidence: Rule 31.11.

(iv) Obtaining admissions for the purpose of reading in evidence at trial is an important aspect of the discovery process: Hancock v. Lunn, 2010 ONSC 3621, at para. 40.

(v) In the class actions context, prior to the common issues trial, the scope of discovery is “defined by and limited to the common issues”: Abdulrahim v. Air France, 2010 ONSC 3953, at para. 13. This is consistent with the CPA’s purpose and bifurcated trial procedure: Doucet, at paras. 29-30.

(vi) The mere fact that individual issues may arise does not open the door to discovery on individual issues prior to the common issues trial: Abdulrahim, at para 13.

(vii) In adapting discovery procedures for class proceedings, the Legislature adopted the recommendations of the Ontario Law Reform Commission in its Report on Class Actions (1982). In particular, defendants should have the same right of discovery against representative plaintiffs as in ordinary civil proceedings, but defendants should not have a right to discover additional class members without leave of the court: Lipson v. Cassels Brock & Blackwell LLP, 2019 ONSC 5524, at para. 38.

(viii) Discovery of individual class members is the exception rather than the rule: Caponi v. Canada Life Assurance Co., 2013 ONSC 1672, at para. 23.

(ix) Parties can move for discovery of additional class members pursuant to ss. 15(2) and 15(3) of the CPA if the information they are seeking is either: (a) not available to the representative plaintiff(s); and/or (b) not readily obtained: Andersen v. St. Jude Medical Inc., 2006 CarswellOnt 9773 (S.C.), at para. 19.

The stage of the proceeding and the issues to be determined at that stage – s. 15(3)(a)

108I first consider the stage of the proceeding and then review the issues to be determined at that stage.

(i) The stage of the proceeding

109Under s. 15(2), additional class member discovery is only available after discovery of the representative plaintiffs is complete.

110A motion under s. 15(2) can be brought prior to a common issues trial. It is not necessary to wait until after the trial of the common issues: Coburn and Watson’s Metropolitan Home v. Bank of Montreal, 2018 BCSC 897 (“Coburn SC”), at paras. 24-25, aff’d 2018 BCCA 432 (“Coburn (CA)”).

111Similarly, in Levac v. James, 2020 ONSC 2606, the court ordered examination for discovery of additional class members after certification and before the common issues trial, when the questions to be answered by the additional class members were “relevant to the common issues as defined in [the ‘judgment reconstituting the common issues in this already certified class action’]”: at paras. 2 and 25.

112Uber brought this motion at the earliest possible opportunity, after discovery of the representative plaintiffs was complete, which occurred in March 2026 when the representative plaintiffs completed their answers to undertakings. The additional examinations will not require any adjournments or delay to the proceeding because no trial date has been set.

113Further, Uber advised the court that it is prepared to conduct the examinations for discovery as quickly as reasonably possible to ensure the common issues trial is not unduly delayed.

114Consequently, Uber has brought the motion at the appropriate stage.

(ii) The issues to be determined at this stage

115This factor addresses the common issues for which the proposed examination for discovery evidence is sought.

116The plaintiffs dispute the relevancy of the information sought from the additional Class Members. The plaintiffs submit that what Uber is seeking is individual evidence from each of the proposed Class Members, relevant only to individual issue trials.

117The plaintiffs submit that the proposed examinations do not relate to a dispute on common facts (as in Berry v. Pulley, 2007 CanLII 16446 (Ont. S.C.), at paras. 49-50) or determining an issue in common (as in Levac).

118I do not agree with the plaintiffs’ submission.

119As I have discussed above, the common issue in the present class action is whether the court can find, at a common issues trial, that (i) the Class Members are either employees or independent contractors, or (ii) the court requires a case-by-case analysis to determine the employment relationship for each Class Member. The individual Class Member evidence is relevant to the common issue as it will provide the court with: (i) evidence relevant to the court’s understanding of the licensing agreements, Community Guidelines and app functionality, and (ii) how each Earner’s individual choices impact the classification test, such that a class-wide determination is not possible.

