Webinar | If Everyone Has AI, Who Wins SEP Licensing!
3 September | 12:00 PM to 1:00 PM ET: Register Now

Instagram X-twitter Linkedin-in Youtube
GET IN TOUCH

The Prior Art Search is step one.
Most of the technical work comes after.

Once references are in, the technical questions multiply at every stage, under every deadline. Which references hold up against prosecution-narrowed scope? What does the record say about each disputed term? Who builds the citation record before the expert phase? GreyB stays in the case from search through trial, at technical rates.

Discuss your matter
See all modules
What litigators describe, repeatedly
Before contentions

Three references looked strong on claim language. Two failed after prosecution history review. Grounds were filed based on the surface read.

Before Markman
A narrowing was buried in office action 9, page 12. Opposing counsel found it at the hearing. The brief was already drafted.
Expert phase

The expert billed 12 hours at $700/hour assembling the state-of-the-art record. None of it required expert judgment. Only domain familiarity.

Litigation initiation
A §112 nonce-word ground was available from day one. It stayed unused because prior art charting consumed the entire schedule before the four-ground analysis was ever run.

Lawfirms impressing clients with our help

hogan-lovells
ab-inbev-1
lafargeholcim
ups-global-innovations
swisscom
norton-rose-fulbright
pierce-atwood
pruefer-partner
saint-gobain-1
sandoz
yetter-coleman
bakerbotts
bugnion
cooley
dla-piper
dority-manning
fenwick-west
fives-group
frys-electronics
goldman-ismail-tomaselli
grab
greenberg-traurig
guardian-ip-consulting
healthbeacon
henkel
holland-hart
ksb-se-co.
lesaffre
merck
natura-co

Patent litigation has a bundling problem nobody talks about.

Every matter mixes two fundamentally different types of work. For most of the industry’s history, both were billed the same way, by the same people, at the same rates. Not by design. Because there was no alternative.

Judgment work

Strategy, argument construction, claim construction decisions, ground selection, deposition positioning

Requires legal judgment built over decades. This is what clients pay attorney rates for, and rightly so.
Investigation work

Prosecution history mapping, reference evaluation, state-of-the-art citation assembly, priority chain verification, system prior art documentation

Requires deep technical domain knowledge and focused investigative time. Not legal judgment. The billing model does not distinguish between them.

Corporate clients are not refusing to pay for judgment. What they are increasingly refusing to pay for is investigation work billed at judgment rates. Hourly rates have climbed significantly over three years. Case budgets have not moved. The pressure is falling on the bundle.

The firms navigating this well are not cutting scope. They are separating two things that were bundled together by habit, not by logic. The result is a better technical record at every stage and a billing structure clients can actually defend internally.

The shift

The technical thread in a patent case does not begin and end with prior art. It runs through the entire matter. Litigation+ keeps that thread live from search through trial.

Three verdicts. One missing technical layer each time.

Prior art alone is not enough. The prosecution history, claim scope, and mechanism analysis around it are where cases are decided.

$940M+

VirnetX v. Apple

VirnetX disclaimed indirect VPN across pages of prosecution history. No one read it at trial. PTAB found Kiuchi (1996), a direct-VPN reference the disclaimer brought squarely within claim scope. Federal Circuit affirmed. $502M verdict vacated.

One prosecution exchange. Not hidden. No one read it.

$100M

Ericsson v. TCL
TCL moved for §101 summary judgment before trial. Denied. Jury returned $100M. Federal Circuit stripped the device nouns: “controlling access to resources.” Abstract idea, no inventive concept. Verdict vacated. The §101 reduction test was available from day one.
The strongest defense went unused because the test was never run.

$355K

Nike v. Lululemon

March 2025: jury verdict for Nike. Lululemon’s parallel IPR: PTAB invalidated all claims August 2025. JMOL issued March 2026 and erased the verdict. Warp knitting and cutting/incorporating were established techniques before the patent.
The obviousness case was there before trial. The SOTA was never built.
Discuss your matter

How we do this

Active litigation

You have 30 references. How do you know which two will actually hold up in an IPR after prosecution history review?

Associates spend days evaluating references against claim elements. In practice, cells get marked strong or weak based on terminology overlap. The prosecution history that quietly excludes a reference is read later, if at all. False positives are discovered after filing, not before.

