How to build a case law research matrix for Insurance Bad-Faith Matters

Build an insurance bad-faith research matrix by defining the legal questions, selecting authorities for the governing jurisdiction and claim posture, and recording each decision’s holding alongside its factual limits and subsequent treatment. Connect each proposition to the policy and claim-handling record rather than collecting favorable quotations. The deliverable is a source-linked table showing which arguments the authorities support, which decisions require distinction, and what legal or factual verification remains before counsel drafts a motion, advice memorandum, or settlement analysis.

When to use this workflow

Use this workflow when counsel must evaluate a denial, delayed investigation, disputed valuation, or asserted defense and turn preliminary research into a supported written position. It serves policyholder counsel, insurer counsel, and in-house teams reviewing outside counsel’s analysis.

Define the forum, proposed governing substantive law, policy type, and first-party or third-party posture before collecting cases. For third-party matters, identify the challenged conduct—for example, defense handling or settlement decisions—rather than treating every liability-policy dispute alike. If governing law is contested, maintain separate research tracks.

This guide uses a California first-party example. Federal procedural sources apply within their federal district-court scope, not automatically in state proceedings or arbitration. The research cutoff is October 13, 2026; that date does not certify subsequent treatment of every authority. This is a general professional workflow, not a nationwide bad-faith survey.

What you need

Gather enough information to formulate questions without treating allegations as established facts:

• The complete policy and endorsements, identifying policy periods and relevant coverage provisions.
• Pleadings or a concise statement of proposed claims, defenses, requested relief, and the immediate research assignment.
• Denial and reservation correspondence, investigation requests, expert reports, payment records, and material settlement communications.
• A chronology separating event dates, receipt dates, and later claim-file entries.
• Forum information, applicable orders, and unresolved choice-of-law questions.
• Full opinions, subsequent-treatment research tools, and historical legal materials relevant to the challenged conduct.

Give documents stable identifiers. DEN-01, p. 2 is more useful than “denial letter.” Identify missing records explicitly: an absent medical report may change both the factual comparison and the next research question.

Use counsel’s issue list to distinguish coverage interpretation, allegedly unreasonable handling, causation, damages, and defenses. These are research categories, not a universal statement of bad-faith elements. Keep remedy questions separate when the assignment requires them; an investigation decision may not address the requested damages.

Use the template

The companion template is Insurance Bad-Faith Matters — Case law research matrix. You can copy this table into your working document or spreadsheet. Use one row per material proposition, even when several rows concern the same opinion.

Authority / pinpoint contains the full citation, direct source link, and supporting page or paragraph. Jurisdiction / status identifies the court, relevant publication or precedential status, and its proposed role in the actual forum.

Policy / posture records distinctions such as underinsured-motorist coverage and summary judgment. Holding / limit separates the court’s decision from broader discussion and counsel’s inference. Facts / record pairs legally significant similarities and differences with document citations and unresolved gaps.

In Adverse / treatment, record contrary authorities, the treatment-search date, and unresolved questions—not merely “good law.” Action / owner assigns the next task and states whether the row is usable, conditional, or excluded. Stable row IDs preserve references when the table is sorted.

Step by step

1. Convert the assignment into bounded legal questions

Replace “research insurer bad faith” with questions tied to counsel’s immediate decision. For example: Does an asserted genuine dispute justify summary judgment where the insurer allegedly failed to investigate conflicting medical information? Which facts distinguish the authorities supporting that defense?

Separate the merits question from discovery needed to establish the investigation record. Also distinguish liability, defenses, and remedies. Write a brief scope statement identifying the intended work product and excluded issues so the matrix does not become a collection of unrelated insurance decisions.

Record: Enter the row ID, narrow issue, intended use, and governing-law uncertainty. Identify additional issue rows needed before the assignment can be answered.

2. Establish each authority’s jurisdiction and role

Read the full opinion rather than relying on a search-result summary or headnote. Identify the court, decision date, policy type, procedural posture, disposition, and relevant precedential status. Assess its role for the actual forum; online availability does not establish authority weight.

