IN RE: SOCIAL MEDIA ADOLESCENT ADDICTION/PERSONAL INJURY PRODUCTS LIABILITY LITIGATION

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
SAN FRANCISCO DIVISION
IN RE: SOCIAL MEDIA ADOLESCENT
ADDICTION/PERSONAL INJURY
PRODUCTS LIABILITY LITIGATION
Case No. 22-md-03047-YGR (PHK)
ORDER RESOLVING STATES’
MOTION TO COMPEL META TO
PRODUCE FURTHER METADATA
FOR CERTAIN TRIAL EXHIBITS
Re: Dkts. 3023 and 3346
INTRODUCTION
This MDL has been referred to the undersigned for all discovery purposes. See Dkt. 426.
Now pending before the Court is a joint letter brief regarding a dispute between the State Plaintiffs
and Meta about whether Meta should be required to produce additional metadata for certain trial
exhibits. [Dkt. 3023]. After receiving a procedurally improper email from counsel regarding this
dispute, the Court directed the Parties to meet and confer further and then file a Further Joint
Discovery Letter Brief. 	[Dkt. 3338]. 	The Parties submitted their Joint Further Discovery Letter
Brief on August 7, 2026. See Dkt. 3346. Neither Party has requested oral argument, and the Court
finds this dispute suitable for resolution without oral argument and resolves the dispute as set forth
herein. See Civil L.R. 7-1(b).
LEGAL STANDARDS
A party seeking discovery bears the burden of establishing that its request satisfies the
relevancy and proportionality requirements under Rule 26(b)(1). La. Pac. Corp. v. Money Mkt. 1
Inst. Inv. Dealer, 285 F.R.D. 481, 485 (N.D. Cal. 2012). The resisting party, in turn, has the burden
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 1 of 11

2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
to show that the discovery should not be allowed. Id. The resisting party must specifically explain
the reasons why the request at issue is objectionable and may not rely on boilerplate, conclusory, or
speculative arguments. Id.; see also Blankenship v. Hearst Corp., 519 F.2d 418, 429 (9th Cir. 1975)
(“Under the liberal discovery principles of the Federal Rules defendants were required to carry a
heavy burden of showing why discovery was denied.”).
The relevant legal standards for discovery are well-known. Rule 26(b)(1) delineates the
scope of discovery in federal civil actions and provides that “[p]arties may obtain discovery
regarding any nonprivileged matter that is relevant to any party’s claim or defense and proportional
to the needs of the case.” Information need not be admissible to be discoverable. Id. Relevancy for
purposes of discovery is broadly defined to encompass “any matter that bears on, or that reasonably
could lead to other matter that could bear on, any issue that is or may be in the case.” In re Williams-
Sonoma, Inc., 947 F.3d 535, 539 (9th Cir. 2020) (quoting Oppenheimer Fund, Inc. v. Sanders, 437
U.S. 340, 350-51 (1978)); see also In re Facebook, Inc. Consumer Privacy User Profile Litig., No.
18-MD-2843 VC (JSC), 2021 WL 10282215, at *4 (N.D. Cal. Sept. 29, 2021) (“Courts generally
recognize that relevancy for purposes of discovery is broader than relevancy for purposes of trial.”)
(alteration omitted).
While the scope of relevance is broad, discovery is not unlimited. ATS Prods., Inc. v.
Champion Fiberglass, Inc., 309 F.R.D. 527, 531 (N.D. Cal. 2015) (“Relevancy, for the purposes of
discovery, is defined broadly, although it is not without ultimate and necessary boundaries.”).
Information, even if relevant, must be “proportional to the needs of the case” to fall within the scope
of permissible discovery. Fed. R. Civ. P. 26(b)(1). The 2015 amendments to Rule 26(b)(1)
emphasize the need to impose reasonable limits on discovery through increased reliance on the
commonsense concept of proportionality: “The objective is to guard against redundant or
disproportionate discovery by giving the court authority to reduce the amount of discovery that may
be directed to matters that are otherwise proper subjects of inquiry. The [proportionality
requirement] is intended to encourage judges to be more aggressive in identifying and discouraging
discovery overuse.” Fed. R. Civ. P. 26 advisory committee’s note to 2015 amendment. In evaluating
the proportionality of a discovery request, the Court considers “the importance of the issues at stake
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 2 of 11