What we deliver

A visual matrix of all references against all asserted claim limitations, color-coded by coverage strength. Every cell carries a pin-cite and a written explanation of why the reference teaches or does not teach the limitation, evaluated against post-prosecution claim scope. False positives are flagged at the mechanism level before grounds are committed.

Case study / Wireless networking

Two grounds dropped before filing. One held at institution.

Three references all taught “dynamic allocation” on the face of the claim. Prosecution history revealed the applicant had argued “dynamically” means per-packet decisions, excluding per-session allocation entirely. Two references failed. One survived. The petition was restructured around that single ground. The patent owner’s response had no mechanism to attack it.

Reference Triage / Heat Map

SAMPLE OUTPUT
US'769B1 CLAIM 01 BEN-SHAUL ET AL. DEHAAN ET AL. VISWANATH ET AL.
1a — [Method step]
1b — [Allocate]
1c — [Dynamic]
1d — [Per-packet]
1e — [Condition]
Full explicit disclosure
Partial / implicit
Gap

Prosecution history flag

Applicant argued "dynamically" means per-packet allocation in OA response dated [date]. Ben-Shaul and DeHaan allocate per-session. Both fail on prosecution-narrowed scope despite surface coverage of 1c.

Claim Term Information Package (CTIP)

Sample output

Claim Term

GreyB Notes

Estoppel

"AAA module"

Feature added via amendment to distinguish Sayers. Applicant argued Sayers does not disclose an AAA module including both HLR and CLR. Term is prosecution-narrowed to a single structural entity managing HLR/CLR functions. Infringement contention maps this to distributed cloud services collectively — contradicts narrowing.

Action: Search for art grouping HLR/CLR in a single AAA module at one local node.

YES

"First home location register (HLR)"

No prosecution estoppel on the term itself. HLR is well-understood in telecom as a local subscriber database. Infringement contention maps HLR to modern 4G HSS and 5G UDR. The novelty lies in its relationship to the CLR, not in HLR itself.

YES

"First community location register (CLR)"

Critical. CLR is a coined term. Applicant argued Sayers’ nodes do not store information associated with mobile stations registered with other BTSs — this is the entire basis of the CLR. CLR must be a local cache of remote subscriber data. A shared centralized UDR is directly excluded by this argument.

Estoppel: CRITICAL. Search for P2P pre-fetches, caches of remote subscriber data at local nodes.

CRITICAL

Active litigation

Markman briefing starts next month. Do you know where every disputed term stands in the prosecution history right now?

Prosecution history review for each disputed term is thorough work done under the worst possible conditions: while drafting the brief, preparing for the hearing, and coordinating with the expert. A narrowing argument buried in office action 9 gets missed. Opposing counsel finds it at the hearing. You move from proactive framing to reactive damage control.

What we deliver

A dossier per disputed term: technical meaning at the priority date, specification passages in context, prosecution history statements with exact citations, estoppel flags, and the non-infringement or invalidity position that follows from each plausible construction. Delivered before drafting begins, so you enter Markman knowing the full prosecution picture for every term.

Case study / 5G distributed architecture

Infringement theory contradicted the patentee's own prosecution arguments.

Three references all taught “dynamic allocation” on the face of the claim. Prosecution history revealed the applicant had argued “dynamically” means per-packet decisions, excluding per-session allocation entirely. Two references failed. One survived. The petition was restructured around that single ground. The patent owner’s response had no mechanism to attack it.

Active litigation

Your expert is billing at $700/hour. How much of that time is spent on research that does not require expert judgment?

The background section gets written from general knowledge first, then citations are found to support what was already written. Date discipline slips and post-priority-date sources appear. The expert bills 10 to 15 hours at expert rates assembling background material that requires domain familiarity, not expert opinion.
What we deliver
A problem-focused technical narrative tracing the evolution of the relevant technology up to the priority date. Citation-first from patents, standards bodies, textbooks, and technical papers. Every statement dated before the priority date. Formatted for direct use in IPR petitions and expert declarations. The expert’s time goes to forming opinions, not building the record.

Case study / Software / Bot detection

Expert hours dropped from 12 to 3. Citation record built before first session.

