Electronic discovery in IP litigation runs on the same federal rules as any other civil case, but the material at issue — source code, engineering files, trade secrets handed to a competitor — forces procedures that ordinary discovery never touches: attorneys-only confidentiality tiers, standalone source code review computers kept with an escrow agent, patent prosecution bars on the lawyers who look at the code, and proportionality fights driven by data volumes that can reach terabytes. The sequence runs from preservation through production, and a misstep at any stage can mean sanctions, lost evidence, or waived privilege.
What Counts as Electronic Evidence in an IP Case
The range of digital material at issue is wider than in most civil litigation, and the technical complexity is higher. Software patent disputes often require production of source code in languages like C++ or Python, which functions as the blueprint for the patented technology. Hardware and consumer product cases pull in Computer-Aided Design files and manufacturing schematics, so the two products can be compared side by side; small differences in geometry or tolerances can decide the case.
Trademark cases lean on marketing material. Internal brand style guides and advertising plans show how a company presented its identity to the public, which matters when the dispute turns on consumer confusion. Sales spreadsheets and customer databases quantify damages by establishing how many units moved during the infringement period. And across every type of IP case, internal communications on email, Slack, and Microsoft Teams often reveal whether a company knowingly copied a competitor or developed its product independently.
Auto-deleting messages have become a flashpoint. Many companies use ephemeral messaging platforms where conversations disappear after a set period, and courts have made clear that the duty to preserve relevant data extends to those platforms. If engineers or product managers discuss development on tools with auto-delete features, policies need to capture that data before litigation is on the horizon.
When the Duty To Preserve Begins
The obligation to preserve electronically stored information attaches when a party reasonably anticipates litigation. The standard is objective: if a reasonable person in the same position would have expected a lawsuit, the duty is triggered whether or not a complaint has been filed. For a defendant, receiving a cease-and-desist letter from a patent holder or a trademark demand letter almost certainly starts the clock. For a plaintiff, the trigger can come earlier, often the moment the company consults an attorney about a potential claim.
Once the duty attaches, the company issues a litigation hold, a written directive instructing employees to stop deleting, overwriting, or altering potentially relevant data. The hold should identify the categories of documents at risk and name the custodians most likely to have relevant files. In IP cases, those custodians typically include lead engineers, product managers, patent liaisons, and marketing directors who worked on the disputed product.
A thorough data map matters as much as the hold notice itself. The map identifies every location where relevant files might live: on-premises servers, cloud storage accounts like AWS or Google Drive, individual workstations, and backup tapes. Chain-of-custody records should log who accessed files, when they were copied, and where they were stored, so the opposing party cannot challenge authenticity at trial.
Personal devices complicate things. Engineers and developers often use personal phones, tablets, or home computers for work-related communications. Whether the company must produce data from those devices depends on how the court defines “control” under Rule 34, which allows discovery of information in a party’s possession, custody, or control.1Legal Information Institute. Federal Rules of Civil Procedure Rule 34 – Producing Documents, Electronically Stored Information, and Tangible Things Most federal courts apply a “legal right” test; a minority, including some in the Second Circuit, use a broader “practical ability” test that asks whether the company could realistically obtain the data regardless of who owns the device.
What Happens When Evidence Disappears
When ESI that should have been preserved is lost because a party failed to take reasonable steps, and the data cannot be restored through other means, Rule 37(e) gives the court two tiers of response.2Legal Information Institute. Federal Rules of Civil Procedure Rule 37 – Failure to Make Disclosures or to Cooperate in Discovery; Sanctions If the opposing party was prejudiced, the court can order remedial measures proportional to the harm, such as allowing additional discovery or precluding certain arguments. The heavier sanctions are reserved for intentional destruction: on a finding that a party deliberately destroyed evidence to prevent its use, the court may instruct the jury to presume the lost data was unfavorable, dismiss the case, or enter default judgment. The intent requirement for that second tier is a meaningful threshold, and courts scrutinize whether the destruction was targeted or the result of routine data management.
Planning Discovery Under Rule 26(f)
Before documents change hands, the federal rules require the parties to meet and confer about how discovery will work. The Rule 26(f) conference is where the practical details get hammered out. The parties must develop a discovery plan that addresses the disclosure, preservation, and production of ESI, including the format in which it will be produced.3Legal Information Institute. Federal Rules of Civil Procedure Rule 26 – Duty to Disclose; General Provisions Governing Discovery
IP cases load extra weight onto this conference. The parties need to discuss the technical architecture of their data systems so both sides understand what exists and how hard it will be to retrieve. In a software patent case, that means version control systems, code repositories, and build environments. In a trademark case, it may mean identifying which marketing platforms and customer relationship management tools hold relevant data. The plan should also address a privilege-protection agreement, which becomes critical given the volume of communications between IP counsel and clients during product development. Advisory committee notes to the rule tell counsel to become familiar with their client’s computer systems before the conference so the resulting plan reflects actual capabilities and limitations.