3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
in the action, the amount in controversy, the parties’ relative access to the information, the parties’
resources, the importance of the discovery in resolving the issues, and whether the burden or expense
of the proposed discovery outweighs its likely benefit.” Fed. R. Civ. P. 26(b)(1).
With regard to proportionality, “[t]he parties and the court have a collective responsibility
to consider the proportionality of all discovery and consider it in resolving discovery disputes.” Fed.
R. Civ. P. 26 advisory committee’s note to 2015 amendment. “A party claiming undue burden or
expense ordinarily has far better information—perhaps the only information—with respect to that
part of the determination. A party claiming that a request is important to resolve the issues should
be able to explain the ways in which the underlying information bears on the issues as that party
understands them. The court’s responsibility, using all the information provided by the parties, is to
consider these and all the other factors in reaching a case-specific determination of the appropriate
scope of discovery.” Id.
The Court has broad discretion and authority to manage discovery. U.S. Fidelity & Guar.
Co. v. Lee Inv. LLC, 641 F.3d 1126, 1136 n.10 (9th Cir. 2011) (“District courts have wide latitude
in controlling discovery, and their rulings will not be overturned in the absence of a clear abuse of
discretion.”). The Court’s discretion extends to crafting discovery orders that may expand, limit, or
differ from the relief requested. See Crawford-El v. Britton, 523 U.S. 574, 598 (1998) (holding trial
courts have “broad discretion to tailor discovery narrowly and to dictate the sequence of discovery”).
For example, the Court may limit the scope of any discovery method if it determines that “the
discovery sought is unreasonably cumulative or duplicative, or can be obtained from some other
source that is more convenient, less burdensome, or less expensive.” Fed. R. Civ. P. 26(b)(2)(C)(i).
As part of its inherent discretion and authority, the Court has broad discretion in determining
relevancy for discovery purposes. Surfvivor Media, Inc. v. Survivor Prods., 406 F.3d 625, 635 (9th
Cir. 2005) (citing Hallett v. Morgan, 296 F.3d 732, 751 (9th Cir. 2002)). Similarly, a district court’s
determination as to proportionality of discovery is within the district court’s discretion. See Jones
v. Riot Hospitality Group LLC, 95 F.4th 730, 737-38 (9th Cir. 2024) (finding district court did not
abuse discretion on proportionality ruling). Ultimately, “the timing, sequencing and proportionality
of discovery is left to the discretion of the Court.” Toro v. Centene Corp., 202 WL 6108643 at *1
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 3 of 11

4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
(N.D. Cal. Oct. 14, 2020).
DISCUSSION
The dispute here centers on certain documents produced by Meta during discovery for which
the metadata indicated there was no custodian for each such document (so-called “non-custodial
documents”). Under the ESI Protocol entered in this case, a Party producing documents is required
to produce certain specified “metadata and coding fields” for those documents, including the
metadata field “ALLCUSTODIANS”. 	[Dkt. 690 at 21-22]. 	For these non-custodial documents,
Meta identified “Meta Platforms” in the “custodian” metadata field to reflect Meta’s assertion that
there were no individuals who were custodians of each of these documents. [Dkt. 3023 at 6].
The instant dispute has been narrowed to 300 Meta documents which are apparently on the
States’ trial exhibit list and which the States refer to as “trial critical”. [Dkt. 3346 at 7]. Although
not all are specified by Bates numbering or exhibit numbers in the briefing, the Parties indicate these
300 trial exhibits were stored in Google Drive. 	According to the States, when Google exports
documents from Google Drive (such as happened here for the documents to be produced), Google’s
policies indicate that Google includes the “Collaborator” and “Viewer” metadata fields for such
exported Google Drive documents. [Dkt. 3023 at 7]. The ESI Protocol does not require or include
these metadata fields as among the listed metadata fields to be produced. The States now request
an Order requiring Meta to produce the “Collaborator” and “Viewer” metadata for the 300
documents in dispute. Id.; [Dkt. 3346 at 7].
RELEVANCE
As noted above, the burden is on the States to establish that the requested material is relevant
for purposes of discovery. As to relevance, the States argue that this additional metadata for these
300 trial exhibits is required because they intend to rely on such metadata to lay foundations as to
the authenticity of and admissibility of these trial exhibits. [Dkt. 3346 at 6]. The States argue that
Meta has refused (as of the date of the briefing) to stipulate to authenticity or admissibility of these
documents, and has reserved objections to such. 	Id. 	The States thus assert that this metadata is
required and relevant precisely because Meta will not admit to the authenticity or admissibility of
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 4 of 11

