Intake Is Not an Afterthought: It Is the Foundation of Case Management

Most firms draw a bright line between legal intake and case management. Intake collects the lead. Case management begins after an attorney accepts the matter and someone opens a file.

Before a case exists in the case-management system, the firm may already have a claimant's name and aliases, opposing parties, key dates, signed consent, and supporting documents. Those details already influence conflict review, qualification, urgency, and investigation. They are the beginning of the case record.

The central idea: intake is where a matter's data model begins. If information enters as inconsistent notes, disconnected uploads, or duplicated fields, every later workflow inherits that disorder.

Case management begins before a case is opened

A prospective client arrives with a story, not a schema. A date may appear in both a narrative and a medical record. A brand name may obscure the relevant subsidiary. One file may support several facts, while another contradicts the caller's account.

Good intake preserves the story while giving its operational parts defined places. This becomes especially important in mass tort intake, where one campaign may apply the same criteria and evidence requirements across a large claimant population. At a minimum, the record should distinguish:

1. what the prospective client reported,
2. what a document or other source appears to show,
3. what the firm has reviewed, corrected, or accepted, and
4. what remains unknown or requires follow-up.

Those distinctions matter immediately. The ABA's intake guidance identifies the prospective client's full name and aliases, the opposing party, counsel, location, area of law, and general nature of the matter as information commonly needed for conflict analysis. It also places the conflict check before legal advice or an engagement letter.

The collection boundary matters too. The comments to ABA Model Rule 1.18 say a lawyer may need information to assess conflicts and whether to take a matter, but should limit the initial consultation to what those decisions reasonably require. Intake needs both structure and restraint.

Bad intake creates compounding rework

Weak intake rarely fails in one dramatic moment. It creates small ambiguities that are copied into more places.

An intake specialist records an incident date in a narrative note. A case manager retypes it into a spreadsheet. A paralegal finds a different date in an attachment and updates the case-management system, but not the spreadsheet. The reporting team later sees two values and no record of which source resolved the discrepancy.

Each handoff has added effort while reducing confidence. The same pattern appears across a case:

Intake defect	Immediate consequence	Downstream effect
Names entered differently	Possible matches require manual review	Weaker conflict checks and duplicate claimant records
Dates buried in notes	Staff must read and re-key them	Conflicting chronologies and unreliable deadline workflows
Uploads stored without a type	Evidence cannot be routed consistently	Missing-document requests and repeated review
Blank fields with no status	No one knows whether the answer was skipped or unknown	Generic follow-up and incomplete attorney packages
Free-text claim categories	Similar matters are labeled differently	Distorted reporting, routing, and portfolio analysis

Thomson Reuters describes legal teams moving among spreadsheets, inboxes, and multiple platforms while chasing information and re-keying data. Its guide targets legal departments, but the failure is familiar to plaintiff firms: disconnected systems make staff reconstruct context instead of advancing the claim.

Rework also compounds silently. A duplicate claimant record can split correspondence and documents. A party name without aliases can weaken a conflict search. A missing source link makes a correct field harder to trust. Automation then repeats those weaknesses at greater speed.

A form is not automatically structured intake

A digital questionnaire can improve access and reduce transcription, but it is still an interface. It becomes structured intake only when its fields have durable meaning inside the firm's case model.

Define fields around the case type

The useful unit is a governed set of questions tied to the facts, parties, documents, and decisions relevant to a case type. Product-liability, employment, and motor-vehicle matters may share identity fields, but should not force their material facts into the same notes box.

Use conditional questions

Questions should follow the facts already provided. If the claimant reports treatment, ask for provider and record details. If a product is involved, request the model, purchase information, and available photographs. Conditional logic reduces irrelevant collection and makes missing evidence easier to identify.

Distinguish unknown from unanswered

An empty field is ambiguous. The claimant may not know, may have skipped the question, may intend to return later, or may never have seen it. Those states drive different follow-up. Model them explicitly instead of treating absence as a single value.

Validate without trapping the claimant

Dates, phone numbers, and identifiers benefit from format validation, but rigid rules can reject legitimate exceptions or encourage invented answers. Allow an "unknown" state, a qualified answer, or human review. The goal is dependable data, not a technically satisfied form.

Preserve documents and provenance

Keep the original answer or file. A structured fact should point back to the page, message, or upload that supports it and show whether it was reported, extracted, corrected, or verified. NIST's Data Governance and Management Profile concept paper treats accuracy, completeness, relevance, timeliness, and consistency as dimensions of data quality, and warns that uncertain provenance and lineage magnify risk. Those are general data-governance principles, not legal-specific rules, but they map directly onto trustworthy case records.

Collect in stages

Do not make first contact carry the entire case. Collect what is appropriate for conflict review and initial qualification, then request deeper factual and evidentiary detail as the matter advances. Staging respects the prospective client's time, reduces unnecessary collection, and gives the firm defined checkpoints for consent and attorney review.

