# How Should Counsel Build Evidence for SEP Royalty Negotiation in 2026?

iprs.cloud · September 25, 2026

> Direct Answer Counsel should build a SEP royalty-negotiation record that can support both the commercial proposal and, if necessary, later FRAND...

## Direct Answer

Counsel should build a SEP royalty-negotiation record that can support both the commercial proposal and, if necessary, later FRAND proceedings. The evidence file should connect the asserted patents to a specific standard feature, show the implementer’s technical use and volume, establish a defensible royalty base, and justify the proposed rate through comparable licenses and economic analysis. It should also preserve failed settlement communications without turning ordinary negotiation into an admission that a particular rate, jurisdictional framework, or licensor interpretation is legally correct.

**Also worth reading:** [What is the definitive IP SaaS contract negotiation checklist for legal counsel and product teams in 2026?](https://iprs.cloud/knowledge/what_is_the_definitive_ip_saas_contract_negotiation_checklist_for_legal_counsel_and_product_teams_in_2026.php) · [What Is the Real Blockchain IP Registry Cost Comparison in 2026, and Which Option Gives Counsel the Best Evidence per Dollar?](https://iprs.cloud/knowledge/what_is_the_real_blockchain_ip_registry_cost_comparison_in_2026_and_which_option_gives_counsel_the_best_evidence_per_dollar.php) · [What is an AI model provenance evidence bundle and why do IP counsel need one?](https://iprs.cloud/knowledge/what_is_an_ai_model_provenance_evidence_bundle_and_why_do_ip_counsel_need_one.php)

The central discipline is to maintain two records at once: a transaction record designed to make the offer understandable and acceptable, and a litigation-readiness record designed to test whether every material assertion can be proved. As of 26 September 2026, that matters because SEP disputes are no longer confined to a binary choice between withholding a license and suing. China’s evolving royalty-determination practice, the Delhi High Court’s treatment of SEP enforcement in Bansal v. Philips, and continuing global FRAND case law all place greater pressure on parties to articulate their positions with evidence rather than generalized assertions about essentiality, non-essential patents, or fair remuneration. No file is “winning” merely because it is large; the better file is organized around the exact propositions the negotiating or adjudicating body must decide.

## What Counts as SEP Royalty Negotiation Evidence?

Useful evidence falls into four connected categories: technical evidence, market evidence, legal evidence, and negotiation evidence. Technical records identify the relevant standard version, claim language, and implementation mapping. Market records establish product shipments, installed base, net sales, geographic coverage, and the products that contain the patented technology. Legal records address the essentiality claim, FRAND obligations, jurisdiction, and the treatment of non-essential portfolio elements. Negotiation records document proposals, counteroffers, comparisons, and the parties’ stated positions.

The most persuasive evidence is not one category in isolation. A claim chart may establish that a patent reads on a standard feature, but it does not establish the number of products that use that feature or the appropriate royalty base. Sales records may show substantial revenue, but they do not prove that the asserted patent is essential. A license comparison may refer to a similar technology, but differences in portfolio scope, geographic coverage, patent life, or exclusivity can make the number misleading. Counsel should therefore create an indexed evidence map in which each factual conclusion points to a dated source document and identifies the inference that the source supports.

Evidence quality also depends on what can be independently verified. Underlying source files, native spreadsheets, system audit trails, and authenticated technical records are generally stronger than screenshots, summaries, or declarations prepared after the dispute. A declaration from an engineer may be useful, but it is stronger when accompanied by source-code versions, test results, specifications, or a demonstrable testing protocol. Similarly, a royalty model should distinguish assumptions from observed inputs, state its currency and valuation date, and show how errors or alternative assumptions affect the result. The objective is not to create maximum volume; it is to reduce avoidable uncertainty.

## Building the Technical and Essentiality Record

The technical file should begin with a precise definition of the technology at issue. Counsel should identify the standard, its publication date, the relevant version, the feature or interface involved, and the patent family under assertion. The analysis should then map the standard’s requirements to the patent’s limitations, while avoiding the common error of treating similarity, compliance, or technical importance as proof that the patent is necessarily essential. A standard may define a technical result while leaving several possible implementation paths open, and a patent may describe one of those paths without covering every compliant product.

