The intake agent is the first filter in a PI law firm's case pipeline. When they apply that filter well, attorneys receive a consistent stream of pre-qualified, retainable cases. When they apply it poorly, attorneys spend time on calls that should never have been scheduled — and turn away cases that could have been won.
Most intake training programs focus on script adherence and call handling. Qualification criteria — the framework agents use to decide whether a case merits attorney review — are typically undertrained. The result is significant variance across agents, unpredictable case quality, and persistent disagreements between intake and the legal team about what "qualifies."
This guide covers how to build a qualification rubric, train agents to apply it consistently, and measure performance in a way that actually improves decision-making over time.
Why Qualification Training Gets Skipped
Qualification decisions feel subjective. Unlike script adherence (did the agent ask all required questions?) or call handling (was the caller treated professionally?), qualification requires judgment — and judgment is harder to teach and harder to score.
The result is that most intake managers default to teaching agents what questions to ask, without teaching them how to evaluate the answers. Agents learn to complete the intake form accurately. They don't learn to assess the case.
This leads to three predictable failure modes:
1. Leniency Drift
Agents gradually accept more marginal cases over time. Declining a caller feels confrontational. Accepting the case and letting the attorney decide feels safe. Over months, acceptance thresholds drift down — and attorney review queues fill with cases that have no realistic path to retainer.
2. Conservatism in Unfamiliar Case Types
Agents trained primarily on standard auto accident intake frequently decline or undervalue cases they don't recognize: slip and falls with unusual circumstances, commercial vehicle accidents where the liability picture is complex, product liability claims that present as injury calls. The unfamiliarity reads as a qualification problem, when it's really a training gap.
3. Inconsistency Across Agents
Without a shared rubric, different agents make different decisions on the same call profile. An agent who worked at a high-volume intake operation accepts more aggressively. An agent with a legal background declines more conservatively. Neither approach matches the firm's actual criteria.
The Four-Factor Qualification Rubric
Virtually every PI case qualifies or disqualifies on the same four factors. A rubric that scores each factor gives agents a consistent decision framework and gives managers a basis for calibration and QA.
Factor 1: Liability (1–3 points)
Is there an identifiable at-fault party with arguable legal responsibility? Liability is the foundational question — without it, nothing else matters.
| Score | Liability Signal |
|---|---|
| 3 | Clear liability (rear-end collision, red light violation, employer liability with police report) |
| 2 | Arguable liability (intersection dispute with witnesses, premises with notice evidence) |
| 1 | Disputed or unclear liability (single-vehicle, no police report, caller may share fault) |
Factor 2: Damages (1–3 points)
Did the caller sustain documented injury requiring medical treatment? Damages are the economic engine of the case.
| Score | Damage Signal |
|---|---|
| 3 | ER visit, hospitalization, surgery, fracture, or confirmed specialist treatment ongoing |
| 2 | Urgent care or primary care treatment, physical therapy, documented lost wages |
| 1 | No treatment yet or only one visit with pain complaints not yet diagnosed |
Factor 3: Causation (1–3 points)
Is there a clear connection between the accident and the injuries claimed? Causation problems surface later in litigation and reduce case value — they should be flagged at intake.
| Score | Causation Signal |
|---|---|
| 3 | Injuries are consistent with accident mechanism, treatment started within 72 hours, no relevant pre-existing condition |
| 2 | Treatment delayed but explainable (uninsured, afraid), pre-existing condition that was aggravated |
| 1 | Significant gap in treatment, injuries inconsistent with described accident, prior claim for same area |
Factor 4: Collectability (1–3 points)
Is there a source of recovery? A strong liability case with real injuries means nothing if there's no money to collect.
| Score | Collectability Signal |
|---|---|
| 3 | Defendant has confirmed liability insurance, commercial carrier, government entity, or employer liability |
| 2 | Caller has uninsured/underinsured motorist coverage, defendant's insurance status unknown but probable |
| 1 | Defendant confirmed uninsured with no known assets, caller has no UM/UIM coverage |
Total scores range from 4 to 12. A firm's specific thresholds will vary, but a common calibration is: 10–12: accept, 7–9: escalate to supervisor, 4–6: decline. This forces explicit discussion of borderline cases rather than agent-level guessing.
Implementation note: The rubric score should be recorded in the intake CRM alongside the intake notes. This creates a reviewable audit trail when attorney and intake disagree on a case, and allows managers to pull decision data by agent, case type, and time period.
Training Agents to Apply the Rubric
A rubric is only useful if agents understand how to score it reliably. Abstract criteria don't produce consistent decisions. Concrete examples do.
Phase 1: Case Type Orientation (Week 1)
Agents should begin with case type training before touching the rubric. Every case type that will come through the intake line (standard MVA, commercial vehicle, slip and fall, premises liability, dog bite, product liability, workplace injury) deserves its own overview: how liability typically presents, what injuries are common, what evidence is usually available, what collectability issues arise.
This prevents the conservatism-in-unfamiliar-case-types problem. An agent who has been briefed on what a commercial vehicle case looks like won't decline it because the call doesn't sound like a standard car accident.
Phase 2: Scored Call Reviews (Weeks 2–3)
Agents listen to recorded calls — a mix of accepted cases, declined cases, and borderline escalations — and score each one using the rubric before hearing the actual outcome. The trainer then reviews the scoring together with the agent, explaining where their assessment differed and why.
This training method builds calibration faster than any other approach. Agents learn not just the criteria but how to weigh ambiguous information, how to probe for additional facts, and what language patterns indicate issues that should change the score.
