How to build a claim chart with source references for Patent Infringement Matters

Build a patent infringement claim chart by preserving the issued claim language, separating each limitation, and mapping it to identified evidence about a specific accused product or process version. Each entry should distinguish what the source establishes from counsel’s inference, expert interpretation, and unresolved factual questions. The deliverable is a limitation-by-limitation chart with reproducible source references, a stated infringement theory, contrary evidence, and assigned follow-up—not merely quotations that resemble the patent’s terminology.

When to use this workflow

Use this workflow when litigation counsel must assess an infringement allegation, prepare or test infringement contentions, or identify evidence missing from a technical mapping. In-house teams can use it to coordinate product knowledge with outside counsel without treating engineering descriptions as legal conclusions.

The scope is US federal district-court litigation involving utility patents. Northern District of California Patent L.R. 3-1 supplies a district-specific example: it requires asserted claims, statutory infringement subsections, identified accused instrumentalities, and limitation charts, and addresses literal infringement, equivalents, and indirect infringement. Rule 1-3 allows court modification. These are not nationwide defaults; check the actual forum’s rules and orders. See the Patent Local Rules. (cand.uscourts.gov)

This workflow organizes evidence; it does not replace infringement or claim-construction analysis. Do not apply its federal procedural assumptions unchanged to ITC, PTAB, state, or arbitral proceedings.

What you need

Assemble a controlled working set:

• Patent materials: issued patent, asserted claim list, dependency chains, relevant prosecution materials, and applicable construction orders or disputed constructions.
• Accused implementation details: product names, model numbers, software releases, operating configurations, relevant periods, and differences among product families.
• Technical evidence: manuals, architecture documents, schematics, inspection records, test results, available source-code materials, and testimony.
• Source controls: document IDs, Bates numbers, native-file locations, code revisions, transcript pagination, confidentiality designations, and provenance.
• Counsel’s instructions: represented side, proposed theories, chart purpose, disclosure requirements, and restrictions on technical materials.

An early chart can identify gaps before production is complete. Keep “not yet evidenced” distinct from “not present in the accused implementation.” Record an evidence cutoff and the chart’s revision date so readers understand the assessment’s boundaries.

Use the template

The companion template is Patent Infringement Matters — Patent claim chart. You can copy this table into your working document or spreadsheet. Add a header identifying the patent, purpose, evidence cutoff, revision, reviewer, and confidentiality restrictions.

Use stable row IDs for comments and revisions. Exact limitation preserves issued wording; paraphrases belong in Mapping / basis. Record inherited limitations through explicit row references, ensuring the reader can reconstruct the complete dependent claim.

Source / pinpoint should identify a document and Bates page or figure, transcript page and lines, or controlled code revision, path, function, and line range. Distinguish printed pages from PDF pages. Links help navigation but should supplement, not replace, durable identifiers.

Label the basis as direct documentary support, testimony, inference, or expert interpretation. Theory / construction states the position being tested and its construction assumptions. Use Action / owner for the decision question, responsible reviewer, and verification status; these are working fields, not prescribed court formatting.

Step by step

1. Define the chart’s purpose and comparison units

Decide whether the chart is internal analysis, a draft contention disclosure, or a response testing an opponent’s mapping. Identify the patent, asserted claims, accused implementations, relevant operating configurations, and evidence cutoff. Avoid one undifferentiated entry for a product family whose releases may operate differently.

For each row, identify the implementation actually supported by the evidence. If representative treatment is proposed, document its factual basis and leave unsupported versions unresolved. Separate charts may be clearer where architectures differ substantially.

Template output: Populate row IDs, claim numbers, accused versions, and the theory evaluated. Record purpose and scope in the header.

2. Preserve claim wording and dependencies

Transcribe claims from the issued patent and check the transcription against that source. Segment the text into manageable analytical limitations without deleting qualifiers, relationships, sequence language, or antecedent references. Retain preamble language and flag questions about its treatment for counsel rather than dropping it during transcription.