For each asserted right, the file should include a claim chart, prosecution history, standards declaration or commitment record where available, and an explanation of any design-around identified by the implementer. The claim chart should identify the exact claim element, the corresponding standard text or proposed feature, the implementation evidence, and the unresolved factual questions. Where validity or enforceability may be disputed, the team should record the relevant statutory and case-law positions without assuming that later amendments, claim construction, or jurisdiction-specific outcomes are global constants.

A practical threshold is to require at least one independent technical confirmation for every material mapping. Depending on the technology, that confirmation could be a laboratory test, a protocol trace, an interoperability requirement, or a signed expert report. The exact number of confirmations cannot be prescribed universally, because some standards are narrow while others involve complex software and hardware interactions. Counsel should document why a proposed test is technically appropriate and what result would support or undermine essentiality. A weak test that merely demonstrates one implementation should not be presented as proof that every product in a broad product category is infringing.

## Establishing the Royalty Base and Rate

A royalty proposal is stronger when it explains both the denominator and the rate. The base might be net sales of identified products, the portion of a product reasonably linked to the patented technology, or another measure permitted by the governing law and licensing record. Counsel should not silently move between gross revenue, net sales, component value, or an arbitrary percentage of total product value. Each choice can materially change the result, particularly when a product includes unrelated hardware, services, data, or post-sale revenue.

The file should preserve source invoices, revenue schedules, unit data, discounts, taxes, and the definition of “net.” Where a product has a transparent price breakdown, a component-based calculation may be more defensible than applying a rate to total revenue. Where the connection cannot be separated cleanly, the party should disclose the assumption and show a sensitivity analysis. For example, if the proposed rate is 2% of identified net sales, the model should show the outcome at 1%, 1.5%, and 2.5%, and explain how those alternatives affect the aggregate amount. Specific numbers are not universally appropriate examples, but explicit ranges make the assumptions auditable.

Rate comparisons should be normalized. Two licenses may appear to have the same percentage but differ in duration, territory, portfolio size, grant-back rights, royalty base, advance, audit rights, or treatment of affiliates. A comparable license is strongest when it involves a similar technology, comparable patent strength, overlapping markets, and a similarly situated licensor and licensee. Where no close comparator exists, counsel should explain the method used to adjust the evidence. The WIPO discussion of FRAND economics, including bottom-up valuation methods, provides a useful reason to present the calculation as a reasoned model rather than a single unsupported number.

## Using Comparables, Economics, and Legal Authority

A negotiation package should contain a curated set of comparables, not an indiscriminate collection of license announcements. Each comparison should state the publication date, parties, technology field, countries, portfolio scope, term, exclusivity, royalty base, rate, and any available payment structure. The team should classify comparables by similarity and explain whether the proposed rate derives from the closest cases, a broader benchmark, or a bottom-up valuation. A single outlier should not control the result, and a settlement reached under urgent commercial pressure may be informative without being determinative.

Legal authority should be organized around the propositions being established. The file may need materials addressing essentiality, FRAND commitments, jurisdiction, patent exhaustion, applicable law, damages, and remedies. The IAM guide on China’s evolving royalty determination and the WilmerHale FRAND Quarterly are useful starting points for current comparative research, but neither replaces primary authority or jurisdiction-specific analysis. Similarly, the reported Delhi High Court position in Bansal v. Philips should be examined through the actual judgment and subsequent treatment before being cited for a broad proposition.

The evidence should be candid about uncertainty. A negotiated rate can reflect commercial risk, timing, and portfolio breadth as well as a purely economic valuation. That is not automatically improper, but it should not be misrepresented as a judicial determination. Counsel should distinguish historical license evidence, party admissions, internal business analysis, and publicly reported positions. The Federalist Society discussion warning about deterioration of appropriate remedies in patent disputes is relevant to strategic planning because threats of exclusion or other extreme remedies can affect settlement leverage, but it should be used as analysis rather than as proof of a particular party’s conduct.

## Practical Steps Before and During Negotiation

Before opening negotiations, counsel should assemble a short evidence memorandum identifying the patents, products, jurisdictions, proposed royalty base, and unresolved factual issues. The memorandum should be revised as discovery, testing, and sales analysis continue. A dedicated evidence register should record the source, date received, custodian, authenticity status, relevant proposition, and confidentiality designation. This prevents an apparently strong number from losing credibility because the underlying spreadsheet cannot be produced or explained.