120I address each use of the evidence below.

(a) The evidence is relevant to the court’s understanding of the licensing agreements, Community Guidelines and app functionality

121The plaintiffs rely on the Service Agreements, Uber Apps, Community Guidelines, and Uber’s policies and procedures to ask the common issues judge to find a common employer-employee relationship.

122However, the flexibility and entrepreneurialism of a Class Member cannot be determined solely through a review of what is permitted under the documents. Uber will not only rely on what the documents say; they will also rely on the Class Members’ evidence about how, when, and where they chose to work, and the reasons behind these choices.

123For example, dual-apping is permitted under the Service Agreements. However, in international decisions, courts have rejected Uber’s attempts to rely only on the Service Agreements as evidence of what Earners can and cannot do. Courts in New Zealand and the United Kingdom have minimized or disregarded an Earner’s contractual ability to dual-app on the basis that it is not, in fact, practical to do so: Rasier Operations BV v. E Tū Incorporated, [2025] NZSC 162 at para. 124; Uber BV Ors. v. Aslam Ors., [2021] UKSC 5, at paras. 135-137.

124Uber is entitled to lead evidence to respond to similar concerns which may be raised by the plaintiffs in the present class action.

125A Class Member who chooses to dual-app frequently, with multiple apps open at the same time, could provide evidence relevant to a common finding of independent contractor status. Uber will submit that every Class Member has the freedom to choose how to operate. The ability to make decisions within the Uber Apps is a factor Uber is entitled to rely on to ask the court to make a common finding that all Class Members are independent contractors. In contrast, a Class Member who never used the dual-app feature may have had less of an entrepreneurial role, which may support the plaintiffs’ position that they had employee status.

126Evidence on how and why a Class Member chooses to dual-app, or make any other choices relating to how, when, and why they work, will assist the common issues court in understanding how Uber’s constating documents, which govern the Uber structure, operate in reality.

127The common issue before the court is whether the Uber structure directs and controls the work of the Class Members (as the plaintiffs submit) or whether the Class Members exercise control as to how, when, and where to work as independent contractors (as Uber submits), and whether such a determination can be made in common. That decision by the common issues court requires evidence as to the actual choices made by Class Members, and not merely a hypothetical consideration of the language in the constating documents.

(b) The evidence is relevant to how each Earner’s individual choices impact the classification test, such that a class-wide determination is not possible

128The common issues judge will have to determine whether a common finding can be made as to whether the Class Members are employees or independent contractors. In his certification decision, Justice Perell held that one possible answer to this question is that the determination of employee or independent contractor status can only be done on a case-by-case basis. The evidence of the choices made by the additional Class Members is relevant to this determination.

129The plaintiffs submit that the seemingly flexible structure, which allows Earners to choose when and where to work, ignores the realities of the relationship. For example, the plaintiffs plead that Uber “incentivizes Class Members to be consistently available for work and penalizes them…if they are not sufficiently available.”

130In light of the plaintiffs’ theory, which asks the court to look beyond just the written agreements, Uber must lead evidence on Class Members’ responses to incentives Uber offers. Uber may need this evidence to establish that the relationship is one of an independent contractor or one that must be determined on a case-by-case basis.

131By way of example, Mr. Ruokonen rejected or cancelled over 85% of the offers made to him. Consequently, his evidence might not support the plaintiffs’ submission that Uber “incentivizes [him] to be consistently available for work” and that Uber penalizes Earners who are “not sufficiently available.” If Mr. Ruokonen gives evidence that he was unmoved by Uber’s incentives, and faced no penalty for not being “consistently available”, this will support Uber’s position that Class Members are independent contractors, or that a case-by-case analysis is needed..

132For the above reasons, I find that this factor strongly supports Uber’s request for leave under s. 15(2).