For dependent claims, trace the entire dependency chain and identify inherited rows. Keep technical explanations outside the exact-language field. Put construction disputes beside the affected text, with a reference to the relevant order or proposed construction.

Template output: Enter exact text, dependency references, and construction questions. Identify the patent page or claim location in the chart header or source field.

3. Establish reproducible source references

Assign stable source identifiers and verify that the cited copy is the reviewed copy. For manuals, capture edition; for code, capture revision and location; for testing, identify the configuration and report section. A code excerpt should be read with enough surrounding context to assess conditions, calls, and execution paths.

Retain provenance for later foundation work. Federal Evidence Rule 901 requires sufficient evidence supporting a finding that an item is what its proponent claims; testimony from a knowledgeable witness is one listed method. Do not treat this authentication inquiry as a complete admissibility review. See Evidence Rule 901. (uscourts.gov)

Template output: Add exact pinpoints and assign unresolved source identity, version, or foundation questions to a reviewer.

4. Map technical operation, not matching words

Explain how the source’s component, operation, or relationship bears on the limitation. A manual mentioning a controller does not establish every function attributed to the claimed controller. Trace inputs, processing, outputs, and component locations where those details matter.

Separate source statements from the reasoning connecting them to the claim. Identify the person supplying an interpretation and its assumptions. For testimony, distinguish what the witness observed from what the witness understood from other records. Record disabled features, alternative configurations, and evidence limiting a document’s applicability.

Template output: Write a concise mapping, label its basis, and identify any unsupported inference or contrary evidence. Avoid an unqualified “met” where only part of the limitation is supported.

5. Separate theories and construction assumptions

Identify whether each mapping tests literal infringement or an equivalents position. State the claim meaning used. Where competing constructions change the comparison, create linked alternatives rather than silently selecting one. Do not use an equivalents label merely to conceal missing literal support; assign the necessary legal and technical analysis separately.

Keep indirect-infringement allegations distinct from the technical mapping. The Northern District of California’s Rule 3-1 separately addresses indirect infringement and literal-versus-equivalents positions, illustrating why a technical table alone may not complete the required disclosure. See Patent L.R. 3-1. (cand.uscourts.gov)

Template output: State the theory, construction assumption, and unresolved legal or factual dependency for affected rows.

6. Convert gaps into focused follow-up

Replace “need more evidence” with a question that could change the mapping: Did comparison occur locally or remotely? Which release introduced the function? Was the option enabled? Identify a likely document, test, or knowledgeable witness rather than requesting every engineering record.

Federal discovery under Rule 26(b)(1) concerns nonprivileged matter relevant to claims or defenses and proportional to the case; information within that scope need not be admissible to be discoverable. Counsel should apply that framework when evaluating requests, not assume every chart gap warrants production. See Civil Rule 26. (uscourts.gov)

Template output: Assign an action, owner, and decision question to each material gap. Preserve uncertainty when evidence remains unavailable.

7. Reproduce mappings and prepare the intended deliverable

Have a second reviewer reproduce material mappings from the cited records. Check claim wording, implementation version, source context, and reasoning. If a source supports only part of a limitation, narrow the statement and retain the unresolved portion.

Before external use, compare the chart with applicable rules, orders, disclosure instructions, and handling restrictions. Maintain a controlled internal version containing investigation notes and an separately approved external version where appropriate. Review newly obtained evidence against affected rows rather than overwriting prior reasoning without a revision record.

Template output: Record verification status and reviewer, correct defects, and resolve or expressly assign remaining issues before approving the intended deliverable.

Worked example

Fictional scenario: Alder Circuit Labs alleges that Beacon Systems’ BS-20 sensor gateway practices claim 1 of a fictional utility patent. Counsel is evaluating firmware 4.2. The following invented limitations are excerpts, not a complete claim. All parties, documents, identifiers, and technical facts are fictional.

The working set includes a gateway manual, architecture memo, and laboratory report. The manual describes receipt of sensor packets but does not identify firmware 4.2. The memo places threshold comparison in a cloud service without establishing release coverage. The laboratory report identifies a firmware 4.2 test unit and records an alert, but does not locate the comparison or establish its causal connection to transmission.