5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
these trial exhibits.
In opposition, Meta argues that the “Collaborator” and “Viewer” metadata fields merely
indicate whether or not someone had authorization to access a document when it was in Google
Drive. Id. at 5. Meta argues that “in many instances” this would consist of “lists of dozens, if not
hundreds, of employees staffing entire teams, pillars, and platforms at Meta—a list too sweeping to
provide any real benefit to the AGs in establishing foundation at trial.” Id. Further, Meta argues
that “nothing about authorization to open a document substantiates anything resembling personal
knowledge” and thus “the requested metadata would not give the AGs the information needed to
lay the foundation to admit documents through a witness at trial.” Id. On this basis, Meta argues
that the States have failed to show that the requested metadata would likely lead to relevant evidence.
Id.
As this Court has advised the Parties throughout this MDL, the standard for relevance for
purposes of discovery is not the same as the standard for relevance at trial. See, e.g., In re Social
Media Adolescent Addiction/Personal Injury Prods. Liab. Litig., 2026 WL 1692218, *7 (N.D. Cal.
June 10, 2026) (finding that “documents are within the scope of relevance for purposes of discovery
(but [the Court] makes no finding as to relevance for purposes of trial or dispositive motions”); see
also C&C Jewelry Mfg, Inc. v. West, 201 WL 3943673, *1 (N.D. Cal. Oct. 7, 2010) (“For discovery
purposes, however, relevance is construed more broadly for discovery purposes than for trial, and
information is discoverable even if it might not itself be admissible.”). While lists of hundreds of
persons who may have been authorized to access a particular document is of limited relevance for
purposes of discovery, Meta provides no indication as to how many of the subject documents would
have metadata resulting in such extensive lists. 	However, Meta admits that there are documents
“where fewer individuals had access”, [Dkt. 3346 at 5], and the relevance of such information is
heightened precisely because the universe of people who worked with or accessed such documents
while at Meta is more constrained and identifiable.
Fundamentally, Meta’s objections to relevance are directed to whether the metadata would
be relevant and useful (or even successful) at trial in laying the foundation necessary for
admissibility, and such objections are not persuasive with regard to the standards for discovery. See
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 5 of 11

6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
Fed. R. Civ. P. 26(b)(1) (“Information within this scope of discovery need not be admissible in
evidence to be discoverable.”). The undersigned makes no ruling or finding on admissibility issues
for purposes of trial. Food & Watter Watch, Inc. v. U.S. EPA, 2024 WL 4035924, *8 (N.D. Cal.
Jan. 18, 2024) (“Of course, relevance for discovery purposes is not the same as admissibility at
trial.”). Rather, the record shows that the requested metadata for the documents at issue would be
within the scope of relevance for purposes of discovery. Because Meta (as the Party in possession
of the documents and metadata) does not separately identify those documents for which the metadata
would be “extensive lists” of hundreds of people (to substantiate the claim that the relevance is
diminished) versus a more limited number of people, there is no basis to differentiate the documents
and as such the record shows that the metadata requested here would disclose information that is
within the scope of relevance for purposes of discovery.
Proportionality
Most of the Parties’ briefing focuses on the timeliness of the States’ motion to compel and
the availability to the States of other information and means to lay foundation or otherwise
authenticate these exhibits for trial.
As to timeliness, Meta relies on the fact that the ESI Protocol in this case was entered on
March 18, 2024 as the result of the Parties’ Stipulation. [Dkts. 689, 690]. The context here is further
instructive. Prior to the entry of the ESI Protocol, the undersigned was presented with and resolved
multiple letter briefs on disputes between the Parties on the terms of the ESI Protocol. See Dkt. 623
(Order resolving ESI Protocol Disputes raised in Dkts. 352 and 534). The terms of the ESI Protocol
were discussed at multiple hearings and Discovery Management Conferences the Court held with
the Parties early in the discovery process, given the number of disputes raised at that time (which
even necessitated the Court developing a bespoke chart for the Parties to organize their dozens of
disputes more clearly). 	See, e.g., Dkt. 503 (Discovery Management Order No. 1 at Section V).
Suffice to say that the selection of metadata fields which are required to be produced pursuant to the
ESI Protocol was the product of extensive, reasoned discussions between the Parties (in their
negotiations) and with the Court (in resolving the disputes). 	The fact that the metadata now
requested is not part of the ESI Protocol is neither an oversight nor a drafting error.
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 6 of 11

