Smart scan

The smart scan functionality allows you to provide data in multiple formats without any prior knowledge of the data structure. Lune analyses this input, deduces the transactions, and calculates the associated emissions.

This functionality is beneficial for cases where you have no concrete information regarding transactions you want to estimate, but have several pieces of data that can help guide this discovery such as receipts, invoices, and emails. By leveraging various data sources, smart scan enables a more comprehensive understanding of your financial activities and their environmental impact, even when dealing with fragmented or diverse information.

Note: This feature is in beta. Please verify results manually. Contact us to help improve its accuracy for your specific needs.

Input data

The smart scan functionality accepts data in two main formats:

1. Structured transaction estimate data

   • Similar to Transaction estimates
   • Allows you to define known properties of the estimates
   • Guarantees these fields will be respected regardless of other input

2. Unstructured data in text format

   • Can contain any relevant information about the transactions
   • Lune parses this to extract relevant data
   • Combined with structured data to produce a final estimate
   • Examples: OCR data extracted from invoices, receipts, etc.

How It Works

1. Lune analyzes both structured and unstructured data
2. Extracts relevant information from unstructured text
3. Combines this with any provided structured data
4. Produces a final emission estimate

Example Use Case

• Transaction: US dollar payment for cloud computing services
• Scenario: Full transaction details are on an invoice, not known by the client
• Data: Mindee was used to extract information from the invoice

This approach allows for accurate emission estimates even when complete transaction details are not initially available to the client.

Sample request body

{
    "currency": "USD",
    "merchant": {
      "category": "cloud computing"
    },
    "unstructured_data": {
      "key_value": {
        "prediction": {
          "locale": {
            "language": "en",
            "currency": "USD"
          },
          "line_items": [
            {
              "description": "AWS Lambda",
              "total_amount": 10
            },
            {
              "description": "AWS EC2",
              "total_amount": 200
            }
          ]
        }
      }
    }
}

The smart scan estimate

A smart scan estimate provides:

• Overall transaction information:

  • mass: Total emissions
  • quote: Appropriate offset quote in Lune

• Detailed item information:

  • line_items: Contains at least one element with details of detected items

Currently, only transaction type estimates are available. Future updates will include activity-based emissions (e.g., flights).

The smart scan accepts:

1. Structured transaction estimate data
2. Unstructured text data (e.g., OCR from invoices)

Lune analyzes both, extracting and combining information to produce the final estimate.

Note: Sample responses omits some fields for readability.

Sample emission response (only the relevant parts)

{
  "mass": {
    "unit": "t",
    "amount": "0.03612"
  },
  "quote": {
    "currency": "GBP",
    "estimated_quantity": "0.03612",
  },
  "line_items": [
    {
      "mass": {
        "unit": "t",
        "amount": "0.00172"
      },
      "type": "transaction",
      "emission_factor": {
        "name": "Data processing and hosting",
        "region": "United States of America",
        "source": "epa",
        "category": "cloud computing",
        "gas_emissions": {
          "co2": "0.139",
          "co2e": "0.172",
          "other": "0.008",
          "methane": "0.001",
          "nitrous_oxide": "0"
        },
        "numerator_unit": "kg",
        "denominator_unit": "USD",
      },
    },
    {
      "mass": {
        "unit": "t",
        "amount": "0.0344"
      },
      "type": "transaction",
      "emission_factor": {
        "name": "Data processing and hosting",
        "region": "United States of America",
        "source": "epa",
        "category": "cloud computing",
        "gas_emissions": {
          "co2": "0.139",
          "co2e": "0.172",
          "other": "0.008",
          "methane": "0.001",
          "nitrous_oxide": "0"
        },
        "numerator_unit": "kg",
        "source_version": "1.1.1",
        "denominator_unit": "USD",
        "publication_year": 2022
      },
      "exchange_rate_date": null
    }
  ]
}