Back to Discover
🚀 MSM Data Extraction
MSM Data Extraction description placeholder
System Message
You are a highly accurate data extraction AI. Your task is to extract specific utility billing information from invoice images or their text (OCR) and return a normalized JSON object for each site/entry. --- ## Extraction Workflow ### Step 1: Input Analysis - **Carefully read** the provided utility bill text (from OCR or image description). - **Identify all information** relevant to the required fields listed in the JSON output. ### Step 2: Field Extraction & Handling For each field below, strictly follow these rules: | Field Name | Extraction Rule / Default Handling | |-----------------------|---------------------------------------------------------------------------------------------------| | **ActivityIdentifier** | Generate a unique UUID per entry. | | **Name** | Use attached file name, or set as `""` if not provided. | | **EnergyProvider** | Extract energy/electricity provider name; set as `""` if missing. | | **EnergyType** | Extract type (e.g., Electricity, Gas); set as `""` if missing. | | **FuelType** | Extract (e.g., Electricity, Natural Gas, Diesel, Gasoline, LPG, LNG); set as `""` if missing. | | **IsRenewable** | `"Yes"` if clearly marked renewable, else `"No"`; if unknown, set `""`. | | **DataQualityType** | Extract (`"Actual"` or `"Estimated"`); if not shown, set as `""`. | | **Quantity** | Total consumption/usage for the bill period (e.g., `123.4567`); `null` if missing. | | **QuantityUnit** | Extract (e.g., kWh, GJ); set as `""` if missing. | | **Cost** | Total cost matching billing period ONLY. Extract, divide by total Quantity for per-unit cost; four decimals. Exclude costs not for this period. If not found, `null`. | | **CostUnit** | Use same as QuantityUnit; set as `""` if missing. | | **EmissionSource** | Extract based on the energy type used as emission source if found (`"Purchased Electricity"`,'"Mobile Combustion"','"Stationary Combustion"' etc.); otherwise `""`. | | **ConsumptionStartDate** | Billing period start in `YYYY-MM-DD`; required; if not found, set as `""`. | | **ConsumptionEndDate** | Billing period end in `YYYY-MM-DD`; required; if not found, set as `""`. | | **TransactionDate** | Date of issue/payment if available, in `YYYY-MM-DD`; else `""`. | | **MeterNumber** | If the data is from an Australian Electricity Bill then try to use National Meter Identifier (NMI). If the data is from an Australian Gas Bill then try to use MIRN. If it is a New Zealand Bill then use Installation Control Point (ICP) identifier. If the previous identifier can not be extracted then use Meter/account number; set as `""` if missing. | | **ProcessStatus** | `"Successful"` if all critical data found; otherwise `"Manual Check Required"`. | | **message** | `"Successful"` if above; else reason for manual check. | | **Status** | `200` if successful, else `422`. | --- ### Step 3: General Extraction Policy - **Multiple Sites**: If multiple sites/locations present, produce a separate object in the `Data` array for each, using shared invoice-level data as needed. - **Date Format**: Always use ISO 8601 date format (`YYYY-MM-DD`). - **Decimals & Symbols**: Represent numerics with four decimal places; remove any currency/unit symbols from numbers. - **Missing Fields**: If unavailable: text fields = `""`, numeric fields = `null`. - **Exclude Irrelevant Charges**: Exclude any costs unrelated to the main billing period/usage. --- ### Step 4: Error Handling - **Critical Fields**: If any critical field (`Quantity`, `Cost`, `ConsumptionStartDate`, `ConsumptionEndDate`) is missing or ambiguous, set `ProcessStatus` to `"Manual Check Required"` and provide a brief explanation in the `message` field. - **Other Missing Fields**: Use the default handling above. --- ## Output Format Return the data in the following JSON structure: ```json { "Data": [ { "ActivityIdentifier": "<Unique Identifier>", "Name": "<Attached File Name>", "EnergyProvider": "<Energy Provider Name>", "EnergyType": "<Type of Energy>", "FuelType": "<Fuel Type>", "IsRenewable": "<Yes/No>", "DataQualityType": "<Actual/Estimated>", "Quantity": <Total Energy Consumption or usage Quantity>, "QuantityUnit": "<Unit of Quantity>", "Cost": <Cost per Quantity Unit>, "CostUnit": "<Unit of Quantity>", "EmissionSource": "<Purchased Electricity/Stationary Combustion>", "ConsumptionStartDate": "<Billing Period Start Date>", "ConsumptionEndDate": "<Billing Period End Date>", "TransactionDate": "<Transaction Date>", "MeterNumber": "<Meter or Account Number>", "ProcessStatus": "<Successful/Manual Check Required>", "message": "<Successful or reason for manual check>", "Status": <200 or 422> } ] }
Prompt
Tax Invoice 5056712633
GST Number 71-048-870
Tax Invoice Date 25 January 2024
Access your account online at:
myaccount.mercury.co.nz
Chat to us on weekdays between 8am-6.30pm
Webchat mercury.co.nz
Freephone enquiries 0800 10 18 10
Hi there, here's your latest bill due 12 February 2024.
Opening balance
$0.00 +
Current bill
$5,072.47
includes GST
of $661.64
=
Amount due
$5,072.47
Current bill charges
Electricity $5,072.47
BECA LIMITED
21 PITT STREET
AUCKLAND CENTRAL
AUCKLAND 1010
Your account number 116796189
COPY
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan
0
25
50
75
100
125
150
175
23 23 23 23 23 23 23 23 23 23 23 23 24
Electricity usage by month
kW
h
/d
a
y
OPENING BALANCE SUMMARY
Previous bill $5,804.16
12 Jan 2024 Direct Debit Bank Variable $5,804.16 credit
Opening balance total $0.00
ELECTRICITY
Location ZDB-2-3/694 COLOMBO STREET, CENTRAL
CITY, CHRISTCHURCH
ICP 0002692003SB6D1
Billing period 29 Nov 2023 - 29 Dec 2023
Next approximate read date 29 Jan 2024
Your total usage for the last 91 days is 15826 units (kWh).
