> ## Documentation Index > Fetch the complete documentation index at: https://docs.embedreach.com/llms.txt > Use this file to discover all available pages before exploring further. # Data Sync Setup > Phase 3: Define schemas for your data structure ## Overview Phase 3 focuses on defining how your data maps to Reach's system. This is accomplished through flexible schema definitions that match your existing data structure—no transformation required. ## The Two-Part Process Tell Reach how your data is structured by providing sample JSON Historical import followed by ongoing sync (covered in Phase 4) ## Schema Creation Process ### Step 1: You Provide Sample Data Send Reach example JSON objects of your customer, transaction, and other important business records **as they exist in your system**. ```json theme={null} { "id": "cust_12345", "email": "john@example.com", "phone": "+14155551234", "first_name": "John", "last_name": "Doe", "created_at": "2024-01-15T10:30:00Z", "city": "San Francisco", "state": "CA", "total_lifetime_value": 5280.50, "tags": ["vip", "email_subscribed"] } ``` ```json theme={null} { "id": "txn_67890", "customer_id": "cust_12345", "amount": 150.00, "created_at": "2024-03-20T14:22:00Z", "status": "completed", "items": [ { "name": "Service A", "price": 100.00 }, { "name": "Service B", "price": 50.00 } ] } ``` ```json theme={null} { "id": "loc_001", "name": "Downtown Office", "address": { "street": "123 Main St", "city": "Portland", "state": "OR", "zip": "97201" }, "phone": "+15035551000", "google_business_profile_id": "ChIJ..." } ``` Send real examples from your database—not idealized versions. Include all fields, even optional ones. ### Step 2: Reach Creates Custom Schemas Reach defines schemas in the system that accept your exact structure. Each schema is assigned a category: **Schema Categories:** * **`contacts_schema`**: Customer/contact records * **`transactions_schema`**: Transaction/billing records * **`locations_schema`**: Location/address records * **Additional schemas**: Custom schemas as needed for your business model **No data transformation required.** Reach accepts your format as-is. ### Step 3: Reach Maps Critical Fields Reach identifies which fields in your structure serve key purposes: **Customer Schema Mappings:** * Which field is the email? → `email` * Which field is the phone? → `phone` * Which field is the customer ID? → `id` * Which field is the creation date? → `created_at` * Which fields for segmentation? → `tags`, `city`, `state`, `total_lifetime_value` * Which fields for merge fields? → `first_name`, `last_name` These mappings enable: * Segmentation for targeted campaigns * Personalization in emails and SMS * Attribution logic **Transaction Schema Mappings:** * Which field is the amount? → `amount` or `total` * Which field is the transaction date? → `created_at` or `completed_at` * Which field links to customer? → `customer_id` * Which field indicates status? → `status` * Which fields contain line items? → `items` array These mappings enable: * ROAS calculations * Revenue attribution * Conversion tracking **Location Schema Mappings:** * Which field is the location name? → `name` * Which field has the address? → `address` object * Which field links to Google Business Profile? → `google_business_profile_id` These mappings enable: * Multi-location reputation management * Location-based segmentation * Google Business Profile integration ### Step 4: You Get Schema-Specific API Endpoints Reach provisions endpoints that accept your format: ```bash theme={null} POST /api/resources/contacts POST /api/resources/transactions POST /api/resources/locations ``` Each endpoint accepts your JSON structure directly—no reformatting needed. ## What Data You'll Send ### Customer/Contact Data (Required for All Industries) **Minimum Required:** * Email address (primary identifier) * Unique customer ID from your system **Highly Recommended:** * Phone number (enables SMS, improves attribution) * First and last name * Creation date/timestamp * Address (city, state, zip for geographic targeting) **Optional but Valuable:** * Custom attributes for segmentation * Lifetime value or spend metrics * Customer status (active, inactive, churned) * Opt-in/opt-out status for email and SMS * Tags or categories ### Transaction/Billing Data (Varies by Industry) **E-commerce, Bookings:** * Order data with line items * Sent after purchase completion * Include product/service details ```json theme={null} { "externalId": "order_4567", "customerId": "customer_789", "amount": 250.00, "createdAt": "2024-04-15T11:30:00Z", "status": "completed", "items": [ { "name": "Product A", "quantity": 2, "price": 100.00 } ] } ``` **SaaS, Memberships:** * Agreements/proposals after finalization * Can be amended over time as contracts change * Include contract value and terms ```json theme={null} { "externalId": "contract_789", "customerId": "customer_456", "amount": 12000.00, "createdAt": "2024-01-10T09:00:00Z", "startDate": "2024-02-01", "endDate": "2025-02-01", "status": "active" } ``` **B2B Services:** * Invoice data (even if unpaid) * More important than payment completion * Describes the sale before money changes hands ```json theme={null} { "externalId": "invoice_123", "customerId": "customer_789", "amount": 5000.00, "invoiceDate": "2024-03-01", "dueDate": "2024-03-31", "status": "pending", "services": [...] } ``` ### Location Data (If Applicable) **Required for:** * Reputation product (Google Business Profile mapping) * Location-based segmentation in Engage **Skip if:** * Only using Measure/Acquire * Single-location businesses ## Key Decision: What Counts as a Transaction? Work with Reach to define your "conversion event"—the moment that matters for attribution: **When:** Customer schedules/reserves **Pros:** * Shows impact quickly * Captures intent early **Cons:** * Not actual revenue yet * May include cancellations **When:** Customer pays **Pros:** * Actual revenue * Most accurate attribution **Cons:** * Longer attribution window * Delayed insights **When:** Invoice is issued **Pros:** * Describes the sale * Works for B2B/net-terms **Cons:** * May not reflect payment * Requires status updates You can send updates as transactions progress through stages (booked → paid → fulfilled). This provides both early visibility and eventual accuracy. ## Schema Evolution Schemas can evolve as your needs change: **Adding New Fields:** * Send updated sample data with new fields * Reach adds fields to schema * Existing records unaffected * New fields available for segmentation and merge fields **Changing Field Types:** * Discuss with Reach team via Slack * May require data migration * Reach handles complexity **Adding New Schemas:** * Send sample data for new resource types * Reach creates new schema and endpoints * Integrate at your own pace ## Phase 3 Checklist Real JSON examples of customers, transactions, locations Custom schemas created matching your structure Critical fields identified for attribution and segmentation Custom endpoints ready to accept your data Decision made on what counts as a transaction ## Next Steps With schemas defined, you're ready to choose a sync method and begin sending data. Choose sync method and implement ongoing data flow