Federal Rule of Civil Procedure 26(c): Protective Orders

A protective order under Federal Rule of Civil Procedure 26(c) is a court order that limits or blocks discovery when disclosure would cause real harm. Any party, or any non-party hit with a subpoena, can ask for one. The court will grant it only after the moving side shows “good cause” and certifies that it first tried to work the dispute out directly with the other side. Once entered, the order can do anything from sealing a deposition transcript to walling off trade secrets behind an attorneys’-eyes-only designation.

Who Can File, and in Which Court

Rule 26(c) is not limited to the named parties. It extends standing to “any person from whom discovery is sought,” so a non-party served with a subpoena for documents or a deposition can move for protection on the same terms as a plaintiff or defendant.¹ That matters in practice, because subpoenas regularly reach employers, banks, and other third parties with their own confidentiality interests.

The motion is normally filed in the court where the lawsuit is pending. When the dispute concerns a deposition, though, the rule gives an alternative venue: the court for the district where the deposition will take place.¹ The alternative exists so a non-party witness in one state is not forced to litigate a protective order in a distant forum where the underlying case sits.

You Have to Meet and Confer First

Going straight to court is not an option. Before filing, you must try to resolve the dispute directly with the other affected parties. The motion itself has to include a written certification stating that you conferred in good faith, or at least attempted to confer, with the opposing side before involving the judge.¹

This is not a formality. Courts routinely deny motions when the certification is missing or when the effort was clearly a box-checking exercise. A single email saying “we object” does not cut it. The expectation is a genuine discussion about narrowing the dispute, whether that means limiting who sees a document, redacting specific fields, or agreeing on a timeline that lowers the burden.

What Good Cause Actually Means

Good cause is the substantive test, and it is where most motions succeed or fail. The court will not grant protection based on a vague claim that information is sensitive or private. You have to show, with specifics, that disclosure of the particular information at issue would cause one of the harms the rule recognizes: annoyance, embarrassment, oppression, or undue burden or expense.¹

“These documents are confidential” accomplishes nothing. “These documents contain our proprietary pricing algorithms, and the requesting party is a direct competitor whose designated expert formerly consulted for us” tells the court exactly what the harm is and why it is real. The burden sits on the party seeking protection, and the showing must be particularized.

The court has wide discretion in weighing that showing. A judge can consider the relevance of the information to the claims, the availability of the same information from less sensitive sources, and whether a less restrictive form of protection could address the concern. If information is only marginally relevant and the potential harm from disclosure is significant, the balance tips toward protection. If it goes to the heart of the case, even serious confidentiality concerns may not be enough to block discovery outright.

What the Court Can Order

Rule 26(c) gives courts a menu of eight specific remedies, and the court can combine them. The options run from a complete ban on discovery to fine-tuned conditions on how information is shared:

Forbidding the disclosure or discovery altogether. Courts reserve this for information that is highly sensitive and only marginally relevant.
Setting the terms, including the time and place, and allocating costs. This is the most common form of relief.
Requiring a different discovery method than the one chosen.
Limiting the scope of discovery to certain matters or forbidding inquiry into certain topics.
Designating who may be present during a deposition or inspection.
Sealing a deposition transcript, to be opened only by court order.
Shielding trade secrets or other confidential research, development, or commercial information, or requiring that it be disclosed only in a specified way.
Requiring specified documents to be filed in sealed envelopes, opened only as the court directs.

The list is illustrative rather than exhaustive. Courts have inherent authority to craft protections beyond the listed categories, and the rule’s phrasing signals as much.¹

Trade secrets get their own subsection, but they do not get automatic immunity. Courts weigh the holder’s privacy interest against the requesting party’s need for the information, and the usual result is limited protection rather than a complete bar.¹ In practice, trade secret protective orders create tiered access. Materials designated “confidential” are viewable by the parties and their counsel. The most sensitive materials are designated “attorneys’ eyes only,” meaning outside counsel and retained experts can see them, but the parties themselves cannot. That lets the litigation proceed on a full record while keeping proprietary formulas, source code, or customer lists out of the hands of a competitor-party.