A petitioner needed to show behavioral bot detection was not novel. The SOTA traced convergence across five independent tracks from 1997 to 2007: early CAPTCHA at DEC, CMU’s Gimpy, PayPal’s fraud protection deployment, UC Berkeley’s 2002 attack, and reCAPTCHA’s probabilistic confidence scoring. Every statement dated before the priority date. The expert spent 3 hours forming opinions instead of 12 hours assembling citations.

State of the Art / Technology Evolution S

Sample output

Percussive therapy: 2008 to 2018 (Search cut-off: Dec 26, 2018)

2008

Thumper Massager (product manuals)

Motor-driven reciprocating percussive action via eccentric mechanism. Long ergonomic handle. Single-head output with interchangeable attachments. Handle-output angle near-obtuse.

2009

HoMedics Percussion Massager

Defined housing with ergonomic handle. Motor and cam/eccentric reciprocating nodes. Interchangeable attachment system. Commercial-scale deployment.

2012

US8512265B2 - Percussive massager

Housing with percussive drive unit, motor, and reciprocating multi-node head. First patent family establishing core architecture.

2017

TimTam Power Massager 1.5

90-degree articulating (rotatable) head enabling obtuse angle between handle and output shaft. Enables reach to various muscle groups. First commercially documented rotatable-head percussive device.

2018

Wahl Deep Tissue Percussion Massager

Ergonomic handle and percussion-head arrangement exhibiting obtuse-angle formation. Detachable heads. Commercially available before Dec 26, 2018 cut-off date.
Every milestone dated and sourced to a public document before the priority date.

Priority Chain Analysis

Sample output

Original filing

US 16/711,340

Filed: Dec 11, 2019 → Issued as US 11,796,833B2 (Oct 24, 2023)
CIP filed while parent pending ↓

Original filing

US 16/711,340

Filed: Dec 11, 2019 → Issued as US 11,796,833B2 (Oct 24, 2023)

Continuation filed while parent pending ↓

Original filing

US 16/711,340

Filed: Dec 11, 2019 → Issued as US 11,796,833B2 (Oct 24, 2023)

Priority finding

“Second sensor” (external sensors) first appears in dependent Claim 18 of US11835798. Independent claims of US12216339 are fully supported by the Jan 2020 CIP filing. Effective prior art cutoff: December 2018.

Any stage

The patent claims a priority date through a continuation chain. Does that claim actually hold for every asserted element?

A 2-day overrun in the Paris Convention window permanently destroys a priority claim. A single claim element first disclosed in a CIP shifts the effective date forward and opens prior art that was previously blocked. These breaks are invisible on the face of the patent and invisible in the bibliographic data. The search is run against the wrong date.

What we deliver

A patent-by-patent, claim-by-claim priority chain audit: every statutory deadline verified to the calendar day, every application transition checked for proper referencing and inventor continuity, every asserted claim element mapped to the earliest specification that actually discloses it. Each gap documented with the prior art exposure it opens.

Case study / Consumer electronics / Smart eyewear

Effective date confirmed. Search calibrated against the actual cutoff.

A smart eyewear patent claimed priority through a continuation to a 2019 filing. A CIP introduced the key “second sensor” element first in a dependent claim. The audit confirmed this element was fully supported in the CIP parent, setting the effective prior art cutoff at December 2018. Art between 2018 and 2020 was correctly excluded. The search was calibrated against the actual effective date, not an assumed one.

Litigation initiation

The §112 analysis is sitting in the backlog. When will it get done, and what is the cost of finding the strongest ground after the petition strategy is already locked?

§112 grounds get treated as a fallback and run last, if at all, while prior art charting consumes the schedule. A nonce-word limitation with no disclosed algorithm is one of the strongest grounds available, and it stays unused because the four-ground analysis was never run before the deadline passed.
What we deliver
A four-ground §112 analysis per asserted claim: written description, enablement, definiteness, and means-plus-function, with an invalidity-strength matrix flagging the grounds worth pursuing. For each §112(f) trigger: nonce-word identification, corresponding-structure linkage, and the disclosed-algorithm gap. You know the strongest ground before petition strategy is locked.

Case study / Consumer electronics / Percussive therapy

Strongest ground identified before prior art charting started.