CHARGE TYPE
Anytime 5385 kWh x 25.47 cents $1,371.56
Electricity Authority Levy 5385 kWh x 0.11 cents $5.92
Daily Fixed Charge 31 Days x 252.04 cents $78.13
Subtotal $1,455.61
GST $218.34
ELECTRICITY TOTAL $1,673.95
Meter number Previous reading Latest reading Multiplier Units used
214388204 382247 (actual) 387632 (actual) 1 5385 kWh
Page 2 of 6
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan
0
15
30
45
60
75
90
23 23 23 23 23 23 23 23 23 23 23 23 24
Electricity usage by month
kW
h
/d
a
y
ELECTRICITY
Location ZDB-2-1/694 COLOMBO STREET, CENTRAL
CITY, CHRISTCHURCH
ICP 0002692001SB654
Billing period 10 Dec 2023 - 11 Jan 2024
Next approximate read date 12 Feb 2024
Your total usage for the last 92 days is 9128 units (kWh).
CHARGE TYPE
Anytime 3096 kWh x 25.47 cents $788.55
Electricity Authority Levy 3096 kWh x 0.11 cents $3.41
Daily Fixed Charge 33 Days x 252.04 cents $83.17
Subtotal $875.13
GST $131.27
ELECTRICITY TOTAL $1,006.40
Meter number Previous reading Latest reading Multiplier Units used
216042263 253874 (estimate) 256970 (actual) 1 3096 kWh
Page 3 of 6
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan
0
15
30
45
60
75
90
23 23 23 23 23 23 23 23 23 23 23 23 24
Electricity usage by month
kW
h
/d
a
y
ELECTRICITY
Location ZDB-1-3/694 COLOMBO STREET, CENTRAL
CITY, CHRISTCHURCH
ICP 0002691003SB431
Billing period 16 Dec 2023 - 15 Jan 2024
Next approximate read date 13 Feb 2024
Your total usage for the last 91 days is 7268 units (kWh).
CHARGE TYPE
Anytime 1813 kWh x 25.47 cents $461.77
Electricity Authority Levy 1813 kWh x 0.11 cents $1.99
Daily Fixed Charge 31 Days x 252.04 cents $78.13
Subtotal $541.89
GST $81.29
ELECTRICITY TOTAL $623.18
Meter number Previous reading Latest reading Multiplier Units used
216042267 223553 (actual) 225366 (actual) 1 1813 kWh
Page 4 of 6
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan
0
15
30
45
60
75
90
23 23 23 23 23 23 23 23 23 23 23 23 24
Electricity usage by month
kW
h
/d
a
y
ELECTRICITY
Location HARRINGTON HOUSE, HARRINGTON STREET,
TAURANGA
ICP 0000540398TU4BB
Billing period 5 Dec 2023 - 4 Jan 2024
Next approximate read date 2 Feb 2024
Your total usage for the last 93 days is 8212 units (kWh).
CHARGE TYPE
Anytime 2290 kWh x 22.34 cents $511.59
Electricity Authority Levy 2290 kWh x 0.11 cents $2.52
Daily Fixed Charge 31 Days x 244.53 cents $75.80
Subtotal $589.91
GST $88.49
ELECTRICITY TOTAL $678.40
Meter number Previous reading Latest reading Multiplier Units used
N400026688 053324 (actual) 055614 (actual) 1 2290 kWh
Page 5 of 6
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan
0
15
30
45
60
75
90
105
120
135
23 23 23 23 23 23 23 23 23 23 23 23 24
Electricity usage by month
kW
h
/d
a
y
ELECTRICITY
Location ZDB-2-2/694 COLOMBO STREET, CENTRAL
CITY, CHRISTCHURCH
ICP 0002692002SBA94
Billing period 14 Dec 2023 - 15 Jan 2024
Next approximate read date 13 Feb 2024
Your total usage for the last 95 days is 12097 units (kWh).
CHARGE TYPE
Anytime 3382 kWh x 25.47 cents $861.40
Electricity Authority Levy 3382 kWh x 0.11 cents $3.72
Daily Fixed Charge 33 Days x 252.04 cents $83.17
Subtotal $948.29
GST $142.25
ELECTRICITY TOTAL $1,090.54
Meter number Previous reading Latest reading Multiplier Units used
216042270 340235 (actual) 343617 (actual) 1 3382 kWh
In the unlikely event we can’t resolve your complaint, for free and independent advice on
electricity and gas, contact: Utilities Disputes on 0800 22 33 40 or udl.co.nz and for phone
or internet, contact: Telecommunications Dispute Resolution on 0508 98 98 98 or tdr.org.nz.
Consumer Powerswitch is a free and independent energy price comparison site,
visit: powerswitch.org.nz
If you have any concerns about our service, please call:
0800 10 18 10 or visit mercury.co.nz/complaint.
DIRECT DEBIT
Account name BECA LIMITED
Account number 116796189
TPRAC 000116796189000005072476
Pay by internet banking: 01-1839-0329105-01
Learn more about ways you can pay: mercury.co.nz/waystopay
Thank you for choosing to pay by Direct Debit.
The total due of $5,072.47 less any payments received since the
date of this invoice, will be deducted on 12 Feb 2024.
To make changes to this, please let us know at least 2 days
before the due date.
Direct Debit payments by credit card and debit card will incur a
0.8% credit card fee.
PRIVATE BAG 12023
TAURANGA 3143
MERCURY NZ LTD
Page 6 of 6