Keep case-law rows distinct from rules and agency guidance. Supporting materials can remain in the same matrix if clearly labeled, but they should not obscure the opinions answering the substantive question. Trace a proposition to the deciding court’s analysis rather than attributing a quoted party argument or dissent to the majority.

Record: Complete the jurisdiction, status, and posture fields. Label noncase sources by function, and flag unresolved questions about authority weight.

3. Extract the holding without enlarging it

Write a narrow paraphrase with a verified pinpoint. In Wilson v. 21st Century Insurance Co., the California Supreme Court explained that the genuine-dispute doctrine does not excuse thorough and fair investigation. It concluded that the record presented a triable issue concerning denial without further medical investigation. See Wilson, 42 Cal.4th 713, 721–724. (law.justia.com)

The decision concerns California first-party underinsured-motorist coverage at summary judgment. It affirmed reversal of judgment for the insurer; it did not finally adjudicate bad-faith liability. Record those limits alongside the proposition rather than relegating them to a general disclaimer. See Wilson, opinion and disposition. (law.justia.com)

Record: Enter the proposition, pinpoint, disposition, and limitation on its proposed use. Split separate holdings into separate rows when their relevance differs.

4. Compare investigation facts, including unfavorable ones

Identify the investigation choices relevant to the analogy: information available at denial, reports requested or received, unanswered questions, supervisory review, and the stated reason for stopping inquiry. Separate contemporaneous knowledge from later explanations.

Test the other side’s strongest distinction. A similar denial may follow materially different investigation steps. Conversely, matching policy categories does not establish matching facts. Link each comparison to a document passage and distinguish “not produced” from “did not occur.” Do not infer an insurer’s receipt of medical information solely from the sender’s copy.

Record: Add favorable and unfavorable comparisons, document IDs, and gaps. State whether each gap limits reliance or requires another research branch.

5. Search adverse authority and subsequent treatment

Search the precise proposition as well as the case name. Look for decisions narrowing the reasoning, distinguishing investigation facts, or addressing another procedural stage. Read relevant citing opinions; a research-service signal does not explain their effect on the proposed argument.

Create a separate row for each material adverse decision and cross-reference the affected proposition. If treatment remains unchecked, make reliance conditional. Record the search scope and date so another lawyer can identify what was actually examined, rather than assuming an exhaustive review.

Record: Enter authorities reviewed, their effect, and unresolved treatment questions. Assign a reviewer and identify whether completion is required before drafting or circulation.

6. Separate substantive standards from supporting materials

The California Department of Insurance’s disaster-adjusting guide describes section 2695.7(d) as requiring diligent, thorough, fair, objective investigation and limiting persistent requests for unnecessary or immaterial information. Treat it as a property-claims research lead; verify applicable regulatory text and separately research any asserted private remedy. See the 2026 guide, printed page 19. (insurance.ca.gov)

A missing investigation report may instead create a procedural research question. In federal district court, Rule 26(b)(1) limits discovery to nonprivileged matter relevant to claims or defenses and proportional to the case’s needs; discoverability does not determine admissibility. See Federal Rule of Civil Procedure 26(b)(1). (uscourts.gov)

Record: Label each source’s function. Keep regulatory applicability, remedy authority, and forum-specific discovery questions separate from the substantive case-law conclusion.

7. Turn the matrix into a bounded research conclusion

Group completed rows by issue, not by favorable and unfavorable outcomes. For each issue, synthesize the strongest authority, strongest distinction, and unresolved factual dependency. Identify questions for which the collected cases do not provide an answer.

Use dispositions such as “usable for narrow proposition,” “treatment review required,” or “different policy context; analogy unproven.” Avoid numerical scores that conceal the reason an authority matters. A colleague should be able to retrace the conclusion without repeating the entire search.

Record: State each row’s proposed use and limitation. Assign open issues to team members and identify which prevent a final conclusion.