The intake record should become the case record

Acceptance should change the record's status, permissions, ownership, and required work. It should not trigger a manual rebuild in a second disconnected system.

The accepted matter should carry forward original answers and uploads, source-linked facts, corrections, consent history, conflict status, review history, and unresolved questions. Matter-opening can add controls, but should not require rediscovery.

This requires one governed source of truth. Other tools may remain useful, but they should reference the same claimant, matter, parties, and accepted facts instead of maintaining competing versions.

The handoff should therefore be a state transition:

• a prospective claimant becomes an accepted client or a declined prospect,
• a proposed fact becomes verified, corrected, disputed, or remains open,
• a received document becomes classified and connected to the requirement it satisfies,
• an unanswered question becomes assigned follow-up, and
• the complete history remains available to the people authorized to review it.

Continuity is what makes the intake record operational. Without it, "integration" may still be a staff member copying fields between screens.

AI makes intake quality more consequential

AI can classify documents, extract candidate names and dates, summarize narratives, and flag apparent gaps. Those actions depend on clear inputs, known case types, defined fields, preserved sources, and firm-approved review rules.

If an upload is unclassified, an extraction has no source link, or the record does not distinguish "unknown" from blank, automation must guess. A polished summary can then hide uncertainty rather than resolve it.

The safer pattern is proposal, review, acceptance:

1. preserve the original submission,
2. generate structured proposals from it,
3. connect each material proposal to its source,
4. route low-confidence, conflicting, or exceptional items to a human,
5. record the accepted value and the review history, and
6. let downstream workflows act on the governed record.

Human review applies judgment, resolves conflicting sources, and controls which information can drive a deadline, communication, or case package. One unchecked ambiguity can otherwise propagate across many automated actions.

What clean intake looks like operationally

Clean intake moves a prospective matter toward review without losing the source, state, or owner of important information.

In practice, that means:

• Staged collection: ask only for what the current decision requires, then deepen the record as the matter progresses.
• Identity resolution and deduplication: search names, aliases, contact details, and existing records before creating another claimant.
• Document classification: preserve each original upload, identify its type, and connect it to the claimant, event, and evidence requirement it supports.
• Required-evidence tracking: show what has been received, requested, waived, or remains missing for the case type.
• Source-linked corrections: keep the previous value, the supporting source, the reviewer, and the reason for the change.
• Clear ownership: every follow-up, exception, and attorney decision has a responsible person and a visible status.
• Exception queues: route conflicts, incomplete records, duplicates, and uncertain extractions to review instead of allowing them to disappear into notes.

Client experience belongs in this model. Clio's 2024 Legal Trends research reported that only 33% of 500 firms in its secret-shopper study responded to email inquiries, while many responses lacked clear next steps. These vendor-reported findings illustrate the cost of weak first contact. A structured record helps a firm explain what was received, what is missing, and who owns the next step.

Metrics that reveal intake quality

Conversion rate alone cannot show whether intake is producing usable matters. A firm should define mass tort intake metrics that expose record quality and operational delay, then monitor:

• Time from submission to first review: how long information waits before a responsible person evaluates it.
• Missing-document rate: the share of matters reaching each stage without the evidence the case type requires.
• Correction and re-keying rate: how often staff must repair or manually re-enter information after initial capture.
• Duplicate rate: how often new submissions match an existing claimant or matter.
• Intake completion rate: where prospective claimants stop and which questions or requests create friction.
• Attorney-ready rate: the percentage reaching attorney review without remediation, duplicate resolution, or basic evidence follow-up.

These measures reveal different failure modes. Fast review with a high correction rate may indicate weak capture. High completion with a low attorney-ready rate may mean the form collects many answers but not the right ones. Useful metrics lead back to the field, rule, evidence requirement, or ownership gap that caused the problem.

Intake is the foundation of case management

Acceptance changes duties, access, ownership, and the work a firm has agreed to perform. It should not sever the information lifecycle.

Intake is the first and cleanest opportunity to establish reliable case structure. The source is still close. Questions can still be clarified. Duplicates can be resolved before records split. Missing evidence can become visible work. Corrections can be linked to the facts that justify them.

Treat intake as an afterthought and case management begins with cleanup. Treat it as the first layer of case management and every downstream team starts from a record it can understand, review, and trust.

OBE is built to carry structured intake, source-backed corrections, evidence requirements, and human review into the case record. Map Your First Case to see what that continuity could look like for one of your firm's workflows.

Book a demo →

Sources

• American Bar Association: How the Legal Client Intake and Conflict Check Process Works
• American Bar Association: Comment on Rule 1.18, Duties to Prospective Client
• NIST: Data Governance and Management Profile Concept Paper
• Clio: 2024 Legal Trends Report findings
• Thomson Reuters: Modernising Legal Intake and Matter Management