--- title: "Fact-Pattern Decision Research" description: "With the Docket Navigator MCP, you can ask your organization's AI provider platform to find decisions based on the factual, substantive, and procedural pattern you are researching." sidebar: badge: text: "Coming Soon" variant: note --- import SkillDownload from '../../../../components/SkillDownload.astro'; ## Find decisions by describing the research problem With the Docket Navigator MCP, you can ask **your organization's AI provider platform** to find decisions based on the factual, substantive, and procedural pattern you are researching. You do not need to determine the right combination of Docket Navigator filters before you begin. Describe the assignment the way you would explain it to another member of your legal team: what happened, what issue matters, what type of ruling you need, which outcomes interest you, and any court, judge, or time period that matters. Your organization's AI provider platform can use the Docket Navigator MCP to translate that request into Docket Navigator's structured patent litigation data and retrieve matching decisions. The goal is simple: **Describe the decisions you need, not the filters you think will find them.** :::note **Important:** The Docket Navigator MCP connects Docket Navigator data and research tools to your organization's AI provider platform. The AI platform you use interprets your request, decides how to use those tools, and generates the response you see. Results may therefore vary depending on the AI provider and model your organization uses. Review important results for factual, doctrinal, and procedural relevance, and use the Docket Navigator citations and links to verify the underlying source material. ::: ## When to use fact-pattern decision research Use this workflow when you want to find decisions based on one or more of the following: - A factual pattern - A substantive or procedural issue - A type of motion or other procedural posture - A particular outcome - A court or judge - A time period - Any combination of these factors You can use familiar legal terminology when you know it, or simply describe the problem. For example: > Find decisions excluding damages opinions for failure to apportion the value of the patented feature. Or: > Find decisions where a damages expert valued the whole accused product even though the patent covered only one component. Both describe a research problem that your organization's AI provider platform can translate into Docket Navigator research. This can also be useful for procedural scenarios that may otherwise require detailed knowledge of Docket Navigator's taxonomy, such as decisions arising from motions to reconsider or clarify or, where represented in the relevant Docket Navigator library, orders entered *sua sponte*. ## Who can use this workflow Fact-pattern research can begin with anyone responsible for the research assignment. A partner or senior associate may use the MCP to explore the issue, identify the first group of authorities, and save the research before handing it to another member of the team. A junior associate, law clerk, paralegal, research attorney, librarian, or knowledge management professional can use the same conversational workflow to continue the assignment without first reconstructing how the original Docket Navigator search was built. This means a supervising attorney can describe the assignment conversationally, begin the research, and save it to a binder. Another researcher can then pick up that binder and continue refining the same body of work. ## How it works Your organization's AI provider platform interprets the request and uses the Docket Navigator MCP to identify the corresponding concepts in Docket Navigator's structured data. Depending on the question, those concepts might include: - The type of motion or decision - The legal issue being decided - The result - Court or judge - Date - Other Docket Navigator criteria relevant to the request This is different from relying only on the words that happen to appear in a document. Docket Navigator's legal-issue coding can help your AI provider platform focus research on the substantive concepts represented in Docket Navigator's structured data. This can be particularly valuable when the same terminology appears across many decisions but the researcher is interested in how a particular issue was treated in a specific factual or procedural context. Returned decision data can also include Docket Navigator annotations describing the court's reasoning, along with links to the underlying Docket Navigator material. Those annotations allow the AI provider platform to evaluate **why a decision is responsive**, rather than relying solely on a keyword match or case name. ## Example: Researching a damages expert's failure to separate the value of a patented feature Assume you are a senior associate **defending a patent infringement case involving a patented feature in a large multi-component industrial machine**. The plaintiff's damages expert calculates a reasonable royalty using revenue from sales of the complete machine but does not separately value what the patented subsystem contributes. You are considering a challenge to the expert's opinion and want to find decisions addressing the same problem. You might ask: > I'm defending a patent infringement case involving a patented feature in a large multi-component industrial machine. The plaintiff's damages expert ran a reasonable royalty off revenue from the entire machine and never separated out what the one patented subsystem actually contributes. I want to move to exclude him. Find me the district court decisions where courts have thrown out or cut back a damages expert for exactly that failure, and show me how the courts described what was wrong with the expert's method. ### What the MCP can help identify Even if the request does not use every relevant term of art, your organization's AI provider platform can use the MCP to recognize the concepts behind the assignment. In this example, the research involves both: - A challenge to expert testimony - Damages apportionment The reference to using revenue from the entire machine may also make the Entire Market Value Rule relevant to later refinement. The important point is not whether the attorney knows those terms. A senior patent litigator probably does. The value is that the attorney does not have to translate the assignment into Docket Navigator's taxonomy, locate the right branches, and combine the corresponding search criteria manually before the research can begin. A less experienced researcher can also describe the underlying facts and reach the same structured research workflow. ## See why a decision matches Fact-pattern research should give you more than a list of case names. For example, one decision that addresses this scenario is *Wirtgen America, Inc. v. Caterpillar, Inc.