A claim chart example is more useful than a description of one, because the format is simple and the discipline is where charts succeed or fail.
Two columns. Claim language on the left, evidence on the right.
One row per limitation, a cited source in every row, and an honest grade including "unverified" where the evidence does not exist.
Three worked examples below — a clean infringement chart, one with a gap, and an invalidity chart.
The columns
| Column | Contains |
|---|---|
| Element | The exact claim words, one limitation per row |
| Accused feature / prior art | The specific corresponding thing |
| Evidence | Source with page, figure or paragraph |
| Grade | Present / absent / arguable / unverified |
| Notes | Construction applied, source date |
A two-column chart is the minimum. The grade and notes columns are what make it usable months later by someone who was not there.
Example 1: a clean infringement chart
Claim 1: A fluid separator comprising: a housing defining a flow path having an inlet and an outlet; a baffle disposed within the flow path, the baffle defining an aperture; and a biasing element coupled to the baffle and configured to vary an area of the aperture in response to a flow velocity within the flow path.
| # | Element | Accused feature | Evidence | Grade |
|---|---|---|---|---|
| 1a | A housing | Cast aluminium body | Service Manual SM-400 rev 3, fig. 2 | Present |
| 1b | defining a flow path | Internal bore, inlet to outlet | SM-400, fig. 2, items 4–7 | Present |
| 1c | having an inlet and an outlet | Ports A and B | Spec sheet DS-400, p.1 | Present |
| 2a | A baffle | Deflector plate | SM-400 §4.2 | Present |
| 2b | disposed within the flow path | Mounted in the bore | SM-400, fig. 3 | Present |
| 2c | the baffle defining an aperture | Central 12 mm opening | SM-400, fig. 3, item 11 | Present |
| 3a | A biasing element | Compression spring | SM-400 §4.3 | Present |
| 3b | coupled to the baffle | Direct mechanical coupling | SM-400, fig. 3, item 14 | Present |
| 3c | configured to vary aperture area | Aperture opens with flow | Their US 11,XXX,XXX, para. 41 | Present |
| 3d | in response to a flow velocity | Velocity-driven | Their US 11,XXX,XXX, para. 44 | Present |
Ten rows from three claim elements. Splitting at every limitation is what makes the chart checkable.
Rows 3c and 3d cite the target's own patent application. Companies describe their approach accurately in their filings, and those documents are public and dated.
The product also has a heater and a digital display. Irrelevant — the transition is "comprising", so extra features never avoid the claim.
Example 2: a chart with a gap handled properly
Same claim, a different product.
| # | Element | Accused feature | Evidence | Grade |
|---|---|---|---|---|
| 1a–1c | Housing, flow path, ports | Moulded body | User Manual UM-22, p.4 | Present |
| 2a–2c | Baffle with aperture | Internal disc | Teardown photos, 12 Mar 2026 | Present |
| 3a | A biasing element | Not observable externally | — | Unverified |
| 3b | coupled to the baffle | Assumed from operation | — | Unverified |
| 3c | varies aperture area | Aperture visibly changes in operation | Test video, 12 Mar 2026 | Arguable |
| 3d | in response to flow velocity | Correlates with flow rate | Test data, 12 Mar 2026 | Arguable |
What the chart says about itself
| Finding | Note |
|---|---|
| Elements 3a–3b unverified | Internal mechanism not observable |
| What would settle it | Full teardown, service manual, or discovery |
| Elements 3c–3d arguable | Behaviour observed, mechanism inferred |
| Conclusion | Not established on public evidence |
This chart is more useful than one asserting the mechanism exists. It tells the reader exactly what is missing and what would close it.
Grading honestly protects the chart. One assumption presented as fact makes every other row suspect.