Stipulated and Blanket Orders

Not every protective order comes out of a contested motion. In many cases, the parties negotiate an order and submit it for the judge’s approval. These stipulated orders are common in commercial litigation where both sides know at the outset that they will exchange confidential business information. Rather than fighting document by document, they agree on a framework: what counts as confidential, who can see it, and what happens at the end of the case.²

A stipulated order is not enforceable just because both sides signed it. It becomes a court order only when the judge issues it, and the court must still find that good cause supports the protection. Courts have refused to endorse stipulated orders where the parties failed to show any basis beyond mutual convenience.²

In complex litigation with massive productions, courts often enter blanket or umbrella protective orders. These let a producing party designate materials as confidential based on a good-faith belief that good cause exists, without a document-by-document ruling upfront. The trade-off is that blanket protection is provisional. If the other side challenges a specific designation, the producing party carries the burden of proving good cause for that particular document. Because blanket orders are inherently overinclusive, courts consider them especially open to later modification.²

Fees, Sanctions, and the Real Stakes

Rule 26(c) incorporates the fee-shifting provisions of Rule 37(a)(5). If the court denies a motion to compel (which functionally resembles granting the requested protection), the court must order the losing side to pay the opposing party’s reasonable expenses, including attorney’s fees, incurred in responding. The same principle runs in the other direction: if a motion to compel is granted and the protective order effectively denied, the party who unsuccessfully resisted discovery faces the same fee exposure. When a motion is granted in part and denied in part, the court can split the costs.³

There is a safety valve. The court will not award fees if the losing position was “substantially justified” or if other circumstances make an award unjust. The exception is narrower than it sounds. A motion that rested on reasonable grounds, even if ultimately unpersuasive, can avoid fees. A baseless motion filed to stall, or stonewalling of a legitimate request, will not.³

Violating a protective order carries far heavier consequences. Under Rule 37(b), the court can impose escalating sanctions on a party that disobeys a discovery order:

Deeming certain facts established in the opposing party’s favor.
Prohibiting the violating party from introducing specified evidence or from pursuing or defending particular claims.
Striking part or all of a pleading.
Staying the proceedings until the order is obeyed.
Dismissing the case or entering default judgment against the violator.
Treating the violation as contempt of court, which can carry fines or incarceration.

On top of any of those, the court must order the violating party, the attorney, or both to pay the reasonable expenses and attorney’s fees caused by the violation, unless the failure was substantially justified or the award would be unjust.³ Courts generally impose sanctions proportionally, starting lesser and escalating if noncompliance continues. Dismissal and default judgment are reserved for willful, bad-faith violations or repeated defiance.

Changing or Lifting an Order

Protective orders are not permanent. A party can move to modify or dissolve one, but the standard depends on the type of order and the reliance interests involved. Blanket orders, because of their inherent overinclusiveness, are more readily modified than narrowly tailored orders entered after a contested showing of good cause. A party that produced documents under a blanket order has a weaker reliance argument than one that disclosed trade secrets only because a specific, carefully negotiated order promised protection.²

Courts weighing modification consider factors like the confidentiality interests at stake, whether the requesting party’s need for the information has changed, whether public health or safety concerns favor disclosure, and whether the party that obtained the order would suffer real prejudice from modification.⁴ The analysis stays flexible and fact-specific. There is no rigid test, and the court retains broad discretion.

How to Challenge a Ruling You Lost

Discovery rulings, including decisions on protective orders, are generally not immediately appealable. Because they are interlocutory, the losing side normally has to wait until the case ends and raise the issue on appeal from the final judgment. That creates an obvious problem: if a court wrongly denies protection and forces disclosure of a trade secret, waiting until after trial means the harm is already done.

The primary workaround is a petition for a writ of mandamus, an extraordinary remedy asking a higher court to order the trial judge to change course. Mandamus is available only in narrow circumstances. The petitioner generally must show that the trial court committed a clear legal error, that waiting for a final appeal would cause irreparable harm that cannot be undone, and that no other adequate remedy exists. Disclosure of privileged or trade-secret information is one scenario where courts have recognized mandamus can be appropriate, precisely because wrongful disclosure cannot be unscrambled after the fact.

If protection is denied and the court orders discovery, the party facing disclosure must comply or risk sanctions. In rare cases, a party may choose to accept a contempt finding specifically to create a final, appealable order, but that is aggressive and risky, and not a step to take without careful counsel.

One Boundary Worth Knowing

Rule 26(c) governs unfiled discovery material exchanged between the parties. It does not govern what happens when a document actually gets filed with the court. Sealing documents in the court file is a separate question, and courts apply a much more demanding standard there, because a strong presumption of public access attaches to judicial records. Good cause is enough to protect materials in discovery; sealing filed documents typically requires compelling reasons.⁴

1
Legal Information Institute. Federal Rules of Civil Procedure Rule 26 – Duty to Disclose; General Provisions Governing Discovery
2
Federal Judicial Center. Confidential Discovery – A Pocket Guide on Protective Orders
3
Legal Information Institute. Federal Rules of Civil Procedure Rule 37 – Failure to Make Disclosures or to Cooperate in Discovery; Sanctions
4
United States Courts. Case Law on Entering Protective Orders, Entering Sealing Orders, and Modifying Protective Orders