7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
The States argue at length that the ESI Protocol’s requirement to produce the “custodian”
metadata field must be interpreted to include people who were authorized to access a document
under the “Collaborator” and “Viewer” metadata fields. Meta is correct that the metadata fields for
“Collaborator” and “Viewer” are not necessarily custodians of the document. The fact that a person
could view a document does not equate their being a custodian of the document. The States cite no
case law interpreting “custodian” so broadly and the case law they cite refers to “custodians” as
people who may have authored, edited or accessed a document (not the other way around). [Dkt.
3346 at 6 (citing IQVIA Inv. V. Veeva Sys., 2019 WL 3069203, *1-5 (D. N.J. 2019)]. 	While
custodians may author, edit, or access a document, not every person who accessed a document is a
custodian of that document. Therefore, to the extent the States argue that they are not seeking new
or additional metadata that is not already required by the existing ESI Protocol (on the grounds that
the term “custodian” in the ESI Protocol should be broadly construed), that argument is
unpersuasive. This is not “cleanup” discovery in the sense that the Parties have been engaging in
supplemental voluntary productions of documents by agreement or pursuant to Court Order since
the close of the fact discovery period.
The issue then is whether the request for this new metadata is proportional to the needs of
the case at this time. Proportionality is evaluated by “considering the importance of the issues at
stake in the action, the amount in controversy, the parties’ relative access to relevant information,
the parties’ resources, the importance of the discovery in resolving the issues, and whether the
burden or expense of the proposed discovery outweighs its likely benefit.” 	See Fed. R. Civ. P.
26(b)(1). 	In briefing, the Parties focus on the importance of the discovery and relative access to
relevant information. 	That is, neither Party argues that the issues at stake in the litigation or the
amount in controversy impact the proportionality analysis here.
Further, Meta does not argue that there is undue burden or expense in producing the
requested metadata (other than a broadly phrased argument that producing the metadata now is too
late, since fact discovery closed long ago and trial is imminent). Certainly, Meta never expressly
denies the States’ repeated suppositions that Meta probably already has the requested metadata in
its possession, in light of Google’s policies about exported Google Vault documents. 	And Meta
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 7 of 11

8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
does not submit any declarations or other evidence to try to establish that producing the metadata
would impose burden or expense.
The States argue that the requested metadata is important for resolution of the issues because,
absent the metadata, there will be challenges in authenticating exhibits for admissibility at trial.
Meta argues that the requested metadata will not establish foundation, especially where witnesses
have no recollection of any particular document at issue. 	Meta points out that the States do not
identify any specific documents by Bates number or exhibit number for which the alleged challenges
exist in laying foundation and for which the requested metadata would cure the admissibility issues.
In that regard, Meta points out that only 49 of the 285 documents on the State AG’s Priority
Exhibit List identify “Meta Platforms” as the sole custodian. [Dkt. 3346 at 5]. The States do not
dispute this. That is, while the States’ motion seeks the requested metadata as to 300 trial exhibits
which are allegedly non-custodial, the record indicates that there are 236 exhibits which identify
custodians in the metadata. And, of the 49 documents which only list “Meta Platforms” as the sole
custodian, Meta identifies six exhibits which bear the names of Meta employees on their face and
Meta argues that some of the exhibits have the names of Meta employees in their exhibit
descriptions. 	Id. (identifying Exhibits 1782, 1755, 1253, 598, 417, and 308). 	The States do not
dispute that these six exhibits identify Meta employees on their face. That is, the record indicates
there are 43 trial exhibits for which the metadata identifies only “Meta Platforms” as the custodian
and for which the contents also do not identify any Meta employees on the face of the documents.
This 	also 	impacts 	the 	factor 	of 	access 	to 	relevant 	information, 	for 	purposes 	of
proportionality. The States do not deny that there exists a subset of the 285 trial exhibits from which
the States may ascertain the author of the documents just from the contents of the documents
themselves. Id. at 6. In response, the States argue that even knowing the author of a document from
the face of the document is insufficient because, as the States speculate, “the ‘author’ may have less
or different knowledge of the documents than others who edited or reviewed the document.” Id.
(emphasis added).
Timeliness of a discovery request is not merely a threshold issue (as argued by Meta and
discussed above) but also is a factor in the proportionality analysis. See PersonalWeb Techs. LLC
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 8 of 11

