Configuration

ReturnShield AI works out of the box with sensible defaults. This guide covers the settings you can customize.

AI Action Queue Settings

Auto-Apply Threshold

Set a confidence threshold for automatic application of AI fixes. Changes below this threshold require manual approval.

• Default: Manual approval for all changes
• Recommended: Auto-apply fixes with 90%+ confidence and < $100/mo impact
• Location: Settings → Action Queue → Auto-Apply

Fix Priority Weighting

Customize how the Action Prioritizer ranks fixes:

| Factor | Default Weight | Description | |---|---|---| | Expected savings | 40% | Monthly dollar savings predicted | | Confidence score | 35% | AI confidence in the prediction | | Implementation effort | 25% | How complex the change is |

Alerts & Notifications

Return Spike Alerts

Receive notifications when a product's return rate exceeds its category benchmark.

• Threshold: Configurable (default: 1.5x category average)
• Channels: Email, Shopify admin notification
• Location: Settings → Alerts → Return Spikes

Serial Returner Alerts

Get notified when a customer is flagged as a serial returner.

• Trigger: Return rate > 40% over last 6 months
• Action: Optional automatic order tagging via Shopify Flow
• Location: Settings → Alerts → Serial Returners

Bracketing Detection

Alert when a customer orders multiple sizes of the same item.

• Trigger: 3+ variants of the same product in one order
• Action: Optional fulfillment hold for warehouse review
• Location: Settings → Fraud → Bracketing

Sizing Social Proof Widget

Enabling the Widget

1. Go to Settings → Widget
2. Toggle "Enable Sizing Widget"
3. The widget uses product metafields — no theme code changes needed

Widget Appearance

• Text format: "X% of customers kept size Y" (customizable)
• Minimum data: Widget only shows when 10+ customers have purchased a size
• Placement: Below the size selector on product pages (via metafields)

Shopify Flow Integration

Available Triggers

| Trigger | Fires when | |---|---| | High-risk order placed | Order risk score > threshold | | Return spike detected | Product return rate exceeds benchmark | | Serial returner identified | Customer flagged as habitual returner | | Product risk threshold exceeded | Product risk score > threshold |

Available Actions

| Action | Effect | |---|---| | Add order tag | Tags the order for fulfillment team review | | Update customer metafield | Stores risk tier on customer record |

Example Workflow

Prevent high-risk shipments:

1. Trigger: "High-risk order placed"
2. Action: "Add order tag" → review-before-ship
3. Your fulfillment team reviews tagged orders before shipping

Post-Purchase Emails

"How's it fit?" Email

• Timing: 7 days after delivery (configurable: 3-14 days)
• Content: "How's your [product name]?" with quick feedback buttons
• Data flow: Responses feed back into product risk scores

Post-Return Surveys

• Trigger: After a return is processed
• Conditional logic: "Didn't fit" → sizing questions, "Not as described" → listing feedback
• Data flow: Survey responses flow into AI analysis automatically

Data & Privacy

Data Retention

| Plan | Retention | |---|---| | Free | 90 days | | Starter - Pro | 12 months | | Scale | Unlimited |

GDPR Compliance

• All data processed via Shopify OAuth
• Customer data deletion available on request
• No data shared with third parties
• EU data processing compliant