Phase 3: Live Supervised Intake (Weeks 4–6)
Agents take live calls with a supervisor monitoring and available to step in. After each call, the agent completes the rubric scoring and explains their decision. The supervisor scores independently and the two compare. Discrepancies are discussed immediately while the call is fresh.
Agents who have been through a structured Phase 2 typically reach 80%+ agreement with supervisor scoring within the first week of live calls. Agents who skip Phase 2 and move directly to live calls often take 6 to 8 weeks longer to calibrate.
QA Scoring: Measuring Qualification Performance
Training without measurement produces initial improvement that erodes over time. A monthly QA scoring process keeps calibration tight and surfaces individual performance issues before they affect case quality at scale.
Metrics That Matter
- Accept-to-sign rate by agent: Of the cases an agent accepted and forwarded for attorney review, what percentage were retained? A rate significantly below the team average indicates leniency.
- Escalation rate: Agents who never escalate borderline cases are likely making overconfident accept/decline decisions. An appropriate escalation rate is typically 10–20% of qualified calls.
- Rubric score vs. attorney outcome correlation: Do higher-scored cases sign at higher rates? If not, the rubric needs recalibration or agents aren't applying it consistently.
- Case type accept rate: Are agents accepting commercial vehicle, premises, and product liability cases at appropriate rates relative to standard MVA, or are non-standard case types being systematically undervalued?
Common QA finding: Agents who have been on the phone for more than 18 months without regular QA tend to have accept-to-sign rates 15–25% lower than newer agents who have been through structured calibration recently. Qualification accuracy declines without feedback loops — this is the single most predictable finding in intake QA data.
The Two-Tier Decision Model
Even with strong training, some calls will fall into genuinely ambiguous territory. A two-tier model handles these without burdening attorneys or losing cases unnecessarily.
Tier 1 (Agent): Clear accepts (score 10–12) and clear declines (score 4–6) are resolved at the agent level. The agent explains the decision to the caller and either proceeds to scheduling or declines with appropriate explanation.
Tier 2 (Supervisor or Senior Intake): Borderline cases (score 7–9) are escalated within a defined window (typically 15 to 30 minutes). The caller is told they'll receive a callback. A supervisor reviews the intake notes and rubric score, sometimes conducts a brief follow-up call, and makes the final decision. This review step also provides continuous calibration data for the escalating agent.
Firms that implement the two-tier model report three consistent benefits: fewer attorney-intake disagreements (the supervisor handles the cases both sides argued about before), higher accept-to-sign rates (supervisor review catches leniency errors before they reach the attorney), and lower agent stress on borderline calls (agents aren't forced to make judgment calls alone on calls they aren't certain about).
Avoiding the Five Most Common Qualification Errors
- Accepting on injury severity without confirming liability. An agent who hears "I was seriously hurt" and proceeds without establishing who was at fault has done half the intake. Liability must be confirmed before injury depth matters.
- Declining based on delayed treatment. Callers often wait several days to seek treatment for reasons unrelated to injury severity (no insurance, fear of medical debt, initially underestimated pain). A treatment gap is a flag to probe, not a reason to decline.
- Treating pre-existing conditions as disqualifiers. An aggravated pre-existing condition can be a strong PI case. The question is whether the accident materially worsened the condition — not whether the condition existed before.
- Assuming uninsured defendants make cases unretainable. UM/UIM coverage, employer liability, vehicle owner liability, and other theories frequently provide recovery paths even when the direct defendant is uninsured. Collectability should be scored after probing all available theories.
- Using call length as a quality proxy. Thorough intake takes longer. Rushed intake takes less time. Agents who are rewarded for short calls develop a bias toward quick declines over thorough qualification. Call handling time metrics should never be applied to intake without also tracking accept-to-sign rates.
Frequently Asked Questions
What are the core qualification criteria for a PI intake case? +
The four universal factors are liability, damages, causation, and collectability. All four must score above a firm-defined threshold. Most PI firms decline or escalate cases that score a 1 on any single factor regardless of total score — a case with no identifiable at-fault party, for example, rarely proceeds regardless of injury severity.
How long does it take to train a new agent to qualify cases accurately? +
Most agents reach 80%+ agreement with attorney decisions within 4 to 6 weeks of structured training. The key variable is whether Phase 2 (scored call reviews) is included. Agents who skip directly to live calls take significantly longer to calibrate and reach lower final accuracy levels.
What is the most common qualification error? +
Accepting cases based on injury severity without confirming liability. Agents who hear "I was seriously hurt" often proceed to scheduling without establishing fault. The second most common error is declining non-standard case types (commercial vehicle, premises, product liability) because they don't pattern-match to the standard MVA template agents were trained on.
Should intake agents make final accept/decline decisions? +
Agents should make clear accept/decline decisions, with borderline cases escalated to a supervisor or senior intake specialist. Sending all calls to an attorney for qualification wastes attorney time. Declining all borderline cases without escalation loses revenue. The two-tier model captures the efficiency benefits of agent-level filtering while providing a safety net for cases that require more judgment.
How do you measure qualification accuracy? +
The primary measure is accept-to-sign rate by agent — of cases an agent accepted, what percentage were retained? Supplementary metrics include escalation rate, rubric-to-outcome correlation, and case type accept rate. Monthly QA reviews comparing agent rubric scores against attorney decisions provide the calibration feedback that keeps accuracy high over time.
Ready to Improve Qualification Accuracy?
HQ Intake runs PI case qualification for law firms across the country — with trained specialists, a structured rubric, and QA scoring that gives attorneys consistent case quality.
Schedule a Consultation