Worked example

Fictional scenario: Alder Shore Counsel represents Maya Rowan in a California first-party underinsured-motorist dispute against Harbor Lantern Insurance. The action is pending in California state court, and counsel is preparing to oppose the insurer’s summary-judgment motion. The parties, documents, and facts below are entirely fictional; the cited authorities are real.

The denial, DEN-01, p. 2, says the submitted material does not connect continuing symptoms to the collision. A diary entry, LOG-04, p. 6, refers to requesting an outside medical review. No resulting report appears in the produced file. An email, MED-03, p. 1, transmits a treating physician’s addendum, but the available record does not establish the insurer’s receipt before denial.

Counsel must decide whether the investigation record supports a Wilson analogy. The matrix also preserves rejected research leads so they are not inadvertently restored as support during drafting.

The matrix narrows the proposed memorandum. Missing production cannot establish that no medical investigation occurred. The physician’s addendum cannot support a predenial-knowledge argument until receipt is established. Counsel can develop a conditional Wilson comparison, but must examine adverse treatment and investigation evidence before presenting it as the answer. The guide and federal rule remain excluded leads, not substitutes for applicable authority.

Common mistakes

• Treating coverage breach as the whole analysis. Separate coverage interpretation from challenged handling, and research their relationship under the selected law.
• Collecting favorable language without the disposition. Identify what the court decided and match the proposed argument to that procedural posture.
• Mixing first-party and third-party cases. Explain why the precise proposition transfers despite differences in duties and handling context.
• Using guidance as private-remedy authority. Label its limited function and research the remedy separately.
• Converting missing records into negative facts. Identify the evidence gap and investigate it before asserting that an investigation step never occurred.
• Hiding adverse treatment in notes. Give material contrary decisions their own rows and state how they affect reliance.

Final quality check

Before circulating the matrix, confirm that:

• Every material proposition has a full-source link and verified pinpoint.
• Governing law, forum, policy type, and handling posture are explicit.
• Holdings are separated from broader discussion, dissents, and counsel’s inference.
• Procedural outcomes support the proposed use.
• Favorable and adverse factual comparisons have document citations.
• Subsequent treatment is checked or reliance is expressly conditional.
• Rules and guidance are not presented as cases establishing liability.
• Open issues are resolved, excluded, or assigned with an identified consequence for drafting.

Complete the workflow with DecoverAI

DecoverAI is the Legal Intelligence Platform for litigation and transactions. Its document review, chronology, and evidence-analysis capabilities can support organizing the claim record used to test case comparisons. Explore Chronology Viewer and litigation preparation capabilities for that factual work; counsel retains authority selection, subsequent-treatment research, and legal conclusions. (decover.ai) To discuss organizing claim-file evidence alongside this matrix, book a demo.

Frequently asked questions

Should one opinion occupy one row?

Not necessarily. Separate propositions when they have different pinpoints, limitations, or factual applications. Repeat the source reference so each row remains understandable after sorting.

Can property-insurance cases support a UIM argument?

Treat the comparison as an analogy to justify, not an automatic transfer. Explain why the policy difference affects—or does not affect—the specific investigation proposition being researched.

What if a missing report could change the analysis?

Make the comparison conditional. Identify the record suggesting the report exists, what remains unknown, and who will investigate. Preserve alternative conclusions rather than assuming favorable contents.

Is the matrix ready to attach to a filing?

Not by default. It is an internal research work product containing assessments and open questions. Prepare filing materials separately, verifying citations, factual support, confidentiality, and forum requirements.

Sources

• California Supreme Court: Wilson v. 21st Century Insurance Co., 42 Cal.4th 713, 721–724; primary opinion reproduced by Justia.
• California Department of Insurance: 2026 disaster property-claims adjusting guide, printed page 19.
• United States Courts: Federal Rules of Civil Procedure, Rule 26(b)(1).

This guide is general information about legal workflows, not legal advice for any matter. It was drafted and reviewed with AI assistance. The worked example is fictional. Contract rights and procedure depend on the governing law, the actual wording, and the facts, so check the cited sources and the rules that apply in your jurisdiction before relying on it.