Protective Orders, AEO Tiers, and the Prosecution Bar
IP discovery routinely forces a company to hand its most competitively sensitive information to a direct rival. Rule 26(c) allows the court to issue protective orders that restrict how confidential material is handled, with an explicit provision for shielding trade secrets and confidential commercial information.3Legal Information Institute. Federal Rules of Civil Procedure Rule 26 – Duty to Disclose; General Provisions Governing Discovery These orders define which documents qualify as confidential, who can access them, and what happens if someone violates the restrictions.
The most restrictive tier is typically labeled “Attorneys’ Eyes Only.” Only outside counsel and pre-approved independent experts may view AEO material; employees of the opposing company, including executives, engineers, and in-house counsel, are barred. The point is to let the lawsuit proceed without giving a competitor a free look at proprietary technology or financial data.
Patent cases add the prosecution bar. Attorneys and experts who review confidential technical material under a protective order are often prohibited from prosecuting patent applications in the same technology area for a specified period after the case ends. This prevents someone who has reviewed a competitor’s source code from using that knowledge, even subconsciously, when drafting new patent claims. The length and scope vary by case, but the bar is a standard feature of patent protective orders, and it has to be factored in when deciding who on the team will access restricted materials.
How Source Code Actually Gets Reviewed
Source code is the single most sensitive category of evidence in software patent litigation. It represents years of development and often constitutes the company’s core trade secret. Courts have developed detailed inspection protocols to balance meaningful review against the risk of misappropriation. The District of Delaware, one of the busiest patent courts in the country, maintains a default standard that has become influential across federal courts.
Under that standard, the producing party loads the source code onto a standalone, password-protected computer that is not connected to any network.4United States District Court for the District of Delaware. Default Standard for Access to Source Code The computer sits with an independent escrow agent, with the costs shared by the parties. Access is limited to two outside counsel and two pre-approved experts for the requesting party, all covered by the existing protective order. The code cannot be printed or copied without the producing party’s consent or a court order.
The producing party must also supply a manifest listing every source and executable file on the computer, including file name, location, and an MD5 checksum that can later verify the files were not altered.4United States District Court for the District of Delaware. Default Standard for Access to Source Code The computer must include utilities that allow counsel and experts to view, search, and line-number source files, run text-pattern searches across multiple files, compare two files and display differences, and compute checksums. If the dispute involves whether the code actually compiles into the accused product, the court may require the producing party to include the build scripts and compilers needed to rebuild the application from source.
These restrictions make review genuinely difficult. The code cannot leave the escrow facility, cannot be run through outside analysis tools, and cannot be casually leafed through. Every session happens on site, and every note taken is subject to scrutiny. That is the tradeoff for getting access to someone else’s most valuable asset.
Proportionality and Cost-Shifting
IP cases generate enormous volumes of data, and the cost of collecting, processing, and reviewing that data can dwarf the legal fees. The federal rules impose a proportionality check: discovery must be relevant to a claim or defense and proportional to the needs of the case. Rule 26(b)(1) directs courts to weigh six factors when deciding whether a request goes too far:
- The importance of the issues at stake in the action.
- The amount in controversy.
- The parties’ relative access to the relevant information.
- The parties’ resources.
- The importance of the discovery in resolving the issues.
- Whether the burden or expense of the proposed discovery outweighs its likely benefit.
The default is that each side pays for its own production. But when ESI sits in sources that are “not reasonably accessible” because of undue burden or cost, the court can shift some or all of the expense to the requesting party.3Legal Information Institute. Federal Rules of Civil Procedure Rule 26 – Duty to Disclose; General Provisions Governing Discovery Legacy backup tapes and decommissioned systems are classic examples. The producing party bears the burden of proving the data is genuinely inaccessible, and courts typically require concrete evidence like vendor invoices or technical declarations rather than vague assertions.
These fights come up constantly in IP cases. A patent holder suing a large technology company might demand every version of a codebase going back a decade, spanning terabytes across multiple repositories. The producing party will argue that retrieving archived versions from legacy systems is disproportionately expensive relative to the incremental value of older code. Judges have wide discretion, and the outcome often depends on how well each side documents its costs and explains why the data matters.
Production, Review, and Privilege Protection
Formats and Metadata
The ESI protocol dictates how files are actually exchanged. Source code is almost always produced in native format so it can be functionally tested, compiled, and compared line by line. Emails and office files are often converted to static TIFF or PDF images that can be stamped with identification numbers for use at depositions and trial. The protocol also specifies how metadata is handled. Metadata includes hidden information like the date a file was created, when it was last modified, and who modified it. In IP cases, that timeline can be dispositive, showing whether a design was created before or after the patent’s priority date.