A four-ground analysis on a percussive therapy device patent found “dividing plane” had no antecedent basis anywhere in the specification, a pure claim-only term, flagged as the strongest §112(b) ground for the IPR petition lead. A second ground on the “interior angle is obtuse” limitation showed zero spec guidance on how to measure the angle in three-dimensional space. Both were identified before the filing timeline was consumed.

§112 Invalidity Analysis

Sample output

Percussive therapy: 2008 to 2018 (Search cut-off: Dec 26, 2018)

Limitation

Ground

Strength

Action

“Dividing plane” + proximal end offset geometry (core wherein clause)
§112(b) Indefiniteness

Highest

Challenge

“Interior angle between handle portion and output shaft is obtuse” (Claims 1, 15, 18, 20)
§112(b) Indefiniteness

High

Challenge

Preliminary amendment change from “connector” to “output shaft” + new geometry claims 14–16
Prosecution history estoppel (Festo)

Moderate

Include

Key finding: “Dividing plane” appears only in the claims. Zero occurrences in the Detailed Description. No antecedent basis. Expert testimony can argue POSITA cannot ascertain scope with reasonable certainty under Nautilus.

Prosecution History Insights

Sample output

Flag type

Finding

Impact

Examiner search gap

Examiner limited all search strings to “claims” field only. Broad strategies returned 22,000+ results in several lanes, burying relevant references. Keywords “circular path,” “rotatable arm,” and “datum” never searched.

Open IPR lane

Terminal disclaimer

TD filed Aug 2, 2013 against co-pending ‘701 application. TD requires common ownership at all times. Acquisition during distressed sale period Nov 2024 to Jan 2025 raises gap in common ownership. Discovery target: assignment timing during DRNC acquisition from Edgio Chapter 11 estate.

Enforceability risk

Minimal prosecution record

Entire prosecution history: one Pre-Interview First Action (Jul 2, 2013), one oral FAI interview (Jul 23, 2013), supplemental amendment (Aug 2, 2013), terminal disclaimer, allowance. No second office action. No final rejection. No examiner appeal. FAI program resolved prosecution verbally.

Minimal estoppel risk / Broad IPR lane on contested terms

Any stage

The prosecution history is 400 pages. The decisive move, the narrowing, the terminal disclaimer, the examiner gap, is in there. Where?

The prosecution history gets read under deadline, in parallel with briefing and expert coordination. A narrowing amendment buried in a third office action response gets missed. A terminal disclaimer tying two patents to common ownership is overlooked. An examiner who never ran a §103 combination leaves an open IPR lane. The other side finds it first.

What we deliver

A structured read of the full prosecution history: estoppel-creating amendments with citations, terminal disclaimer and double-patenting risks, §112 and defective-preamble openings, examiner search gaps that leave an open IPR lane, and ownership-chain standing flags. You enter contentions, the petition, or Markman knowing exactly where the prosecution record helps and where it hurts.

Case study / Consumer electronics

Examiner search gaps opened two leads that the standard search missed entirely.

Prosecution history review found the examiner had limited search strings to “claims” only, with broad strategies returning thousands of results that buried relevant references. Missing keywords included “circular path,” “rotatable arm,” and “datum.” Following these gaps led to two leads: a system prior art product with a rotatable head (TimTam, 2017) and a non-gun-format massager exhibiting the claimed obtuse angle (Wahl, 2018). Both pre-date the critical date.

When the case turns on how a mechanism works, we build it.

Standard 2D patent figures ask a judge or jury to hold a moving assembly in their head. In mechanical matters, that is where a sound argument gets lost. We reconstruct the prior art or the accused system as a 3D model from the figures and specification, then map each part to the limitation in dispute. The distinction becomes something the room can see, not something it has to reason through.

3d
Discuss your matter

Eleven modules. Every stage.

Each module targets a specific decision point. Use one on its own or combine them across the matter. Illustrated in detail above for the five most commonly engaged modules.

01

Reference Triage / Heat Map

Active litigation

Prosecution-aware coverage matrix. False positives caught at mechanism level before grounds are committed.

02

Claim Term Information Package

Active litigation

Evidence dossier per disputed term. Prosecution history estoppel flags. Delivered before Markman briefing begins.