Example 3: an invalidity chart
Same claim, mapped against prior art instead of a product.
| # | Element | Prior art disclosure | Evidence | Grade |
|---|---|---|---|---|
| 1a–1c | Housing, flow path, ports | Disclosed | US 6,XXX,XXX (Hallberg), col. 3:12–28 | Present |
| 2a–2c | Baffle with aperture | Disclosed | Hallberg, fig. 4, col. 4:5–19 | Present |
| 3a | A biasing element | Fixed baffle — no biasing | Hallberg, col. 4:22 | Absent |
| 3b–3d | Variable aperture | Not disclosed | — | Absent |
The obviousness combination
| Reference | Supplies |
|---|---|
| Hallberg | Elements 1a–2c |
| JP-XXXXXX (Tanaka), fig. 2 | Spring-biased variable aperture |
| Motivation to combine | Must be shown, not assumed |
Anticipation fails here. Hallberg lacks the biasing element, and a single reference must disclose every element for §102.
Obviousness needs the combination argued, not just charted. A two-reference chart showing all elements between them does not by itself establish that a skilled person would have combined them. See patent invalidity search.
What separates usable from unusable
| Usable | Unusable |
|---|---|
| One row per limitation | Elements combined |
| Specific page or figure cited | "See product literature" |
| Honest grades including unverified | Everything marked present |
| Source dates recorded | Undated |
| Every element addressed | Elements skipped |
| Construction stated where disputed | Silent |
| The full claim charted | Partial |
"See the website" is not evidence. A source citation must let someone else find the same thing.
Date every source. A 2022 manual may not describe the 2026 product, and that gap gets found by the other side if not by you.
Method claims need a performer column
| # | Step | Performed by | Evidence |
|---|---|---|---|
| 1 | Directing the fluid stream | The customer | UM-22, p.9 |
| 2 | Varying the aperture area | The device | SM-400 §4.3 |
| 3 | Reporting the result | The vendor's service | Portal docs, p.2 |
Steps split across parties raise divided infringement. Where no single party performs every step, attribution requires direction or control, and an ordinary vendor-customer relationship generally does not meet that.
Add the column whenever the claim is a method. It surfaces the issue rather than hiding it.
What the chart does not decide
| Question | Where it lives |
|---|---|
| What the claim terms mean | Construction, before the chart |
| Whether a gap is bridged by equivalents | Separate analysis |
| Whether estoppel bars equivalents | Prosecution history |
| Whether the patent is valid | A different chart |
| Whether pursuit is worthwhile | Commercial judgement |
Construe before charting. A mapping built on an unstated reading of a term changes when a court reads it differently.
Check estoppel before relying on equivalents. It is free to establish from a public file wrapper and frequently disposes of the argument. See prosecution history estoppel.
Before charting anything
| Check | Why | Time |
|---|---|---|
| Is the patent in force? | 58.6% are abandoned before term | 3 min |
| Who owns it? | The front page is out of date | 5 min |
| Which claims survived challenge? | Narrowed claims chart differently | 10 min |
| Ipiry Patent Survival Curve v1.0 | Rate |
|---|---|
| Survive the 3.5-year fee (2022 cohort) | 85.8% |
| Survive the 7.5-year fee (2018 cohort) | 64.6% |
| Reach full term (2014 cohort) | 41.4% |
Computed from 27,273,654 USPTO maintenance fee records covering 8,262,336 US utility patents — see the patent survival curve.
Where automation helps
| Task | Automation |
|---|---|
| Finding candidate products | Substantial |
| Retrieving documentation | Substantial |
| First-draft mapping | Substantial |
| Verifying every citation | Human |
| Grading honestly | Human |
| Deciding what to do | Human |
Verify every cited source. A fabricated or misread citation is worse than an empty row, because it makes the whole chart untrustworthy at exactly the moment it matters.
Claim chart example: the checklist
- Two columns minimum, four if the chart will be read later.
- One row per limitation, splitting at every requirement.
- Cite a specific page, figure or paragraph in every row.
- Search the target's own patent filings. Best available evidence.
- Grade honestly. "Unverified" is more useful than an assumption.
- State what evidence would close each gap.
- Chart the full claim. Skipping an element makes it useless.
- Add a performer column for method claims.
- Construe disputed terms before charting, and record the construction.
- Check the patent is in force before doing any of it.