How Do You Organize Material Support Evidence?
Organize material support evidence in an element-by-element matrix that connects each factual proposition to its original source, attribution basis, context, and unresolved gaps. Use the applicable element checklist supplied by your legal reviewer, distinguish observations from inferences, and preserve contradictory material alongside supporting records. This is an evidence-organization workflow, not legal advice.
What is an element-based evidence matrix?
An element-based evidence matrix is a review document that maps source-backed factual propositions to a supplied legal-element checklist. It shows what a record establishes, what remains inferential, and what still needs verification.
The matrix is not a substitute for the underlying evidence. Its purpose is to make a large collection reviewable without turning an investigative theory into an assumed fact.
Material support investigations may involve communications, transfers, travel records, documents, and account activity. Those categories describe evidence, not legal elements. A payment record, for example, may document a transaction without establishing its purpose or what the sender knew.
Keep the legal framework and the evidence inventory separate, then connect them through specific propositions.
Which legal elements should investigators use?
Start with the precise checklist provided by the legal reviewer for the matter under review. Do not reuse an element list from another investigation merely because both matters involve material support allegations.
This article does not supply a universal legal test. Applicable requirements depend on the legal framework and allegations being assessed.
For organization, investigators can maintain factual review buckets such as identity, conduct, recipient, knowledge, timing, and location. These are review categories, not a statement of required legal elements.
Record the checklist's version and review date. If it changes, revisit existing mappings instead of assuming the earlier assignments remain valid. Preserve the earlier version so reviewers can understand why a proposition was originally placed in a particular row.
What belongs in each material support evidence row?
Use one row per factual proposition rather than one row per file. A single conversation may contain several propositions, while one proposition may require multiple independent sources.
A practical row checklist includes:
- Element reference: The identifier from the supplied checklist.
- Factual proposition: A narrow statement that can be checked against a source.
- Source locator: File identifier, page, message identifier, spreadsheet row, or media timestamp.
- Attribution basis: Why the account, device, transaction, or statement is associated with a person.
- Time information: Original timestamp, timezone, and any normalization applied.
- Context and limitations: Missing messages, translation issues, shared access, or incomplete records.
- Assessment and follow-up: Supporting, contradictory, unresolved, or not yet reviewed, plus the next verification task.
Write “the record shows a transfer between these accounts” before considering a broader interpretation. Avoid replacing observable facts with labels such as “funding operation” unless the underlying proposition and supporting evidence are clearly identified.
How is element-based review different from file-based review?
File-based review remains necessary for understanding source structure and completeness. Element-based review adds a second organizational layer; it should not replace source-by-source examination.
| Review approach | Main question | Useful output | Limitation to watch |
|---|---|---|---|
| File-based | What does this return or extraction contain? | Source inventory and review notes | Relevant facts can remain scattered |
| Person-based | Which records are associated with this person? | Attribution notes and person summary | Account association can be mistaken for authorship |
| Element-based | Which propositions relate to this checklist item? | Source-linked evidence matrix | A theory can bias classification |
Use all three perspectives. A reviewer should be able to move from a matrix proposition back to the full source and then examine other records associated with the relevant person or account.
How do you build the matrix from mixed evidence?
How should you inventory and prepare the sources?
Begin with an intake register that identifies each received file, its origin, receipt date, and relationship to other productions. Preserve originals under your established evidence-handling procedures and distinguish working copies from original material.
Lantern Intelligence is an AI evidence platform for investigators. Lantern accepts any evidence file, including warrant returns from any provider, phone extractions, PDFs, scans, photos, video, spreadsheets, email, bank records, and call records. OCR turns scans, images, and handwriting into searchable text.
Searchability is a starting point, not verification. Check OCR-derived names, amounts, dates, and account numbers against the original image before using them in a proposition. Mark unclear handwriting as uncertain rather than silently selecting a plausible reading.
See how Lantern works for the product workflow.
How should you review and classify the records?
Use a repeatable sequence:
- Inventory the source. Note coverage dates, record types, and apparent omissions.
- Identify candidate passages. Search names, identifiers, amounts, dates, and relevant language without treating keyword hits as conclusions.
- Read surrounding context. Include nearby messages, attachments, transaction descriptions, and source explanations.
- Draft a narrow proposition. Separate the record's contents from your interpretation.
- Map and challenge it. Assign the relevant checklist item, record alternative explanations, and identify needed corroboration.
Lantern's live app includes Upload Case, Dashboard, Messages, Persons, Media, IP / Login Activity, AI Insights, Files, Records, Search, and Export, plus an “Ask the evidence” AI chat. AI Insights provides an intel brief and next steps.
Use those views to review the collection while maintaining your element matrix as a controlled review document. Do not assume a source automatically proves an element because an AI summary groups it under a particular theme. Timeline and Relationship graph are coming soon, not live features.
How do you avoid confusing association with attribution?
Attribution is the documented basis for connecting an action or record to a particular person. A name in an account profile is one observation; authorship of a specific message is a separate proposition.
Record what supports the connection and what could weaken it. Shared devices, delegated account access, forwarded messages, and incomplete login history can affect interpretation.
In a hypothetical review, a conversation discusses an amount similar to a later transfer. That similarity may justify further comparison. It does not, by itself, establish that the same person authored the message, initiated the transfer, or intended the purpose alleged.
Keep the conversation, transaction, and identity propositions separate until the evidence supports connecting them. The related guide, Terrorism Financing Investigation: Following Money Across Borders, addresses the financial-record review context.
How should contradictory evidence and gaps be recorded?
Give contradictory and qualifying records the same source-reference quality as supporting material. Do not bury them in a miscellaneous folder or reduce them to a generic caveat.
A useful gap statement is specific: “The production contains the incoming message but not the attachment it references.” A weak gap statement is “Need more evidence.”
Assign each gap a verification task and an owner. Distinguish material not yet reviewed from material not received and from material whose meaning remains uncertain. Those conditions require different follow-up.
Before handoff, ask another reviewer to test whether each proposition survives reading the complete source. Record disagreements rather than forcing consensus through stronger wording.
What should the final review package contain?
Prepare the current checklist, evidence matrix, source inventory, attribution notes, unresolved-gap register, and references to contradictory material. Keep source identifiers consistent across the package.
For AI-assisted work, retain enough review notes to distinguish machine-generated suggestions from investigator-verified findings. An AI summary is a navigation aid, not a replacement for checking the original record.
Lantern runs in private AWS or on-prem on Google Gemma 4; evidence is never sent to an outside AI service. CJIS, FedRAMP, and SOC 2 are coming soon. Evaluate deployment requirements without treating those future milestones as current certifications.
Explore the use cases to understand the review experience, and visit pricing for sign-up information. Free sign-up includes 10 credits.
Put this to work on your own case
Sign up free and get 10 credits — enough to process a typical warrant return and see the messages, people, AI summary and recommended next steps.