Author: Malcolm Souness, Director, 221b Limited
Some time ago the customer specific energy margin was determined and used to identify a number of issues within an energy retailer. This is extremely commercially sensitive, with all data extraction, transformation and analysis performed independently to maintain confidentiality.

A series of sales channels were compared, with one presenting substantially less energy margin per kWh than all others. We also identified a series of historical online billing and direct payment discounts that had been reported against the energy ledger, artificially understating the energy margin by mid six figures each year.

This started out within the Microsoft PowerBI environment, however this proved limited in its ability to chart the margin of >100,000 customers. The R programming language was employed, given its extensive libraries for interrogation of databases, transformation and cleanup of data, and presentation of charts. In addition, the R scripts are readily auditable, and the results are reproducible.

If you suspect your energy margin is being eroded, and cannot clearly explain why - contact Malcolm in confidence on 021 565557.

By Malcolm Souness - Director 221b Limited

In my first role as an energy supply analyst, a whole series of reports for the Board were produced using separate spreadsheets and datasets. One thing that was missing was the ability to cross check and validate data for board reporting. Read More…

Controlled load availability

By Malcolm Souness - Director 221b Limited
During January 2020, we undertook an investigation into electricity meter register configurations on New Zealand electricity networks.

The number of infrequently used register configurations indicates that legacy ripple control schedule durations remain in the Electricity Registry. For example, Vector has over 0.2M controlled meter registers with 1-23 hours availability, however just one has an availability of 5 hours, and one other has six hours available. The balance of probabilities suggests these may not be legitimate hours available. Read More…

Author: Malcolm Souness, Director, 221b Ltd
Originally published 29 October, 2019, updated January 2020

The source data used for this analysis are taken from the Electricity Authority of New Zealand EMI website. The R statistical analysis package is used to transform and present the data in a more useable form, source code available from the 221b Data Refinery Github Repository. The key constraint on this dataset is the exclusion of de-energised sites (ICP Status 001). Read More…

Author: Malcolm Souness, Director, 221b Limited
Originally published 5 June 2019

Electricity industry participants (traders) in New Zealand are required to submit calendar normalised "submitted" energy volume (AV-080 for non half hour data, and AV-090 for half hour data), as well as the "supplied" (billed/invoiced) volume (AV-120). Refer: Reconciliation Manager Functional Specification v9.27. The normalised volumes are used for market settlement and clearance.
Read More…

Author: Malcolm Souness, Director, 221b Limited
Originally published 4 May 2016

Having worked within the energy sector for some time, I never thought that I'd find electricity pricing schedules interesting. Often published tariffs are taken for granted, however published tariffs not without errors. Read More…

Author: Malcolm Souness, Director, 221b Limited
Originally published 17 May 2016

Over the weekend I found myself looking into the monthly wholesale position of electricity retailers in different parts of New Zealand. The information isn't publicly available, however Market Participants can extract this information using the Reconciliation Manager's GR-120 "Report Unaccounted for Energy" (UFE) reports (refer to Reconciliation Manager Functional Specification pg 132). Read More…