*, 715 F. Supp. 3d 587 (D. Del. 2024). Instead of simply accepting the case as responsive, you can ask: > Why is the *Wirtgen* decision in here? Walk me through what the expert did wrong. Your organization's AI provider platform can use the Docket Navigator information associated with the decision to explain why it matches the fact pattern and direct you back to the underlying source. This is an important part of the workflow. The objective is not simply to retrieve something that looks similar. It is to help you evaluate whether the decision is actually relevant to the assignment. ## Look for decisions that cut the other way You can also use the same research to identify contrary authority. For example: > Which decisions in this research allowed the expert's opinion, and what did those courts think the expert did well enough? Or: > Show me the cases where the apportionment challenge failed. Seeing both sides can help you understand not only which arguments have succeeded, but where courts have drawn the line between an admissible methodology and one they excluded. You can also ask your organization's AI provider platform to flag decisions with unusual procedural postures or decisions that are adjacent to, but not exactly the same as, your requested fact pattern. ## Refine the research conversationally Once you have a useful result set, continue the assignment without rebuilding the search from scratch. For example: > Show me the more recent decisions where the court actually excluded or limited the opinion. > Within the original results, focus on cases where the expert used the value of the whole product as the royalty base. > Now show me just the District of Delaware decisions. > Which of these decisions is closest to my fact pattern? > What distinguishes the cases where the expert was excluded from the cases where the methodology survived? The AI provider platform can use the existing research as context while applying the additional Docket Navigator criteria needed for each refinement. ## Save the research to a Docket Navigator binder When you have research you want to keep, ask your organization's AI provider platform to save it: > Save this research as a Docket Navigator binder called "Apportionment Daubert Rulings." The binder becomes **a research artifact that you can refer back to at any time**. Binder results update when new Docket Navigator data matches the saved search criteria, so you can return to the binder to review current results. You can preserve the original research criteria and organize narrower questions into additional binder tabs. That makes it possible to maintain, for example: - The broader set of responsive decisions - A recent-decisions tab - A tab focused on the closest factual issue - A forum-specific tab A supervising attorney can also begin the research, save the binder, and share the working artifact with the person who will continue the assignment. Because creating or editing a binder actually changes saved Docket Navigator content, your organization's AI provider platform should do so only when you ask it to save or modify that research. ## Tips for better fact-pattern research ### Describe what happened Facts often communicate the assignment more precisely than a doctrine name alone. Instead of: > Find apportionment cases. Try: > Find cases where the patent covered one part of a larger product and the damages expert calculated the royalty using the value of the whole product without separating the patented feature's contribution. ### Use terms of art when they are useful Conversational research does not mean avoiding legal terminology. A patent litigator can ask: > Find Daubert decisions addressing apportionment of a reasonable royalty. A junior researcher who does not yet know the terminology can describe the same underlying problem. Both approaches can be effective because the MCP is intended to connect the research request to Docket Navigator's structured concepts. ### Tell the platform what kind of decision matters Procedural posture can materially change the usefulness of a result. For example: > I want decisions excluding the expert. > Find summary judgment decisions on this issue. > What happened on motions to reconsider? > Show me decisions from Judge X. The more clearly the assignment describes what matters, the easier it is to evaluate whether the returned decisions are responsive. ### Ask why results are included For important decisions, ask: > Why does this case match? > What did the court actually say about this issue? > Is this the same procedural posture as my case? > Is this directly responsive, contrary authority, or only an adjacent issue? These questions encourage your organization's AI provider platform to use the Docket Navigator information associated with the decision rather than relying on surface similarity. ## Verify important results Docket Navigator provides the structured patent litigation data and source links that your organization's AI provider platform can use, but the AI platform generates the response. Before relying on an important proposition: 1. Review the cited Docket Navigator decision or document. 2. Confirm that the procedural posture matches your research question. 3. Confirm that the cited discussion actually addresses the substantive issue you care about. 4. Check whether a result is favorable, contrary, or only factually adjacent. 5. Use the underlying source material for any proposition that will appear in work product. This verification step is especially important when a research question depends on subtle distinctions in facts, doctrine, or procedural posture. ## Use the attached Fact-Pattern Decision Research skill If your organization's AI provider platform supports skills, use the **Fact-Pattern Decision Research** skill attached to this article. The skill is designed to help the platform: - Recognize this type of research request - Translate natural-language factual and legal descriptions into Docket Navigator research - Evaluate results using Docket Navigator annotations and source information - Preserve citations and links - Distinguish responsive decisions from contrary authority and near misses - Refine existing research instead of unnecessarily starting over - Save and continue research through Docket Navigator binders when requested The skill does not replace legal review. It helps your organization's AI provider platform use the Docket Navigator MCP more consistently for this particular workflow. ## The key idea Fact-pattern decision research changes where the research process starts. Instead of beginning with: **Which Docket Navigator filters should I use?** you can begin with: **What decisions am I actually trying to find?** Your organization's AI provider platform and the Docket Navigator MCP handle the translation into structured research, while Docket Navigator's citations and source links give you a path to review and verify the result.