The chart redirects the investigation. Counsel should not use the alert test as proof of gateway-local comparison. C1-B contains potentially contrary architecture evidence; C1-C contains an incomplete causal inference. The priority is establishing where comparison occurred in firmware 4.2, not collecting more alert screenshots.

If the memo covers 4.2 and accurately describes the relevant configuration, its location evidence materially weakens the proposed literal mapping under the stated reading. If coverage remains unknown, retain that uncertainty. Neither possibility justifies inventing a local code path or resolving the complete infringement question from these excerpts.

Common mistakes

• Replacing claim language with paraphrases. Preserve issued wording; keep explanations separate so qualifiers remain visible.
• Citing a document without a pinpoint. Identify the exact page, figure, testimony lines, or controlled code location supporting the proposition.
• Combining releases without support. Split version-specific mappings or document the evidence supporting representative treatment.
• Treating silence as absence. Distinguish missing documentation from affirmative evidence that the implementation lacks a feature.
• Hiding contradictory architecture evidence. Record conflicts beside the mapping and identify what could resolve them.
• Circulating the working chart as approved contentions. Review disclosure completeness and restrictions; separate internal investigation notes from the approved deliverable.

Final quality check

• Every asserted claim and dependency is accounted for without altering issued wording.
• Each mapping identifies the implementation, configuration, and relevant version.
• Every source reference resolves to the reviewed copy and exact pinpoint.
• Documentary support, testimony, inference, and expert interpretation are distinguished.
• Contrary evidence and construction-dependent positions remain visible.
• Material gaps are resolved or assigned with a decision question and owner.
• The intended deliverable has been checked against applicable rules, orders, and confidentiality restrictions.

Complete the workflow with DecoverAI

DecoverAI is the Legal Intelligence Platform for litigation and transactions. Its litigation document review and evidence analysis capabilities can support examination of technical records and identification of supporting or contradictory material. Explore its litigation capabilities. The chart remains counsel’s work product: lawyers retain responsibility for claim segmentation, technical interpretation, source verification, and infringement positions. Book a demo to discuss evidence analysis for this workflow. (decover.ai)

Frequently asked questions

Can one chart cover several accused products?

As an organizational choice, yes, but identify the evidence supporting each implementation and check the required disclosure format. Use separate rows when versions, configurations, or component locations differ materially. Shared branding alone is not a basis for treating technical operation as identical.

How should dependent claims be shown?

Preserve their added wording and explicitly reference inherited limitations throughout the dependency chain. The reader should be able to reconstruct the complete comparison without guessing which earlier rows apply. Recheck inherited mappings against the accused version used for the dependent claim.

What if source code is unavailable?

Use manuals, testing, schematics, and testimony at their actual evidentiary strength. Identify the uncertain operation and assign targeted follow-up. Do not describe inferred implementation as observed code behavior. Check handling restrictions before moving restricted excerpts into the chart or another review environment.

Is a completed claim chart itself evidence?

Not automatically. Rule 1006 permits summaries proving the content of voluminous admissible materials that cannot conveniently be examined in court, with underlying originals or duplicates available for examination or copying at a reasonable time and place. Charts functioning only as illustrative aids fall under Rule 107. A claim chart’s technical interpretations require their own analysis; source references alone do not establish admissibility. See Evidence Rules 1006 and 107. (uscourts.gov)

Sources

• Northern District of California Patent Local Rules, particularly Rules 1-3 and 3-1.
• Federal Rules of Civil Procedure, Rule 26(b)(1).
• Federal Rules of Evidence, Rules 901, 1006, and 107.

Scope: General professional workflow, not advice for a particular matter. The procedural sources support the identified requirements, not a complete substantive infringement analysis; the charting and review recommendations are practical methods.

This guide is general information about legal workflows, not legal advice for any matter. It was drafted and reviewed with AI assistance. The worked example is fictional. Contract rights and procedure depend on the governing law, the actual wording, and the facts, so check the cited sources and the rules that apply in your jurisdiction before relying on it.