Technology-Assisted Review
When a case involves hundreds of thousands or millions of documents, manual attorney review is impractical. Technology-assisted review uses machine learning to classify documents as relevant or irrelevant based on a training set reviewed by subject-matter attorneys. The approach was first approved by a federal court in 2012 and has since become widely accepted as a legitimate way to satisfy the “reasonable inquiry” standard under Rule 26(g).5Federal Judicial Center. Technology-Assisted Review for Discovery Requests Courts evaluate the results using two statistical metrics: precision, which measures how many documents flagged as relevant actually are, and recall, which measures what fraction of the truly relevant documents the system found. Neither the rules nor the case law prescribes a specific threshold, so parties often negotiate acceptable rates as part of the ESI protocol. TAR works well in IP cases because the relevant documents often share technical vocabulary and conceptual patterns the algorithm can detect.
Clawback Agreements and Privilege Logs
Large-scale production inevitably raises the risk of accidentally disclosing privileged attorney-client communications. Rule 502(d) of the Federal Rules of Evidence allows the court to order that any inadvertent disclosure in the litigation does not waive privilege, in that case or in any other federal or state proceeding.6Legal Information Institute. Federal Rules of Evidence Rule 502 – Attorney-Client Privilege and Work Product; Limitations on Waiver Parties negotiate a clawback agreement and ask the court to enter it as an order. If a privileged document slips through, the producing party can claw it back without a permanent waiver.
The value of a 502(d) order is hard to overstate. Without one, a single missed privileged document in a two-million-file production could be treated as a waiver, potentially opening the door to every communication on that subject. With the order in place, the producing party can run a faster first-pass review knowing that honest mistakes are recoverable. Most eDiscovery practitioners start here when building a review workflow for an IP case.
Parties must also maintain a privilege log that describes each withheld document in enough detail for the opposing side to evaluate the claim, without revealing the privileged content.3Legal Information Institute. Federal Rules of Civil Procedure Rule 26 – Duty to Disclose; General Provisions Governing Discovery In IP cases, privilege logs can run to thousands of entries because patent prosecution generates heavy attorney-client correspondence. Categorical logging, which groups similar documents rather than listing them individually, is accepted in many courts as a way to manage the volume.
Forensic Experts and What They Cost
Digital forensic specialists do work in IP cases that goes well beyond standard document collection. In trade secret misappropriation disputes, they trace the movement of proprietary files, identifying whether a departing employee copied data to a USB drive, uploaded files to a personal cloud account, or forwarded documents to a personal email address before resigning. That forensic trail is often the strongest evidence of misappropriation.
In patent cases, forensic experts perform detailed code comparisons to identify structural similarities between the patented technology and the accused product. This work sometimes involves a “clean room” analysis, where the expert works in isolation from any proprietary material to show that similarities arose from independent development rather than copying. The clean room approach is designed to produce evidence a court will view as untainted by access to the opposing party’s trade secrets.
The federal rules draw a sharp line between consulting experts and testifying experts. Consulting experts work behind the scenes and their analysis is generally shielded as attorney work product. Testifying experts must submit a written report containing a complete statement of their opinions, the basis for those opinions, the data they considered, their qualifications, their compensation, and a list of cases in which they have testified over the previous four years.3Legal Information Institute. Federal Rules of Civil Procedure Rule 26 – Duty to Disclose; General Provisions Governing Discovery Testifying experts can also be deposed, so everything in the report is fair game for cross-examination.
Hourly rates for digital forensic specialists and expert witnesses in IP cases typically run from $200 to $550 or more, depending on specialty and complexity. Beyond expert fees, the broader eDiscovery costs add up quickly. Industry surveys indicate that hosting fees commonly run under $10 to $20 per gigabyte per month, while processing fees range from under $25 to $100 or more per gigabyte depending on the vendor and the data format. In a case involving terabytes of source code, design files, and communications, total eDiscovery spend can easily reach six or seven figures before the first deposition.
The ITC Timeline Is Different
Not all IP disputes play out in federal district court. Patent holders who want to block infringing imports can file a complaint under Section 337 at the International Trade Commission, where discovery follows the same basic structure but on a compressed timeline.7Congressional Research Service. An Introduction to Section 337 Intellectual Property Litigation at the International Trade Commission The ITC must resolve investigations “at the earliest practicable time,” and the administrative law judge sets a target date for final determination that cannot exceed 16 months without the commissioners’ approval. Preservation, collection, and production all move faster than in district court, and there is far less room to renegotiate ESI protocols or extend deadlines. Companies facing an ITC investigation need their eDiscovery infrastructure in place before the complaint is filed.