03

State of the Art (SOTA)

Active litigation

Citation-first technology evolution narrative. Every statement dated before the priority date. Formatted for IPR petitions and expert declarations.

04

Priority Chain Analysis

Any stage

Claim-by-claim audit of every statutory deadline and application transition. Each gap documented with the prior art exposure it opens.

05

Counter-Argument Diagnostic

Active litigation

Mechanism-level rebuttals to patent owner responses. Stays within the existing record. No procedural exposure from new theories.

06

Non-Infringement Arguments

Active litigation

Limitation-by-limitation analysis at the architecture level. Each distinction supported by evidence of how the product actually works internally.

07

Supplementary Gap-Fill Search

Active litigation

Surgical follow-up search targeting specific weaknesses in the current reference pool. Focused, not a full re-search.

08

§112 Invalidity Analysis

Litigation initiation

Four-ground analysis with invalidity-strength matrix. Strongest ground identified before petition strategy is locked.

09

§101 Invalidity Analysis

Litigation initiation

Alice/Mayo two-step applied per claim. Device nouns stripped. You know whether §101 is your strongest argument before filing.

10

System Prior Art Package

Any Stage

Multi-source package: timestamped video evidence, brochures, NPL, archived pages, pre-assembled and mapped to claim limitations with public-availability dates.

11

Prosecution History Insights

Any Stage

Structured read of the full prosecution history. Estoppel flags, double-patenting risks, examiner search gaps, open IPR lanes, ownership-chain flags.

+

Custom Technical Request

Any Stage

Not sure module fits? Describe the matter and we will point you to the right one.
Discuss your matter

01

Priority Chain Analysis

Any stage

Claim-by-claim audit of every statutory deadline and application transition. Each gap documented with the prior art exposure it opens.

02

§112 Invalidity Analysis

Litigation initiation

Four-ground analysis with invalidity-strength matrix. Strongest ground identified before petition strategy is locked.

03

§101 Invalidity Analysis

Litigation initiation

Alice/Mayo two-step applied per claim. Device nouns stripped. You know whether §101 is your strongest argument before filing.

04

System Prior Art Package

Any Stage

Multi-source package: timestamped video evidence, brochures, NPL, archived pages, pre-assembled and mapped to claim limitations with public-availability dates.

05

Prosecution History Insights

Any Stage

Structured read of the full prosecution history. Estoppel flags, double-patenting risks, examiner search gaps, open IPR lanes, ownership-chain flags.

+

Custom Technical Request

Any Stage

Not sure module fits? Describe the matter and we will point you to the right one.

Discuss your matter

01

Reference Triage / Heat Map

Active litigation

Prosecution-aware coverage matrix. False positives caught at mechanism level before grounds are committed.

02

Claim Term Information Package

Active litigation

Evidence dossier per disputed term. Prosecution history estoppel flags. Delivered before Markman briefing begins.

03

State of the Art (SOTA)

Active litigation

Citation-first technology evolution narrative. Every statement dated before the priority date. Formatted for IPR petitions and expert declarations.

04

Priority Chain Analysis

Any stage

Claim-by-claim audit of every statutory deadline and application transition. Each gap documented with the prior art exposure it opens.

05

Non-Infringement Arguments

Active litigation

Limitation-by-limitation analysis at the architecture level. Each distinction supported by evidence of how the product actually works internally.

06

§112 Invalidity Analysis

Litigation initiation

Four-ground analysis with invalidity-strength matrix. Strongest ground identified before petition strategy is locked.

07

§101 Invalidity Analysis

Litigation initiation

Alice/Mayo two-step applied per claim. Device nouns stripped. You know whether §101 is your strongest argument before filing.

08

System Prior Art Package

Any Stage

Multi-source package: timestamped video evidence, brochures, NPL, archived pages, pre-assembled and mapped to claim limitations with public-availability dates.

09

Prosecution History Insights

Any Stage

Structured read of the full prosecution history. Estoppel flags, double-patenting risks, examiner search gaps, open IPR lanes, ownership-chain flags.

+

Custom Technical Request

Any Stage

Not sure module fits? Describe the matter and we will point you to the right one.

