This guide covers how to check one: what to confirm before you read any row, how to read a single row properly, which rows deserve your time, and what a chart cannot settle no matter how well it is filled in.
Key takeaways
- Check what the chart compares before you check its conclusions. An incorrect claim version can undermine the mapping, while missing product evidence can leave specific questions unresolved.
- Trace each material mapping from the claim language to the supporting product evidence. Following one row is a useful starting exercise, not a substitute for checking the claim as a whole.
- Prioritise unresolved or consequential mappings, including apparently clear findings that drive the outcome. A chart does not need a quota of uncertain rows, but unresolved questions must remain visible.
A complete chart is not a checked chart
The presentation of a generated chart does not tell you whether a mapping is correct. A clear layout helps you inspect the reasoning; it does not validate it.
A person who built a chart may remember difficult elements and missing documents, but memory is not a shared audit trail. Generated and manually prepared charts both need a record of the sources, assumptions, and unresolved questions behind their mappings.
Make the checking deliberate, starting with the comparison baseline.
Before you read a single row
Confirm three things before relying on the rows. They may appear in the chart header, linked records, or the table itself.
Which claim, from which document, in which version. Confirm the publication or patent number, jurisdiction, claim number, and exact text used. Distinguish a published application from the granted patent, and check for later amendments affecting the claim. Record the claim-text source or version separately from the date and source of the legal-status check.
Whether dependent claims include all inherited limitations. A dependent claim incorporates the limitations of the claim it refers to, including any earlier dependencies. The chart can repeat those limitations or clearly cross-reference an existing mapping. What matters is that none is omitted from the assessment.
Which product, at which revision. Identify the configuration and the documents used to describe it. Check whether marketing material, specifications, and test records refer to the same version. Where the version or a relevant feature cannot be established, record that gap rather than treating the available description as a complete product record.
Clarify any material uncertainty in the baseline before relying on the affected mappings. You may still inspect the rows to identify which documents or facts are missing.
Read one row end to end
Pick a material row and follow it from start to finish. Use the same checks for other mappings that affect the assessment, and check the relationships between the elements as well.
Start at the claim language. Is the element quoted accurately, and is it quoted with enough context to be read correctly? An element lifted out of its clause can lose a positional or functional requirement that the full text carries.
Move to the product evidence. Open the cited record. Identify the document, revision, and relevant page, figure, or passage, then check whether it actually supports the mapped feature. A named source is a route to evidence, not proof that the evidence establishes the correspondence.
Read the rationale. Does it explain the correspondence, or does it restate the mapping in different words? "The product includes a motion sensor, therefore the motion sensor element is identified" is a restatement. "The passive infrared module described in specification revision 4 performs the detection function at the location required by the claim" is a reason.
Check the mapping against the rationale. A rationale that describes uncertainty alongside an Identified label is a mismatch to investigate. Do not assume either the label or the explanation is correct. Resolve the discrepancy against the claim language and the product evidence.
These checks make the basis of a mapping easier to discuss with engineering or counsel. They do not settle the legal outcome. Where the source or reasoning cannot be verified, record what is missing and limit reliance on the mapping accordingly.
Put a claim comparison under review. Bring your product description to FTO Search, then inspect the claim-level comparisons against the supporting evidence.
The rows that deserve your time
Use these four patterns to prioritise deeper review, not to exempt the other rows. A finding that a required limitation is absent can be decisive even when it is presented confidently.
Rows where the rationale hedges but the mapping does not. Wording such as appears to, likely, or consistent with may signal uncertainty that the mapping label does not capture. Check whether the distinction is justified by the evidence.
Rows resting on a functional match. A function performed elsewhere in the architecture raises both technical and claim-construction questions. Establish the system boundaries, component relationships, and data flow before asking whether the claimed limitation is met under the relevant interpretation.
Rows with an empty evidence cell. A blank may reflect missing documentation or unfinished mapping. It is not evidence that the product lacks the element. Record what is missing and what would resolve it.
Rows involving a disputed interpretation. Identify any claim term whose meaning could change the outcome in a relevant jurisdiction. Record the interpretation used and the issue that needs professional review, regardless of how confidently the chart presents it.
What the chart cannot settle
Some questions do not live in cells, and a chart becomes misleading when it appears to answer them.
Claim construction. How a term is read depends on the specification, the prosecution history, and the law of the jurisdiction. A chart records the construction it assumed. It does not establish that the construction is correct.
Equivalents. The treatment varies by jurisdiction and cannot be reduced to a column applied across a table.
Validity. An element can map cleanly onto your product while the claim itself is vulnerable. That changes your options without changing a single cell.
Whether the set of charted patents is the right set. A chart tells you about the patents someone decided to chart. It says nothing about the ones that were never screened in.
Work through the factual and search questions with the appropriate engineering and IP specialists, and take the legal questions to qualified counsel in the relevant market. A good chart makes those questions specific; it does not settle them by being complete.
The questions to put to any chart
Whether the chart came from a colleague, an external firm, or a tool, the same short list works.
- Which claim version was charted, and when was legal status last checked?
- Which product record was used, and at what revision?
- Which rows would you change if you had another hour?
- Which rows rely on an assumption about claim construction rather than a technical fact?
- Which patents were screened out before charting, and on what basis?
The third question can help identify where further work would matter. Ask the person responsible for reviewing the chart to explain that priority from the record, rather than assuming the person who generated it knows where every weakness lies.
How claim-level comparisons appear in Eureka
The Eureka FTO Search workflow starts from a product description. The description is broken into technical features, candidate patents are screened, and selected claims are compared with the product features. The resulting comparisons give reviewers a structured starting point for checking the claim language, proposed correspondence, and supporting evidence.
Reviewers should still confirm the claim version, inherited limitations, and product configuration, then check each material mapping against the original records. Missing evidence or uncertain reasoning should remain visible in the review record. A completed comparison is not an infringement determination.
This article is general information about analysis method. It is not legal advice. A claim chart organises preliminary evidence. It is not an infringement determination or a launch clearance. Claim construction, equivalents, validity, and infringement depend on the applicable jurisdiction and facts, and require qualified professional advice.