The presence of certified subclasses – s. 15(3)(b)

133The presence of certified subclasses is not necessary for leave to be granted to examine additional class members: e.g. Kirk v. Executive Flight Centre Fuel Services Ltd., 2023 BCSC 2251, at paras. 14, 33-34.

134However, even if there are no subclasses, the court can consider whether there are different groups of class members: Coburn SC, at paras. 44-48, aff’d Coburn CA, at para. 20. In cases where there are different groups of class members, this factor may weigh in favour of granting leave for the examination of additional class members (even if considered under s. 15(3)(f) instead of s. 15(3)(b)).¹³

135Consequently, without subclasses but with recognized distinctions in how the Class Members operate within the Uber structure, this factor favours leave.

The necessity of discovery – s. 15(3)(c)

136The plaintiffs submit that additional Class Member evidence is not necessary for two reasons: (i) the issues that Uber wishes to obtain evidence on are not in dispute; and (ii) Uber already has the information it is seeking from (a) its own documents, (b) the Class Members’ affidavits it filed when opposing certification, and (c) informal discussions such as roundtables.

137I do not agree. I address each of these submissions below.

(i) The issues on which the proposed Class Members are to be examined are in dispute.

138In their responding factum, the plaintiffs purport to make the following concession:

The Plaintiffs concede that the Service Agreements and the functionality of the Uber App permit Class Members to: (a) operate through the use of a corporation; (b) dual-app with competitors; (c) provide different levels of work to Uber; (d) utilize different strategies to try and maximize their income with Uber; (e) choose when and where to work; (f) consider themselves independent contractors in business for themselves; (g) obtain various levels of training from Uber; and (h) “utilize a variety of different tools in providing transportation or delivery services”. None of this is disputed.

139The above is not a true concession of any of the issues in dispute. The plaintiffs are merely acknowledging the obvious: the language of the Service Agreements and the functionality of the Uber Apps permit the Class Members to make the choices described above.

140However, this case is about what the Service Agreements, other constating documents and practices of Uber “do”, and not what they “say”. The plaintiffs’ case depends on a finding by the court that despite what the relevant documents “say”, the Class Members are, in fact, directed by Uber on how, when and where to work under the Uber structure.

141The following “concessions” in the plaintiffs’ factum are contradicted by the plaintiffs’ Amended Fresh as Amended Statement of Claim (the “Claim”):

(i) Dual-apping: The plaintiffs purport to concede that Class Members are permitted to dual-app with competitors’ apps. However, in the Claim, the plaintiffs allege that Uber penalizes Class Members who refuse requests while logged into the App and are not otherwise transporting customers or deliveries (para. 20(p)).¹⁴

(ii) Work at different levels: The plaintiffs purport to concede that Class Members are permitted to provide different levels of work to Uber. However, in the Claim, the plaintiffs allege that (a) the Class Members are dependent on Uber for their economic well-being (para. 17) and (b) as a so-called Uber “Partner” and/or independent contractor, Heller (and other Class Members) worked between 40-50 hours per week, including weekends and holidays (para. 45).

(iii) Use of strategies to maximize income: The plaintiffs purport to concede that Class Members are permitted to utilize different strategies to try and maximize their income with Uber. However, in the Claim, the plaintiffs allege that (a) Uber directs the Class Members’ routes and penalizes them if they deviate from that route (para. 20(l)) and (b) all prices and “Partner” compensation are fixed by Uber and frequently changed at Uber’s discretion (para. 20(j)).

(iv) Choice of when and where to work: The plaintiffs purport to concede that Class Members are permitted to choose when and where to work. However, in the Claim, the plaintiffs allege that (a) at all material times, the Class Members were directed how, when, and where they could provide driving services for Uber’s customers (para. 43); (b) Uber required the Class Members to consistently be available for work or risk reduced pay and loss of incentives and/or promotions (para. 44); (c) Class Members who reduce their hours of work were denied special promotions, incentives and pay increases (para. 20(o)); and (d) Uber incentivizes Class Members to be consistently available for work and penalizes them through loss of incentives, promotions, and/or access to the App, and therefore their ability to work, if they are not sufficiently available (para. 20(q)).