Discuss your matter

01

Reference Triage / Heat Map

Active litigation

Prosecution-aware coverage matrix. False positives caught at mechanism level before grounds are committed.

02

Claim Term Information Package

Active litigation

Evidence dossier per disputed term. Prosecution history estoppel flags. Delivered before Markman briefing begins.

03

State of the Art (SOTA)

Active litigation

Citation-first technology evolution narrative. Every statement dated before the priority date. Formatted for IPR petitions and expert declarations.

04

Prosecution History Insights

Any Stage

Structured read of the full prosecution history. Estoppel flags, double-patenting risks, examiner search gaps, open IPR lanes, ownership-chain flags.

+

Custom Technical Request

Any Stage

Not sure module fits? Describe the matter and we will point you to the right one.

Discuss your matter

01

Claim Term Information Package

Active litigation

Evidence dossier per disputed term. Prosecution history estoppel flags. Delivered before Markman briefing begins.

02

Prosecution History Insights

Any Stage

Structured read of the full prosecution history. Estoppel flags, double-patenting risks, examiner search gaps, open IPR lanes, ownership-chain flags.

+

Custom Technical Request

Any Stage

Not sure module fits? Describe the matter and we will point you to the right one.

Discuss your matter

01

Reference Triage / Heat Map

Active litigation

Prosecution-aware coverage matrix. False positives caught at mechanism level before grounds are committed.

02

Counter-Argument Diagnostic

Active litigation

Mechanism-level rebuttals to patent owner responses. Stays within the existing record. No procedural exposure from new theories.

03

Non-Infringement Arguments

Active litigation

Limitation-by-limitation analysis at the architecture level. Each distinction supported by evidence of how the product actually works internally.

04

Supplementary Gap-Fill Search

Active litigation

Surgical follow-up search targeting specific weaknesses in the current reference pool. Focused, not a full re-search.

05

System Prior Art Package

Any Stage

Multi-source package: timestamped video evidence, brochures, NPL, archived pages, pre-assembled and mapped to claim limitations with public-availability dates.

+

Custom Technical Request

Any Stage

Not sure module fits? Describe the matter and we will point you to the right one.

Discuss your matter

01

State of the Art (SOTA)

Active litigation

Citation-first technology evolution narrative. Every statement dated before the priority date. Formatted for IPR petitions and expert declarations.

02

Non-Infringement Arguments

Active litigation

Limitation-by-limitation analysis at the architecture level. Each distinction supported by evidence of how the product actually works internally.

03

Supplementary Gap-Fill Search

Active litigation

Surgical follow-up search targeting specific weaknesses in the current reference pool. Focused, not a full re-search.

04

System Prior Art Package

Any Stage

Multi-source package: timestamped video evidence, brochures, NPL, archived pages, pre-assembled and mapped to claim limitations with public-availability dates.

+

Custom Technical Request

Any Stage

Not sure module fits? Describe the matter and we will point you to the right one.

Discuss your matter

01

Counter-Argument Diagnostic

Active litigation

Mechanism-level rebuttals to patent owner responses. Stays within the existing record. No procedural exposure from new theories.

02

Non-Infringement Arguments

Active litigation

Limitation-by-limitation analysis at the architecture level. Each distinction supported by evidence of how the product actually works internally.

+

Custom Technical Request

Any Stage

Not sure module fits? Describe the matter and we will point you to the right one.

Discuss your matter

The Associate's Associate.

GreyB is the technical arm. The associate owns every legal decision. We build the record they argue from, at every stage, under every deadline.

01

One team, the full case

The same team that ran the search has reviewed thousands of references. When the case moves to Markman eight months later, they are not starting from zero. Every technical question comes back to people who already know the technology, the prosecution landscape, and the false positive risk.
02

Technical rates, not legal rates

Investigation work and judgment work are not the same thing. When investigation work is done at technical rates, the cost difference on a complex matter is real. That is the efficiency case, independent of the quality argument.
03

No legal opinions, ever

GreyB provides technical analysis only. The associate owns all legal decisions. This removes any unauthorized practice of law concern. E-billing language that has worked: “Technical advisory support” or “Associate-equivalent technical support.”
04

Evidence-first, not assertion

Every analysis is citation-first. Every cell, every term dossier, every counter-argument carries a pin-cite and a written functional basis. Mechanism-level specificity is what makes the record hold up under cross-examination and at institution.
Discuss your matter

What this looks like in practice.

