Skip to main content

Integration Overview

Eaternity Forecast integrates with your existing restaurant systems to automatically collect sales data and deliver predictions. This guide explains available integration options and helps you choose the right approach.

Integration Options

Best for: Existing Necta ERP customers

Eaternity Forecast launches exclusively with Necta, our strategic ERP partner for professional kitchens.

Key Benefits:

  • Zero integration work — Pre-built connection
  • Automatic data sync — Sales data flows seamlessly
  • Native interface — Predictions appear in familiar Necta planning tools
  • Unified workflow — No system switching for kitchen teams

Setup Time: 1-2 weeks from activation request

Learn more about Necta Integration →

2. Custom API Integration

Best for: Operations with technical resources and non-Necta systems

Connect directly via our REST API for full flexibility.

Key Features:

  • Standard REST API — Industry-standard HTTP/JSON
  • OAuth 2.0 or API keys — Secure authentication
  • Real-time predictions — Access forecasts programmatically
  • Webhook support — Receive updates automatically

Setup Time: 2-6 weeks depending on complexity

View Custom API Integration Guide →

3. Manual Data Upload

Best for: Initial evaluation or smaller operations

Upload sales data manually for simplified testing.

Process:

  • Export sales from your POS weekly
  • Upload CSV/Excel to secure portal
  • Receive predictions via email/dashboard

Setup Time: 1 week

Limitation: Not suitable for long-term production use

See Quick Setup Guide →

Integration Architecture

Data Flow Overview

Your POS/ERP System

[Sales Data Export]

Eaternity Forecast API

[Neural Network Processing]

[Daily Predictions Generated]

[Delivery Layer]

Your Planning System/Dashboard

Required Data Exchange

Outbound (Your System → Forecast)

Daily Sales Data (automated or manual):

Required fields:

  • Date (YYYY-MM-DD)
  • Item name/ID
  • Quantity sold
  • Service period (optional but recommended)

Optional but valuable:

  • Price
  • Category
  • Weather conditions
  • Special events

Format Options:

  • JSON via REST API
  • CSV file upload
  • Excel spreadsheet
  • Database direct connection (custom)

Inbound (Forecast → Your System)

Daily Predictions (automated delivery):

Provided data:

  • Date (next 7 days)
  • Item name/ID
  • Predicted quantity
  • Lower bound (10th percentile)
  • Upper bound (90th percentile)
  • Confidence score
  • Historical accuracy

Delivery Options:

  • REST API endpoints (pull)
  • Webhook notifications (push)
  • Email reports
  • Dashboard interface

Choosing Your Integration Method

Decision Matrix

FactorNecta IntegrationCustom APIManual Upload
Setup ComplexityVery LowMedium-HighLow
Technical ResourcesNone requiredDeveloper neededBasic Excel skills
Automation LevelFullFull (once built)Manual
Ongoing MaintenanceZeroLowMedium (weekly work)
Data FreshnessReal-timeReal-timeWeekly
ScalabilityExcellentExcellentLimited
Best ForNecta customersCustom POS systemsInitial testing

Integration Selection Guide

Choose Necta Integration if:

  • ✅ You are an existing Necta customer
  • ✅ You want fastest, easiest setup
  • ✅ You prefer zero technical work
  • ✅ You want seamless workflow integration

Choose Custom API if:

  • ✅ You use a different ERP/POS system
  • ✅ You have technical development resources
  • ✅ You need custom integration with your workflow
  • ✅ You want full control over data exchange

Choose Manual Upload if:

  • ✅ You're evaluating Forecast initially
  • ✅ You don't have API access to your POS
  • ✅ You have limited technical resources
  • ✅ You can commit to weekly manual uploads

Authentication Methods

Best for: Secure, delegated access

Flow:

  1. Your system redirects user to Eaternity auth page
  2. User grants permission
  3. Eaternity issues access token
  4. Your system uses token for API requests

Benefits:

  • Industry-standard security
  • Token expiration and refresh
  • User-specific permissions
  • Audit trail of access

Contact support for OAuth 2.0 implementation →

API Keys

Best for: Server-to-server integration

Setup:

  1. Generate API key in Forecast dashboard
  2. Store securely in your system
  3. Include in API request headers

Usage:

curl -X GET "https://api.eaternity.org/v1/forecast/predictions?date=2024-01-20" \
-H "Authorization: Bearer your_api_key_here"

Benefits:

  • Simple implementation
  • Suitable for automated scripts
  • Easy to rotate/revoke

Basic Authentication

Best for: Development and testing only

Not recommended for production due to security limitations.