During negotiation, each proposal should be accompanied by a concise basis. If the licensor offers 1.5% of net sales, it should state whether that rate reflects a comparable license, a bottom-up valuation, or a settlement discount. If the implementer challenges essentiality, counsel should request specific technical evidence rather than responding only with a renewed assertion. If the parties disagree about the royalty base, the negotiation can use staged alternatives, but the agreement should define the products, revenue deductions, reporting cadence, currency, interest, and audit procedure in plain language.

A written record of communications is important. Emails should distinguish proposals from admissions, and meeting notes should identify who made each statement and whether it was qualified. Counsel should preserve drafts when they illuminate the parties’ assumptions, but not selectively omit inconvenient material. Internal legal advice may be protected in some circumstances, yet later litigation can require a careful privilege review. The safest approach is to maintain separate legal-strategy files from factual evidence files and to avoid copying legal conclusions into technical or financial records unnecessarily.

A reasonable review cadence is monthly during active negotiations and quarterly for a dormant portfolio, subject to litigation deadlines, product launches, and changes in sales. The 90-day interval is a management convention, not a legal rule. If revenue materializes quickly, the team may need weekly updates; if the product is not yet commercial, a quarterly review may be excessive. The trigger for updating should be a changed fact, not merely a desire to increase the number of documents.

## Comparing Negotiation and Litigation Evidence Strategies

| Feature | Negotiation-first strategy | Litigation-readiness strategy |
| --- | --- | --- |
| Primary goal | Reach a defensible and timely agreement | Preserve proof for a court or arbitral tribunal |
| Evidence depth | Curated package with clear commercial logic | Broad, indexed record with authentication and chain of custody |
| Royalty analysis | Normalized comparables and practical alternatives | Fully reproducible model with sensitivities and legal support |
| Technical proof | Selected mappings sufficient to support negotiation | Complete testing, expert analysis, and rebuttal material |
| Settlement communications | Clear proposals without unnecessary admissions | Precisely documented positions and controlled distribution |
| Main weakness | May omit issues later needed in court | Can consume resources and delay commercial resolution |

A negotiation-first strategy is often appropriate when the patent position is stable, the products are identifiable, and the parties want a transaction quickly. Litigation readiness is more important when essentiality, validity, infringement, or the royalty base is genuinely disputed. The two strategies are not mutually exclusive. A strong negotiation package should contain the core of a litigation-ready record without producing every internal document at the table.
Cost should be treated as a portfolio decision rather than a fixed fee for one report. A focused desk review of public records and existing internal documents may cost substantially less than a multi-month technical validation and royalty-forensics project, but the figures depend on scope, jurisdictions, experts, testing, and data volume. A fixed estimate should define assumptions, number of patent families, number of product versions, included jurisdictions, and whether experts or court-admissible declarations are included. Counsel should not publish an artificial universal price range because SEP work varies so widely. Instead, a request for proposals should separate basic document review, technical analysis, financial modeling, negotiation support, and litigation-grade authentication.

## Common Mistakes That Weaken the Record

The most frequent mistake is treating a declared standard commitment as conclusive proof that every asserted patent is essential. A commitment and claim mapping may be important, but the scope of the commitment and the breadth of the claims still require analysis. Another error is using a single comparable license without adjusting for portfolio differences. Comparing a narrow, exclusive, single-country license to a worldwide, multi-patent agreement can produce a misleading rate even when both agreements state 2%.

A second common mistake is mixing revenue definitions. Applying a net-sales rate to gross receipts, then comparing the result with a component-based benchmark, can create a disputed result that has little to do with the underlying technology. Teams also overstate what engineering evidence shows. A test on one product version does not necessarily establish infringement across an entire family, and a standards-compliance document may establish conformity rather than a particular patent claim’s coverage.

The record can be weakened by hindsight. Analysts should preserve original source files, model versions, assumptions, and change histories instead of presenting a newly prepared calculation as though it had existed throughout the negotiation. Late-added “supporting” declarations should be identified as such and tested for consistency with earlier statements. Finally, counsel should avoid assuming that a high royalty amount proves reasonableness. A large demand may reflect an overbroad base, an unsupported rate, or a portfolio whose essentiality remains uncertain. Conversely, a smaller settlement may be commercially rational if it resolves several rights efficiently, but the rationale should be recorded.