(v) Use of tools: The plaintiffs purport to concede that Class Members are permitted to utilize a variety of different tools in providing transportation or delivery services. However, the plaintiffs allege in the Claim that (a) Class Members must use Ubers’ tools in the form of delivery bags, data, delivery signs, and the Uber App (para. 20(d)); (b) Class Members are told the type of vehicle they must use and the category of license they must possess in order to work for Uber (para. 20(e)); and (c) Uber requires Class Members to use specific types of phones and phone carrier networks and also requires Class Members to have data plans with at least 10GB per month (para. 20(g)).

142The common issue is whether the court can find that the Class Members are collectively employees or independent contractors, or whether a case-by-case approach is required. That issue depends on whether the common issues judge can determine whose business it is on a collective basis. The plaintiffs do not concede that the Service Agreements, the Uber Apps, the Community Guidelines, or any other aspect of the Uber structure (regardless of the text of the constating documents) provide Class Members the ability to direct or control their work, use their own tools, make strategic decisions to maximize their income, or provide different levels of work.

143By way of example, the plaintiffs have taken the position in other proceedings that Uber’s promotions and incentives “have the effect of directing delivery people to work at certain times and in certain places”: see the submissions of the class in OLRB Case No. 2426-21-ES, at para. 45. The plaintiffs take the same approach in this class action.

144The evidence of the proposed additional Class Members is essential for the common issues court to determine the choices that are actually available to Class Members, as opposed to engaging in a hypothetical analysis based on the constating documents. The evidence from the proposed additional Class Members will address the effect of the Uber structure on the employment relationship, which is the seminal contested issue in this case.

(ii) Uber does not have the information it is seeking

145The plaintiffs submit that Uber has the information it needs from (a) its own records, (b) the Earner affidavits (filed in connection with the certification motion), and (c) informal conversations, including roundtables, with Earners. I do not agree.

(a) Uber’s records do not provide the relevant information for the common issues trial

146Uber’s records only show how, when, and where a Class Member provided service as an Earner; the explanation for “why” a Class Member made the choices they did under the Uber structure is missing. This explanation will be relevant for the common issues judge, who will consider factors such as direction and control to determine whether the Class Members are employees or independent contractors, or if a case-by-case analysis is needed.

147For example, Uber knew the type of vehicle driven by Garcia. However, it was only through examination for discovery that Uber learned that: (i) Garcia purchased her car specifically for providing delivery services; (ii) she calculated that she would make more money using a car, rather than a bike; (iii) she concluded that she should buy, and not use a loaner car; and (iv) she made a conscious choice to purchase a fuel-efficient car to increase her earnings.

148A Class Member’s specific choices, and the reasons underlying the choices, may be considered by the common issues judge when determining whether the Uber structure in fact results in an employee or independent contractor relationship. However, as in Garcia’s case, this information may not be revealed through Uber’s own records; they may only emerge through examinations for discovery.

149Similarly, the fact that Uber knows where and when an Earner provides their services does not explain why the Earner made those choices. For example, if an Earner tends to work downtown, the plaintiffs may submit at a common issues trial that it is because the Earners are effectively compelled to do so by Uber’s promotions and incentives. However, an Earner’s individual evidence may be that they work downtown because it is close to home, because Merchants are faster and more experienced, or because Riders coming from downtown tend to tip more. For instance, Ms. Scott stated on a podcast that she preferred to deliver downtown because she knew it would be busy and lucrative.

150These additional facts cannot be found in Uber’s documents but are relevant to the entrepreneurial nature of the Earner’s work.