Each case is anonymized. Technology domain and litigation context are accurate. Parties and courts withheld.

Non-infringement

Software / Cloud

Non-infringement arguments

"We believe the product works differently" is not a defense.

A patent claimed “storing user preferences in a local database synchronized with a remote server.” The accused product used local caching. An associate flagged the terminology difference. A technical analyst documented the actual architecture: the product uses a key-value store in volatile memory, not persistent storage; refreshed on each session start, not synchronized; with no schema or query capability. The non-infringement position moved from “we think it is different” to “here is specifically how and why, limitation by limitation.” The argument held under cross-examination because it was grounded in architecture, not in definitions.

Position grounded in architecture, not assertion. Held under cross-examination.

Motion preparation

Automotive / Sensor fusion

Prosecution history insights

The concession was on page 12 of a 30-page response. It controlled the entire case.

A patent on autonomous vehicle sensor fusion had 14 office actions across a 6-year prosecution. In the 9th office action response, the applicant argued that “real-time fusion” requires processing within 50 milliseconds, distinguishing over prior art that processed in batch mode. This argument appeared in a single paragraph. A GreyB analyst flagged it with the exact citation. The attorney drafting the motion knew about this constraint before the other side raised it at the hearing, and framed the argument proactively rather than reacting to it under pressure.

Prosecution constraint surfaced before hearing. Attorney controlled the narrative.

Expert phase

Semiconductors / OLED

Expert support package

Expert billing: 3 hours of opinion formation, not 15 hours of sorting.

In a case involving OLED display manufacturing, the expert needed to opine on whether a specific deposition technique was well-known at the priority date. Instead of receiving 40 references and being asked to find the relevant ones, the expert received a structured package: the three most relevant references mapped to the specific claim elements at issue, key passages highlighted and pin-cited, and the technical evolution from sputtering to thermal evaporation to PECVD documented chronologically with standards references. The expert spent 3 hours reviewing and forming opinions. The citation trail held up when opposing counsel challenged it during deposition.
Expert billing shifted to opinion formation. Citation trail held under deposition.

False positive

Consumer electronics / Media

Reference triage

Both reduced data. Only one was a compressor. The difference mattered at institution.

A patent claimed a “compressor” in a media processing pipeline. A reference described a coding block that also reduced data volume. The surface read: both reduce data, both look equivalent. A technical analyst found the coding block was an entropy coder sitting after a transform stage, reducing statistical redundancy from already-transformed coefficients. The claimed “compressor” sat earlier in the chain, taking raw signal input, removing perceptual redundancy. Different input, different output, different position in the chain. One sentence in the heat map cell prevented a petition ground from collapsing at institution when the patent owner challenged the mapping.

Functional mismatch caught before filing. Ground survived institution challenge.

Claim construction

High-tech / Consumer

Claim term information package

The word "muting" looked settled. Six months later it controlled the entire case.

A claim used the word “muting.” No one flagged it during chart review because it looked like an ordinary English word. Six months later, the entire case turned on whether “muting” meant full silence or volume reduction to a threshold. The specification described muting as reducing audio gain to zero. The accused product implemented muting as attenuation to a threshold. The parties spent significant resources on expert declarations arguing about a term that was never examined at the outset. A Phase 1 claim term review would have flagged this as a flashpoint: ordinary-language term with ambiguous technical scope, specification uses it in context X, field usage varies, flag before positions are locked.
Flashpoint term identified early. Claim construction position built before the dispute arose.

Priority / §101

Software / Display

§101 analysis + priority chain

The claim looked commercially broad. The breadth was coming from under-definition, not from a well-supported disclosure.

A transparent display patent with a broad Markush-style material list included terms like “transparent exotic material” and “transparent invisible metal.” On the surface the claim read as a high-value umbrella. A §101 analysis stripped the device nouns: the claim was directed to collecting data, analyzing a user’s state, and providing feedback using generic sensor components. No specific technological improvement to sensor processing, wearable functionality, or computer operation was recited. The claim breadth came from under-definition of terms, not from an enabled technical disclosure. The §101 ground was available from the first read of the claim. It was identified before prior art charting began.
§101 ground identified before petition strategy was locked. Strongest argument available from day one.

