©2018 Copyright by Simble Solutions (UK) Limited, a subsidiary of Simble Solutions Limited.

What is Open Energy?

Dan Wright | London, UK

If releasing banking #data for use by approved third parties via Open Banking worked for the banking sector, could it work for the #energy sector too? Let’s pick out the three recommendations made by the Open Energy report authored by the Federation of Small Businesses (FSB) in collaboration with Fingleton Associates and untangle how they might revolutionise the energy market.


1) Machine Learning: Standardising information


The first recommendation addresses the lack of standardisation in marketing and tariff information presented by the Big Six and the independents who rapidly building market share:

Standardising tariffs and other relevant market information in machine-readable formats to allow automated comparisons of energy tariff offerings.

Machine readability is one of the pillars of the open data movement [1]. In a nutshell, any information that is printed, or unstructured (see the Glossary), cannot be processed easily by a computer, while data that is presented digitally and in a structured format can be crawled and specified data pulled out. This could facilitate comparisons of the varying tariffs offered by the 65 or so energy providers in the UK [2]. In theory, this means that both domestic and commercial customers will have access to all the information they need make the best possible decisions at a specific time. It’s unlikely that even the most engaged energy customer will take full benefit of this new information, personally comparing tariffs daily to find the cheapest, greenest, highest rated (etc.) supplier, see Open Energy’s third recommendation for more on this.


2) Restoring the balance: Sharing data


Most of the smart meters that have been installed in the UK as of 2018 are SMETS 1-compliant [3], meaning that the meter is designed to only send data to a specific supplier. The newer SMETS 2 meters allow customers to switch energy provider without their smart meter losing all its smart functionality, however, the data will continue to flow to the provider, but will go no further. If the reasoning underlying the first recommendation was that easily accessible information can support better decision making, what could be the impact of more data? The second recommendation from Open Energy seeks to increase transparency:

Making smart meter data available through a secure standardised API to approved third parties.

Sharing data has been shown to be a valuable method of improving customer experience in markets such as transport (applications like Citymapper uses Transport for London’s open bus and underground schedule for route planning) and banking (applications like Yolt aggregate balances from across bank account and offer spending insights and marketing information on new accounts). What caught my attention was that this data would be available for use by approved third parties. This is common practice and makes sense when considering the sensitive nature of the data, however, thinking about the value of knowing what households are doing and when, will product and service marketers be excluded from access to smart meter data?


A graphical example of how energy data could transition into a more ‘open’ state (reproduced with permission from the Federation of Small Businesses & Fingleton Associates, 2018)

3) Energy contracts that work for you: Delegating control


For suppliers, this new information will allow for time-of-use tariffs based on energy price signals. With accurate information about when energy is being used, engaged domestic customers may be able to load-shift their daily activities such as washing or hot water heating to times of low demand (such as late nights) when on a tariff that has cheaper off-peak energy. As touched upon earlier, Open Energy’s third recommendation hits upon a potential means of fairer energy bills for all:

Allowing energy customers to delegate contract switching powers to third party intermediaries.

It is already possible to opt-in to delegating control of energy supplier switching to third parties (such as Labrador), but what if customers were required to opt-out of automatic supplier switching? If customers were to automatically switch to the tariff which suited their preferences [4] (elected by a system which benefits from machine-readable information), this could mean that hundreds of thousands of households avoid overpaying on their energy bills. This overpayment was estimated to be approximately £2 billion in 2015 [5] and it could be argued that this would mean a revenue cut for energy suppliers…


Opinion


The recommendations made in the Open Energy report seek to address an imbalance in access to data that has continued for decades. The role of energy providers may need to shift into supplying energy as a service rather than charging for energy consumed, and a new ‘intermediary’ sector could arise to support customer convenience and bring automation and machine-learning into the equation. Is Open Energy the answer? I'm not sure, but transparency around energy use could be a valuable tool for suppliers and customers alike.


Dan Wright is a doctoral researcher with the School of Architecture, Building and Civil Engineering at Loughborough University funded by the EPSRC London-Loughborough (LoLo) Centre for Doctoral Training in Energy Demand (Grant No. EP/L01517X/1) and supported by Simble Solutions Limited, an innovative, Australia-based SaaS specialist.

References


[1] The Open Data Handbook

[2] OFGEM - UK Energy Bills, Prices and Profits

[3] Department of Energy and Climate Change. (2014). Smart metering equipment technical specifications version 1.58

[4] For example, some energy customers may be less concerned about price and more interested in their energy coming from renewable sources.

[5] Competition & Markets Authority. (2016). Energy market investigation


Glossary


Unstructured data: While logically structured data can be automatically machine-read through inferences and pattern detection, unstructured data appears as messy to a computer and it may be harder to extract the required data. An example of structured data is an xml-formatted file, while PDF documents are considered unstructured - the underlying code dictates visual positioning rather than structure.


Page crawling: Programmed bots can be used to index the information on a page to enable extraction and summation of machine-readable text. Internet search providers use spiders to crawl millions of webpages, providing ranking of relevance to different terms and presenting the pages in order of relevance when a user inputs a search term.


Application programming interfaces (APIs) are tools for building new programmes utilising an existing framework or database (for example). This allows developers to use underlying data but adding new functionality or interpretation of the data.


Price signals: Time-of-use data recorded by smart meters can be valuable for more than just suppliers and grid controllers. Through each day, there will be periods where energy is in low demand but there is high supply, and conversely times of the day when energy is relatively scarce. In other markets, price signals are used to influence consumption, and in the energy market this could mean that energy becomes cheap to use at certain times of the day to motivate customers to use energy in these periods.