(b) The certification affidavits and informal conversations do not constitute evidence at a common issues trial

151I agree with Uber’s submission that the certification affidavits and informal conversations do not constitute evidence at a common issues trial. I rely on the following submissions from Uber’s reply factum, quoted verbatim:

(i) Both the certification affidavits and the informal conversations contain information from those who volunteer to participate and share their experience and documents. The discovery process is the only means by which the Defendants can compel class members to provide information or documents.

(ii) The certification affidavits and informal conversations cannot be read in as evidence at trial: Rule 31.11. In addition, informal conversations with Earners are not under oath. The fact that Mr. Ruokonen told an Uber employee over the phone that he used multiple apps is no substitute for an examination for discovery under oath which results in a transcript that can be read in at a trial.

(iii) After certification, Defendants’ counsel is prohibited from contacting the class members under Rule 7.2-6 of the Rules of Professional Conduct. There is no reasonable ability for the Defendants to collect information and documents from Earners for the purpose of preparing for trial after certification.

(iv) The evidence in the certification affidavits was obtained for the purpose of the procedural certification test in s. 5 of the Class Proceedings Act. It was not obtained for the purpose of the common issues trial. It would be unfair to retroactively cabin the Defendants’ ability to gather any evidence beyond this.

152Consequently, the s. 15(3)(c) factor strongly favours leave to examine the additional Class Members.

The approximate monetary value of individual claims – s. 15(3)(d)

153The amounts sought in this class action are substantial. The plaintiffs claim $200 million in damages with an additional $200 million for punitive, aggravated and exemplary damages.

154The plaintiffs submit that Uber is seeking to obtain evidence from the proposed additional ten Class Members in order to reduce the plaintiffs’ proposed aggregate damages claim. However, the appropriateness of sampling is not at issue under s. 15(3)(d).

155Instead, under s. 15(3)(d), the court is required to consider a defendants’ potential exposure to damages as a factor which could justify the examination for discovery of additional class members. Put differently, a defendant facing a $200 million judgment may be entitled to more discovery rights compared to a defendant facing a $200,000 claim.

156Consequently, this factor favours leave under s. 15(2).

Whether discovery would result in oppression or in undue annoyance, burden or expense for the class members sought to be discovered – s. 15(3)(e)

157Uber seeks to examine less than 0.003% of the Class Members. There is no evidence that such examinations for discovery “would result in oppression or in undue annoyance, burden or expense for the class members sought to be discovered”, as set out in s. 15(3)(e).

158The proposed additional Class Members have been selected because of Uber’s evidence that these individuals are expected to have unique and relevant evidence. The plaintiffs have led no evidence about the alleged burden on these individuals.

159The plaintiffs submit that the examinations for discovery are likely to be unduly burdensome because: (i) the Class Members are, to class counsel’s knowledge, not already engaged in the issues; and (ii) class counsel does not have any direct relationship with these Class Members. I do not agree.

160First, there is no case law to support the plaintiffs’ submission. In any class action, the examination of additional class members would likely be conducted for individuals who are not necessarily engaged in the issues or have no direct relationship with class counsel, unlike the representative plaintiff. Imposing such requirements under s. 15(3)(e) is not consistent with a leave process that considers factors relevant to the appropriateness of additional examinations for discovery.

161Second, even if the plaintiffs’ proposed requirements were accepted, there is no evidence to support the plaintiffs’ position.

162Uber led evidence that the additional Class Members are “engaged in the issues.” Two of them hold leadership positions in delivery and rideshare worker organizations. Three of them have volunteered information about their experiences to Uber either in telephone calls or at “roundtables” at Uber’s offices. Three of them have active online or social media presences where they speak about their use of the Uber Apps. Four of them have given interviews to the media or spoken at legislative standing committees about their use of the Uber Apps. In 2025, one of them, Ms. Scott, testified at an Ontario Labour Relations Board proceeding against Uber. Additionally, given the Class Members’ documented willingness to speak to the media, legislative committees and Uber about their experiences, there is no evidence that it will be unduly burdensome to contact these Class Members or for these Class Members to participate in discovery.