## When to Act and What to Preserve

Act before the first formal proposal if the licensor cannot clearly identify the relevant standard, patent, product, or royalty base. If the asserted patent is approaching a material limitation or expiry milestone, the team should review the implications for the offer and any ongoing license. In many licensing programs, a six-to-twelve-month planning horizon is sensible, but no universal deadline applies. The immediate triggers are a product launch, a major license expiration, new standards adoption, a customer indemnity demand, a change in sales territory, or a threatened enforcement action.

Preservation should cover technical repositories, standards records, source-code snapshots, product bills of material, sales and accounting systems, licensing agreements, board or committee materials, and relevant communications. Counsel should follow the applicable jurisdiction’s preservation obligations and avoid deleting data once a dispute is reasonably anticipated. Where litigation is possible, forensic collection, metadata, and chain of custody may matter more than the number of emails collected. For ongoing negotiations, a controlled data room with role-based access can reduce both confidentiality risk and inconsistent statements.

The final package should include an executive summary, a patent-to-product map, a royalty-base definition, a reproducible rate model, a comparable-license table, a list of assumptions, an evidence index, and a schedule of unresolved questions. It should be updated when a fact changes and reviewed by both technical and financial subject-matter experts. The goal is not to prove in advance that one side will win. The goal is to make the parties’ positions measurable, distinguish genuine disagreement from avoidable ambiguity, and give any future decision-maker a clear account of how the proposed rate was reached.

## Bottom-Line Evaluation of the Evidence Package

A credible SEP royalty-negotiation file answers five questions: which patent is implicated, why it is treated as essential, what commercial activity is covered, how the rate was calculated, and why the chosen base is legally and economically appropriate. It also identifies what remains unknown. That candor is especially important in 2026, when royalty determination is becoming more fact-intensive and cross-border disputes are drawing on different legal and economic assumptions.

The package should not be judged by document count, the apparent certainty of a headline rate, or the number of cited cases. It should be judged by authenticity, relevance, reproducibility, and the quality of the link between evidence and proposition. A smaller, well-supported record can be more useful than a larger collection containing inconsistent definitions and unexplained assumptions. For counsel and product teams, the practical standard is whether another competent reviewer can follow the chain from the standard and product evidence to the proposed royalty and settlement position. If that chain is incomplete, the next step is targeted testing or data collection rather than a stronger rhetorical assertion.

## Quick answers

### What is the most important evidence in an SEP royalty negotiation?

The most important evidence is the link between the asserted patent, the relevant standard feature, and the commercial products allegedly using that feature. Reliable royalty-base data and a normalized comparable-license analysis are also essential. No single document proves the entire case.

### How should counsel compare SEP license rates?

Counsel should normalize rates for territory, portfolio size, exclusivity, term, royalty base, advance payments, and included technology. A percentage copied from a superficially similar agreement can be misleading. The comparison should state both the similarities and the material differences.

### Is a standards declaration enough to prove SEP essentiality?

It is generally not enough by itself. It may establish a commitment or technical relationship, but counsel should still analyze the claims, standard version, and implementation. Technical testing or expert analysis can strengthen or contradict the mapping.

### When should a company prepare a SEP negotiation record?

Preparation should begin before the first formal offer, especially if products, patents, or licenses are approaching a material change. A six-to-twelve-month review horizon can be useful, but launches, expiry dates, customer disputes, and enforcement threats can require faster action.

### How much does an SEP royalty evidence review cost?

There is no responsible universal price because scope, patent count, jurisdictions, testing, and expert requirements vary sharply. A basic review of existing documents is usually less expensive than litigation-grade technical and financial work. Any estimate should separate document review, technical analysis, modeling, negotiation support, and expert declarations.

Canonical: https://iprs.cloud/knowledge/how_should_counsel_build_evidence_for_sep_royalty_negotiation_in_2026.php
Markdown: https://iprs.cloud/knowledge/how_should_counsel_build_evidence_for_sep_royalty_negotiation_in_2026.php/index.md