AI SNAP Trafficking Detection Playbook

A state SNAP agency's analytics team has flagged 42 retailers with suspicious transaction patterns over 6 months: high-value transactions in low-income zip codes, round-dollar transactions above $100, and multiple transactions per household in a single day. Total benefits transacted: $4.2M.

When to use this playbook

• Use this playbook when the decision looks like the situation above: A state SNAP agency's analytics team has flagged 42 retailers with suspicious transaction patterns over 6 months: high-value transactions in low-income zip codes, round-dollar transactions above $100, and multiple transactions per household in a single day.
• It is a fit when you have source files in hand and need a structured, reviewable analysis — not a generic chat answer about "SNAP Trafficking Detection".
• Do not use it as a substitute for licensed, legal, clinical, or authorized official judgment in the domain.

What you'll need

• SNAP transaction data for all 42 retailers (6 months)
• Retailer authorization records and store type classification
• Household benefit issuance records
• USDA FNS trafficking indicators list
• State OIG referral criteria

Attachments: Documents (Documents)

You are a public benefits fraud analyst investigating SNAP trafficking at 42 flagged retailers. I am attaching:

Work only from the attached source files. If a conclusion is not supported, say so.

Produce:
1. Score each retailer against USDA FNS trafficking indicators: round-dollar transactions, high-value singles, transaction velocity, and household repeat-visit patterns.
2. Identify household clusters showing trafficking-consistent behavior: multiple transactions at the same retailer in one day, transactions immediately after benefit issuance.
3. Calculate the estimated trafficking volume for each retailer: flagged vs. legitimate purchase patterns.
4. Rank the 42 retailers by evidence strength and estimate the total trafficking dollar exposure.
5. Tell me which retailers meet the threshold for OIG referral today and what documentation is needed for a disqualification action.

Call out where independent models are likely to disagree, and list follow-up documents a reviewer should request.

What to expect

• Retailer risk scoring against USDA FNS indicators
• Household clustering analysis
• Estimated trafficking volume per retailer
• Prioritized OIG referral list
• Disqualification action documentation checklist

Review before you act

• Validate this output against source files before relying on it: Score each retailer against USDA FNS trafficking indicators: round-dollar transactions, high-value singles, transaction velocity, and household repeat-visit patterns.
• Validate this output against source files before relying on it: Identify household clusters showing trafficking-consistent behavior: multiple transactions at the same retailer in one day, transactions immediately after benefit issuance.
• Validate this output against source files before relying on it: Calculate the estimated trafficking volume for each retailer: flagged vs. legitimate purchase patterns.
• Validate this output against source files before relying on it: Rank the 42 retailers by evidence strength and estimate the total trafficking dollar exposure.
• Confirm every cited figure, date, counterparty, or requirement against the attached originals — models compress and can drop a qualifier.
• Treat disagreement between models as a review item, especially on classification, materiality, and recommended next action.
• Do not authorize an operational, clinical, legal, credit, or enforcement action solely because the models agree.