What litigators say when the search is done.

From in-person conversations across 40+ litigation practices in 2025 and 2026.

After all these years, that’s what I value most: good judgment, open communication, and a team that’s genuinely committed to getting the search right.

GreyB has invested in AI and modern search technology, and that’s important. But what impressed me more was that they knew when technology could help and when experience and conversation mattered more. They didn’t force one approach. They used the right approach for the problem.

The team explained why they were searching in certain directions, walked me through the references they were finding, listened to my feedback, and adjusted the search as our discussions evolved. That gave me confidence that they were trying to find the best answer, not just deliver a report.

Data is almost free now. What adds value is the judgment on top of the data. That’s the scarce thing.

Daniel Scola
Managing Partner

That kind of insight only comes from really digging into the file wrappers and understanding what was missing. The motivation to combine was only identified by going into the file wrapper and finding the reference that fills exactly the gap the PTO said was missing.

Jennifer Klein Ayers
Founding Partner

I like the idea that you’re getting nuanced about implicit disclosures and inferences. Because if it’s expressed, it’s easy. It’s almost never fully expressed. That’s the hard part, and that’s where the value sits.

Daniel Scola
Managing Partner

Common questions.

No. GreyB takes the investigation work at technical rates. The associate owns every legal decision. GreyB builds the technical record they argue from. The model is the Associate’s Associate, not the associate’s replacement.
No. GreyB provides technical analysis only, never legal opinions. The service sits in the same category as expert technical consulting that has been embedded in litigation for decades. E-billing language that has worked: “Technical advisory support” or “Associate-equivalent technical support.”

GreyB draws exclusively from publicly available sources: USPTO records, PACER filings, PTAB proceedings, published standards, and archived public materials. Client materials provided for a specific engagement are used only for that matter. Standard NDAs are executed at the start of every engagement.

You pay only for what was delivered. Scope and hours are agreed per module at the start. If the case settles before a stage is reached, that stage is not billed. Pro-rated on delivery. No retainer model. No minimum commitment.

Yes. Every module is available independently. If you need a reference triage before contentions and nothing else, that is a complete engagement. The full-lifecycle model is available for firms that want continuous technical support across the matter, but it is never required.

The team has search and analysis depth in:

Wireless / 5G / 6G

Wi-Fi (802.11)

Bluetooth / IoT

Semiconductors

Consumer electronics

HEVC / AV1 codecs

Software / cloud architecture

Medical devices

Life sciences

Automotive / ADAS

Chemical / metallurgy

For domains outside these, we say so directly rather than take a matter we cannot do well.
At technical rates, not legal rates. Scope and hours are estimated per module at the start of the engagement and confirmed before work begins. Billing is pro-rated on delivery. The conversation about rates is easier around a real case than around a hypothetical, so reach out directly and we will take it from there.
For matters where GreyB already ran the prior art search, work can begin within 24 to 48 hours of scope confirmation. For new matters, we typically need 3 to 5 business days to review the patent and prosecution history before committing a scope estimate.
There is no minimum. A single module on a live matter is a complete engagement. We will tell you honestly if the scope is too narrow to be worth the coordination cost on either side.

Questions we hear often.

We already have prior art results from GreyB. Does Litigation+ require starting over?

No. The team that ran the search already knows the technology and the prior art landscape. Litigation+ begins where the search ends. No cold handoff, no re-learning the case.

Our associates handle the technical work internally. Why would we change that?

The question is not whether your team has the capability. It is whether that capability is available at the depth each stage requires without pulling people off other work. Most firms have one or two strong technical associates stretched across multiple matters. Litigation+ fills the gaps specific questions create without displacing what your team does best.

Tell us about your matter.

The fastest way to know whether Litigation+ fits is to look at one specific case together. We will tell you honestly what we can do and what we cannot.

  • Response within one business day
  • Technical rates, scoped to the specific matter
  • No retainer required. Modular or full-lifecycle.
  • NDA executed at the start of every engagement
  • If the matter is not a fit, we will say so directly

Start a conversation

Share where you are in the matter and we will take it from there.