Data Formats

Sales Data Submission

JSON Format (API preferred):

{
"kitchen_id": "your_kitchen_id",
"date": "2024-01-19",
"service_periods": [
{
"period": "lunch",
"items": [
{
"item_id": "pasta_carbonara",
"name": "Pasta Carbonara",
"quantity_sold": 52,
"price": 14.50,
"category": "Main Course"
},
{
"item_id": "caesar_salad",
"name": "Caesar Salad",
"quantity_sold": 31,
"price": 9.00,
"category": "Starter"
}
]
}
]
}

CSV Format (Manual upload):

date,item_id,item_name,quantity_sold,service_period,price,category
2024-01-19,pasta_carbonara,Pasta Carbonara,52,lunch,14.50,Main Course
2024-01-19,caesar_salad,Caesar Salad,31,lunch,9.00,Starter
2024-01-19,grilled_salmon,Grilled Salmon,28,lunch,18.50,Main Course

Prediction Retrieval

JSON Format (API response):

{
"kitchen_id": "your_kitchen_id",
"generated_at": "2024-01-20T03:15:42Z",
"predictions": [
{
"date": "2024-01-20",
"day_of_week": "Saturday",
"items": [
{
"item_id": "pasta_carbonara",
"name": "Pasta Carbonara",
"predicted_quantity": 52,
"confidence_interval": {
"lower": 48,
"upper": 56
},
"confidence_score": 0.92,
"accuracy_last_30_days": 94.2
}
]
}
]
}

System Requirements

Technical Requirements

For API Integration

Your System Must Support:

  • HTTPS requests (TLS 1.2 or higher)
  • JSON parsing
  • OAuth 2.0 or API key authentication
  • Scheduled tasks (cron jobs or equivalent)

Recommended:

  • Webhook endpoint capability
  • Error logging and monitoring
  • Automatic retry logic
  • Data validation

For Manual Upload

Your System Must Support:

  • CSV or Excel export from POS
  • Date range selection for exports
  • Item-level transaction data

Network Requirements

Outbound Access:

  • API endpoint: https://api.eaternity.org
  • Ports: 443 (HTTPS)
  • No proxy restrictions on API calls

Firewall Configuration:

  • Allow outbound HTTPS to *.eaternity.org
  • (Optional) Allow inbound webhooks from Eaternity IPs

IP Whitelisting (if required): Contact support for current Eaternity API server IPs

Data Requirements

Minimum Historical Data:

  • 30 days of sales data for training
  • Daily records (no gaps >2 consecutive days)
  • Item-level quantities (not just revenue)

Recommended Historical Data:

  • 90+ days for optimal accuracy
  • Weather data for same period
  • Event calendar annotations

See Quick Setup Guide →

Integration Workflow

Phase 1: Planning (Week 1)

Activities:

  1. Choose integration method based on decision matrix
  2. Review technical requirements for your chosen method
  3. Assign team resources (technical contact, kitchen liaison)
  4. Schedule kickoff call with Eaternity integration team

Deliverables:

  • Integration plan document
  • Technical contact assigned
  • Timeline agreed

Phase 2: Setup (Week 2-3)

Activities:

  1. Credential setup — API keys or OAuth configuration
  2. Endpoint configuration — API URLs, webhook targets
  3. Data mapping — Match your item IDs to forecast items
  4. Historical data export — Prepare training dataset

Deliverables:

  • Authentication working
  • Test data submitted successfully
  • Historical data imported

Phase 3: Testing (Week 3-4)

Activities:

  1. Data validation — Verify format and quality
  2. Integration testing — Automated data flow
  3. Prediction testing — Receive and validate forecasts
  4. Error handling — Test failure scenarios

Deliverables:

  • Data quality report approved
  • Automated sync working
  • First test predictions received

Phase 4: Training (Week 4-5)

Activities:

  1. Model training — Neural network learns your patterns
  2. Accuracy validation — Compare to baseline
  3. Team training — Kitchen staff learn to use predictions
  4. Process documentation — Document your workflow

Deliverables:

  • Model training complete
  • Initial accuracy benchmarks established
  • Team trained on system

Phase 5: Production (Week 6+)

Activities:

  1. Go-live — Start using predictions daily
  2. Monitoring — Track accuracy and issues
  3. Optimization — Refine based on feedback
  4. Continuous improvement — Weekly model retraining

Deliverables:

  • Daily predictions in production use
  • Performance monitoring in place
  • Feedback loop established

Integration Support

Available Resources

Documentation:

Technical Support:

  • Email: forecast@eaternity.org
  • Response time: 24 hours for integration questions
  • Phone support: For critical production issues

Integration Assistance:

  • Setup consultation calls
  • Technical review of integration code
  • Data quality validation
  • Performance optimization guidance

Common Integration Scenarios

Scenario 1: Restaurant Chain with Central POS

Setup:

  • Central POS system (Toast, Square, etc.)
  • Multiple locations
  • Standardized menu across locations

Recommended Approach:

  • Single API integration at POS level
  • Location-specific predictions via location_id parameter
  • Shared menu item mapping
  • Central management dashboard

Timeline: 3-4 weeks for first location, 1-2 weeks per additional location

Scenario 2: Independent Restaurant with Necta

Setup:

  • Single location
  • Necta ERP already in use
  • Kitchen manager familiar with Necta

Recommended Approach:

  • Activate Necta-Forecast integration
  • Automatic data sync from existing Necta data
  • Predictions visible in Necta planning interface

Timeline: 1-2 weeks total

Scenario 3: Hospital Food Service with Custom ERP

Setup:

  • Large institution with custom ERP
  • IT department available
  • Complex menu with dietary restrictions

Recommended Approach:

  • Custom API integration by IT team
  • Map dietary variants to item categories
  • Webhook notifications for daily predictions
  • Integration with inventory management

Timeline: 4-6 weeks including IT development

Scenario 4: University Canteen with Legacy POS

Setup:

  • Older POS system with limited API
  • IT resources constrained
  • Academic calendar effects

Recommended Approach:

  • Manual CSV export from POS (weekly)
  • Upload via secure portal
  • Email delivery of predictions
  • Future migration to API when resources available

Timeline: 1 week for manual process setup

Security and Compliance

Data Security

Transmission:

  • All data encrypted in transit (TLS 1.2+)
  • Certificate validation enforced
  • No unencrypted HTTP allowed

Storage:

  • Data encrypted at rest
  • Regular security audits
  • Access logging and monitoring

Authentication:

  • OAuth 2.0 for user-delegated access
  • API keys with rotation capability
  • IP whitelisting available

Compliance

GDPR Compliance:

  • Data processing agreement available
  • Right to data deletion
  • Data portability support
  • Audit logs maintained

Data Ownership:

  • You retain full ownership of your sales data
  • Eaternity uses data only for your predictions
  • Opt-in for anonymized research contribution

Data Privacy

What We Store:

  • Historical sales quantities
  • Menu item information
  • Prediction history
  • Accuracy metrics

What We DON'T Store:

  • Guest personal information
  • Payment card data
  • Individual transaction details
  • Identifiable guest data

Next Steps

Getting Started

  1. Review integration options on this page
  2. Choose your integration method based on requirements
  3. Read detailed guide for your chosen method:
  4. Contact us to begin setup: forecast@eaternity.org

Additional Resources

See Also