163Further, there is no evidence that class counsel has attempted to establish a relationship with these individuals.

164Lastly, there is no suggestion that Uber would conduct the examinations in an “oppressive” manner or that would create an “undue” annoyance, burden or expense for the Class Members. To the contrary, Uber seeks to minimize the burden faced by the proposed additional Class Members by allowing examinations to take place virtually or in-person. Additionally, given class counsel’s experience and ability to prepare Class Members, the burden on any individual would not be “undue”.

165Consequently, this factor strongly favours granting leave for examination for discovery of the ten additional Class Members.

Other factors – s. 15(3)(f)

166In this case, the importance of obtaining admissions and read-in evidence is a significant factor. With every Class Member represented by class counsel after certification, Uber has no other way to contact individuals with relevant evidence to lead that evidence at trial. While Uber could summon any Class Member at trial, Uber should not be required to prepare for trial without the essential admissions and read-in evidence available from examination for discovery, particularly because the representative plaintiffs are unable to provide evidence on how those Class Members operate under the Uber structure.

167Consequently, while the s. 15(3)(f) analysis is not necessary for my decision to grant leave (as I find that all the s. 15(3)(a) to (e) factors favour granting leave), I find that s. 15(3)(f) also favours leave being granted.

Conclusion on granting leave

168For the above reasons, I grant leave to Uber to examine the ten additional Class Members for discovery.

Process for additional discovery – the plaintiffs’ alternative submission

169In the alternative, the plaintiffs submit that “should the Court determine that examination of additional Class Members is appropriate in the circumstances, Uber’s proposed process (or lack thereof) is inappropriate and potentially prejudicial to the Class.”

170The plaintiffs propose a process in which (i) the plaintiffs (or a combination of plaintiffs and Uber) would select the additional Class Members; (ii) the number of Class Members would be reduced (although the plaintiffs did not provide a specific number); (iii) the additional examinations for discovery would take place in writing; and (iv) if the additional examinations for discovery proceed orally, they should be for less than the one full day for each additional Class Member.

171Uber proposes a process in which (i) Uber selects the additional Class Members; (ii) there would be ten additional Class Members examined for discovery; (iii) the additional examinations for discovery would take place orally; and (iv) if the additional examinations for discovery proceed orally, they should be for one full day for each additional Class Member.

172I find that Uber’s proposed process is appropriate given the evidence before the court, except for the length of the proposed examinations for discovery, which I would reduce to a half day for each additional Class Member.

173I first consider the applicable law and then review each of the contested elements of the process proposed by Uber.

(i) The applicable law

174In the case law relied upon by the parties, courts have applied different approaches to the discovery process of additional Class Members.

175In Kirk, the court ordered that 26 additional class members be examined to reflect 27 listed specific land use categories. The court adopted the defendants’ proposal that (i) they would nominate one class member for each category and the plaintiff would nominate one as well and (ii) the examinations for discovery of each additional class member would be limited to two hours: at paras. 14-16, 34. The court did not refer to any counter-proposals from the plaintiff which challenged the defendants’ proposal.

176In Cobourn SC, there was no reference to the defendants seeking to select the additional class members. The court ordered the plaintiff to provide a list of eight additional class members to be examined for discovery, based on the eight categories of merchants who comprised the class. The court ordered that each examination for discovery be conducted orally with time limited to one day for each additional class member: at paras. 63, 64, 67. Again, the proposed process does not appear to have been contested.

177In Levac, the court ordered that five additional class members, as selected by the defendant, attend for examinations for discovery, which was to take place in writing, largely because much of the evidence sought related to medical records: at paras. 11, 13, 21, 23.

(ii) Process issue 1: Should Uber select the additional Class Members?

178I agree with Uber that on the evidence before the court, Uber should be entitled to select the additional Class Members for examination for discovery.