9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
v. Int’l Bus. Machines Corp., 2017 WL 11673582, *2 (N.D. Cal. June 16, 2017) (finding discovery
request was not proportional because of, in part, “the advanced stage of the case and the untimeliness
of PersonalWeb’s motion” where “PersonalWeb is attempting to discover new information shortly
before trial, so ‘flexible treatment’ is less warranted.”); see also Sierra v. Covidien LP, 2025 WL
2463800, *2 (D. Mass. Jan. 27, 2025) (“The stage of the case bears upon whether the requested
discovery is proportional to the needs of the case under Rule 26(b)(1).”) (collecting cases).
Fundamentally, Meta argues that the untimeliness of the demand for the additional metadata renders
the request non-proportional. The States argue that, in December 2024, they requested additional
“custodian” data for the non-custodial documents and that the request was therefore timely made.
[Dkt. 3023 at 8-9]. The problem with this argument is that, while the request may have been made
in 2024, the instant motion was filed more than a year later and the request for immediate relief was
raised with the undersigned only recently. 	Meta argues that the States chose not to pursue
depositions of witnesses whose names appear on the documents. Meta argues that the failure to take
depositions of witnesses to authenticate these trial critical documents was a tactical decision and the
States should bear the burden of their tactical decisions.
The Court also notes that, at the earliest Discovery Management Conferences where the
Parties discussed setting discovery limits, the Court explicitly raised with the Parties that they could
use Requests for Admissions for purposes of authentication of documents and that there would not
be limits imposed for that purpose. This is reflected in the resulting Stipulation and Order Governing
Discovery Limitations, entered on March 7, 2024 which states “[t]he limitations on RFAs set forth
in this Order do not include any RFAs that may be needed for purposes of authentication (to the
extent the Parties are not able to stipulate to the authentication of documents through other means).”
[Dkt. 672 at 2 n.2]. The States have not provided a reasonable explanation for their failure to employ
the available discovery tools made explicitly available to them since 2024 to obtain discovery which
would have obviated the need for the instant motion.
Finally, as discussed above, Meta disputes whether the requested metadata would be
sufficient to authenticate the documents at issue, and that the failure of the metadata to cure the
authentication issues renders the requested metadata not merely outside the scope of relevance for
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 9 of 11

10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
discovery but also renders the request non-proportional to the needs of the case. The States argue
that the request for the metadata is proportional because, without the metadata, the States would
have to rely on the “fallible” memories of witnesses for admissibility. [Dkt. 3346 at 6-7].
As discussed above, while the States seek relief as to 300 trial exhibits, the record indicates
that there are in fact 236 exhibits which identify custodians (and thus only 49 exhibits which list
only “Meta Platforms” as the custodian). Id. at 5. And, as detailed above, six exhibits of these 49
exhibits have the names of Meta employees on their face. 	Id. (identifying Exhibits 1782, 1755,
1253, 598, 417, and 308). Based on the record presented, there are only 43 trial exhibits which are
truly non-custodial (for which the metadata identifies only “Meta Platforms” as the custodian) and
for which the contents also do not identify any Meta employees on the face of those documents.
Accordingly, taking into consideration the factors and arguments of the Parties as to proportionality,
the Court FINDS that requiring Meta to produce the requested additional metadata for these 43 trial
exhibits is within the scope of proportionality for purposes of discovery. Fundamentally, the Court
finds that, as to these 43 exhibits, there is no readily available other information for identifying
persons authorized to access the documents and, despite the States’ failure to take steps earlier in
the case, the request as to these documents is proportional to the needs of the case. As a corollary,
the Court FINDS that requiring Meta to produce the requested additional metadata for all the
remaining 242 or so trial exhibits is not proportional because either the metadata produced already
identifies other custodian(s) for each such document or the face of the document identifies Meta
employees (or both).
CONCLUSION
For all the reasons discussed herein, the Court GRANTS-IN-PART the States’ motion to
compel Meta to produce the requested “Collaborator” and “Viewer” metadata for the 43 trial
exhibits which only identify “Meta Platforms” as the custodian in the previously produced metadata
and which also lack identification of any Meta employees in the contents of such documents. Given
the limited number of documents involved, the Court ORDERS Meta to produce this additional
metadata on a daily rolling basis starting on August 11, 2026 and complete the production promptly
(and no later than by close of business on August 13, 2026). The Court ORDERS Meta to produce
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 10 of 11

11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
United States District Court
Northern District of California
this additional metadata as metadata overlays (consistent with Meta’s practice in producing
supplemental metadata previously), but the Parties are free to reach agreement on alternate formats
for this production.
Further, the Court ORDERS the Parties to file a Joint Status Report (no longer than two
pages evenly divided between the Parties) on August 14, 2026 reporting on the completion of this
production of the additional metadata.
IT IS SO ORDERED.
Dated: August 10, 2026
______________________________________
PETER H. KANG
United States Magistrate Judge
Case 4:22-md-03047-YGR Document 3361 Filed 08/10/26 Page 11 of 11