179The plaintiffs submit that Uber has offered “no principled basis” to explain why the ten Class Members were proposed and that it has “merely asserted” that the Class Members have unique and relevant evidence. However, that submission is contrary to the evidentiary record before the court.

180Uber filed two affidavits containing voluminous information about these individuals. Based on this evidence, Uber provided a detailed list of the information and admissions they hope and expect to obtain from these examinations, including information relating to the advantages of operating through a corporation, how and why class members dual-app, financial investments made by the Class Members, and how Class Members react to Uber’s incentives and promotions.

181In these circumstances, the proposed examination for discovery of each additional Class Member is supported by uncontested evidence as to their unique features and engagement with issues in the action. There is no evidence to suggest that Uber has improperly selected any proposed Class Member.

182Consequently, since the uncontested evidence supports a reasonable basis for the selection, I follow the approach in Levac, where the court ordered that the additional Class Members be those selected by the defendant.

(iii) Process issue 2: Should the number of proposed additional Class Members be reduced?

183The plaintiffs rely on Coburn SC, where only eight additional class members were ordered to attend for examination for discovery.

184However, each case must depend on its own facts. In Coburn SC, there were eight distinct categories of merchants: at para. 64. Consequently, the order for additional examination for discovery of class members was based on that evidence (the proposed number was not challenged).

185However, in Kirk, 26 additional class members were ordered to be examined given that such an amount reflected the 27 land use categories of the class members (with the representative plaintiff falling into only one of such categories): at para. 12.

186In the present case, Uber’s evidence establishes the equivalent of ten “categories” of Class Members. While there is some overlap in the choices the additional Class Members made regarding how, when, and where to work, each proposed Class Member has at least one unique feature which supports an examination for discovery to obtain evidence relevant to the common issue.

187In these circumstances, I find that the examination of ten additional Class Members is reasonable.

(iv) Process issue 3: Should the examinations for discovery be conducted orally?

188The plaintiffs ask the court to follow the approach in Levac and order that the additional examinations for discovery be in writing. I do not agree.

189As I have discussed above, in Levac, a major reason why additional examinations for discovery were ordered was to obtain additional medical records. This process would be done more efficiently in writing.

190However, in both Kirk and Coburn, the courts ordered that the additional examinations for discovery be conducted orally (again, a process that does not appear to have been opposed). In those cases, the questions to be asked of the proposed additional class members were either based on land use (Kirk) or specific merchant conduct (Coburn), such that the examinations would not be conducted efficiently in writing.

191Similarly, in the present case, asking the additional Class Members about the choices they made and the reasons for those choices cannot proceed efficiently in writing. There will be nuanced answers to the questions, which I anticipate will require follow-up questions. It will not be possible to properly canvass the direction and control issues without oral discovery.

(v) Process issue 4: The length of the examinations for discovery

192As I have discussed above, courts have ordered varying lengths of examinations for discovery, with two hours for each additional class member in Kirk and a full day for each of the merchants in Coburn SC.

193The present case does not raise the same complexity as the additional discovery in Coburn, which required a thorough review of the economic factors related to credit-card fees for each category of merchant. Instead, the present case is more like Kirk, where each land-use category class member would be asked a more limited series of questions.

194The Class Members in the present case have important evidence on the common issue, but it is confined to their choices and the reasons for those choices. In such circumstances, I am satisfied that a half day is sufficient for the examinations for discovery of each additional Class Member.

(vi) Other process issues

195At the hearing, the plaintiffs raised several other process issues, which I address briefly below:

(i) Scope of questions: In Coburn SC, the court limited the scope of the additional discovery “to enquiries that are reasonably necessary to legitimately test the economic methodologies and models being advanced and relied upon by the plaintiff’s economic expert(s)”: at para. 65. The plaintiffs in the present case ask the court to limit the questions in some form, perhaps to issues such as their decisions, choices, and reasons.

I do not set out specific wording to restrict questions in this case. However, the questions must be relevant to the common issues – if not, then the court may need to resolve such disputes which could be addressed in a summary manner at a case conference.

(ii) Production of documents: Both parties agreed that prior to the additional examinations for discovery, each of the additional Class Members and Uber would produce all documents related to the Class Member which are relevant to the common issues. The parties will attempt to agree on a timetable to do so, failing which I will set a deadline at a case conference.

(iii) Timing of the additional examinations for discovery: The plaintiffs raised a concern about potential delay in moving the action to trial in light of the additional examinations. I advised all parties that I expected them to move promptly toward scheduling dates for the additional examinations for discovery. Each party has a large group of extremely experienced and sophisticated counsel. Time should be found promptly to arrange the examinations for discovery, with a timetable to be agreed upon by the parties or set by the court at a case conference.

ORDER AND COSTS

196For the above reasons, I dismiss both refusals motions and grant Uber’s s. 15(2) motion, subject to the terms I have set out above. Counsel asked for the court’s permission to provide written costs submissions after the release of these reasons, since the parties wished to consider their collective success on the various motions before the court. I agreed to this request.

197If the parties cannot agree on costs, they may each deliver brief costs submissions of no more than three pages, to collectively address all the motions, by August 14, 2026. Each party may deliver responding costs submissions of no more than three pages, to collectively address all of the

motions, by August 28, 2026, The parties may submit a reply costs submission of no more than one page, to collectively address all of the motions, by September 4, 2026.

GLUSTEIN J.

Date: 20260709

CITATION: Heller v. Uber Technologies Inc., 2026 ONSC 4102

Footnotes

1. 1Unless otherwise stated, all statutory references are to the CPA.
2. 2In his examination for discovery, Nickels confirmed that Uber is the ultimate decision-maker about which terms are included in the Community Guidelines.
3. 3Between January 1, 2012 and March 1, 2021, 366,359 Earners provided at least one ride or delivery in Ontario using the Uber App: Heller, at paras. 57, 72.
4. 4See the recent decision in Carcillo v. Ontario Major Junior Hockey League, 2025 ONCA 652, where the court describes the some basis in fact test at para. 41 as follows: “The evidentiary standard — ‘some basis in fact’ — reflects this low bar. This standard requires only minimal evidence showing that the proposed common issues exist and could be addressed class-wide. The appellants need not demonstrate a prima facie case or establish the likelihood of trial success, and the court cannot weigh the evidence or evaluate the merits at certification”.
5. 5The “whose business is it” test (which considers “[w]hether or not the activity of the agent is part of the business organization of the principal for which he works” was stated in Braiden v. La-Z Boy Canada Limited, 2008 ONCA 464, 294 D.L.R. (4th) 172, at para. 34, to be “the heart of the matter” in determining whether a person is an employee or an independent contractor. The court relied on the decision in 671122 Ontario Ltd. v. Sagaz Industries Canada Inc., 2001 SCC 59, [2001] 2 S.C.R. 983, at paras. 43 and 50.
6. 6References in this paragraph refer to the paragraph number from the plaintiffs’ factum.
7. 7If there was an internal document stating that the reason for paying by the piece was to avoid ESA obligations, that document would be relevant and would have to be produced if not privileged. As I discuss above, however, to the extent Uber has any such documents, Uber advised the court that they have produced them.
8. 8Unless those reasons were based on the employment relationship, such as ESA obligations – see footnote 7 above.
9. 9See footnote 8 above.
10. 10See footnote 8 above.
11. 11See footnote 8 above.
12. 12See footnote 8 above.
13. 13In Coburn SC, the court held that “insight into the perspective of Class members” was required to address the “distinctive characteristics” of the conspiracy claim advanced in that case, as a factor under the BC equivalent of s. 15(3)(f): at para. 62.
14. 14In this paragraph, references are to the relevant paragraphs in the Claim.