How to launch a transaction-based carbon footprint calculator in three steps
January 20, 2022 16 minutes read
Transaction-based carbon accounting allows banks to help their customers become aware of their climate impact. Launching such a real-time carbon footprint calculation service on top of a digital bank opens up considerations related to architecture, reliability, and data security. Whether you decide to develop your sustainability service in-house or partner with a third-party solution provider, it’s worthwhile to understand the basic elements of a carbon footprint calculation service.
In our previous articles, we have discussed the methodology behind carbon accounting, and the reasons why banks and issuers should want to fight climate change in the first place. We’ll now share our insights on designing and implementing a transaction-based carbon footprint calculator.
By reading this article, you’ll learn:
- The basics of payment transactions as the basis for carbon accounting
- How to launch a transaction-based carbon footprint calculator – in three steps
- Examples and best practices for implementation
The nuts and bolts of payment transactions
A modern carbon footprint calculator is a standalone, real-time service. It’s connected to core banking or other issuer systems over APIs. In the background, it’s essentially powered by a carbon accounting engine that runs calculations.
Payment transaction data allows issuers to make accurate lifestyle estimates. This is done by tracing consumers’ transactions and assigning them carbon footprints. This is called the “bottom-up approach”. You can read more about it in our blog discussing the methodology of carbon footprint calculations.
By understanding the key elements of users’ transactions, you can better design an accurate and comprehensive carbon footprint calculator.
Delivering a carbon footprint calculation user experience essentially comprises three steps. We’ll discuss these in more detail throughout this article.
- Collecting data based on different dimensions of customer transactions. This is the basis for a carbon accounting engine.
- Using the engine to calculate and make sense of the data. This is done by comprehensively mapping the transaction data to carbon footprint databases.
- Making sure the carbon accounting engine generates relevant insights to users based on accurate calculations and add-on features.
Below, you’ll see how the carbon footprint of a payment transaction is calculated, and what is included in the calculation.
Payment transaction data usually follows a standard format, such as ISO 8583. This makes it easier for issuers to start building their carbon footprint calculator.
Step 1: Collect transaction data from multiple sources
Most consumers today manage their personal finances online. This expands the opportunities to understand their buying, investment, and saving habits in depth. Customers of the digital banking era typically use several accounts and financial services in multiple banks. All of this justifies the idea of integrating a carbon footprint calculation into the payment transactions of an individual – even across their accounts.
When building an engaging carbon footprint calculator, it makes sense to combine many data sources from card payments and everyday bank accounts. You should aim to identify those data sources that have the most detailed information available.
Here’s an overview of data sources that can boost the accuracy of a carbon footprint calculation model.
Card transaction data
This data is fetched either from a card scheme or a card management system.
When the card transaction data is retrieved from a card scheme, it comes in a clearing file format. It contains a broad set of data points such as merchant IDs and addresses. The clearing files of special cards such as fleet cards also contain product-level data.
When the data is retrieved from a card management system, it also provides an extract with less data. Usually, this extract is still sufficient for a carbon footprint calculation.
Open banking data
Real-time open banking data from other banks enables issuers to create a 360-degree view of a person’s spending. By extension, they can create the user’s carbon footprint.
The challenge with using open banking data is that it may be difficult to harmonise data from multiple sources. This is because different banks tend to represent transaction data in different ways.
Line item data from merchants or invoices
Line item data typically has better quality and insight than card transaction data. Electronic invoices tend to follow regional standards, such as the EU standard. This enables issuers to ingest invoice data in a standard and structured format.
Other data sources: KYC data and purchasing patterns
In addition to sourcing transaction data, issuers will benefit from combining selected user data. This makes the calculation more relevant and personalised.
The issuer typically has access to the customer’s Know-Your-Customer (KYC) data, such as their age and location. The issuer can collect more information about their customers’ lifestyles when they subscribe to a carbon footprint calculation service. On those services, the issuer can ask e.g. “how many people live in your household”, “how many square metres does your apartment have”, and other lifestyle-related questions when new users get started with the service.
In addition, you can identify specific patterns from transaction data, energy, or electricity invoices. Upon user consent, the issuer can use this data to make the carbon footprint calculation more accurate.
Step 2: Map the transaction data to carbon footprint databases
Now, how do you build an engine to calculate and make sense of the data you collected? By comprehensively bringing together the transaction data and carbon footprint databases. Here’s how it works.
How to create a robust carbon footprint calculation system
Securing relevant data from various sources is the fuel that powers a credible carbon footprint calculation. Efficient performance requires a robust calculation system. It needs to assign the most accurate possible footprints to customers’ transactions.
You need to define how you’d like your calculator to use customer transaction data. Then, you need to define how you’d like the calculator to use environmental databases in the calculations. This tends to be the hardest part of building a credible carbon footprint calculator. If done well, the calculator can easily extract a relevant footprint from any given transaction, regardless of the transaction’s level of detail.
When it comes to designing a successful mapping logic, there are two key things to consider:
- The logic of bridging transaction units with the carbon footprint database categories.
- The approach to filling eventual gaps in data.
A typical implementation features a dynamic lookup table to help interpret the data. It captures both transaction data units and the CO2 equivalent per unit of money spent in one place. Further inputs such as personal lifestyle data may be introduced as separate factors. This will have an impact on how your custom-build calculation engine will look like.
Choosing parameters for calculating carbon footprints
You can use various parameters to calculate carbon footprints. These include for example transaction-level, merchant-level and product-level parameters. A carbon footprint calculator’s accuracy depends on the parameters’ level of detail.
Transaction-based carbon footprint calculators typically rely on the Merchant Category Code. MCC is a global standard for grouping together merchants that receive a transaction. It provides a reasonable level of detail and is fairly easy to map. Another benefit is that MCC codes are globally standardised. This makes it relatively easy to run databases containing them.
Let’s look at an example where a transaction is mapped with four parameters:
- The MCC is 5411 – Supermarket
- The merchant name is Tesco
- Product category level is Vegetables
- Product-level information is Cucumber
Step 3: The user interface brings your carbon footprint calculations to life
The ultimate purpose of a carbon footprint calculator is to share concrete insights with the end users. This can be achieved for example through a standalone app or an online banking portal. However, the setup is not the main concern. It’s all about providing an exceptional user experience.
Building the user experience for your carbon footprint calculator
Issuers can package emissions data from calculators into engaging user experiences in many ways. Besides showing emissions in black and white, issuers can show deeper insights based on users’ consumption data. This is something that engaging personal finance management apps already do.
Some apps enrich the user experience by providing customers with a comparison of their personal emission levels against national or demographic averages. There’s an example of how My Carbon Action, a transaction-based carbon footprint calculator, does this.
Another exciting dimension is the ability to encourage users with specific calls to action. Using transaction carbon footprint data, you can for example:
- Tell users how their emission levels stack up against peer groups or personal emissions goals
- Showcase how your customers’ consumption habits have evolved over time
- Provide various loyalty-building initiatives such as campaigns, challenges, or compensation options, such as carbon offsetting.
Designing a real-time carbon footprint calculator
A robust architecture is the starting point for an enjoyable user experience. Your engine must have a high-enough availability, frequency, and response time. Listing transactions should never take longer than one second.
How do you achieve this? A modular design is an excellent option. Using small APIs of different services in parallel helps you get the desired performance. If you rely on a third-party calculator, the integration points should be built on standard APIs.
You should aim for a design that enables sophisticated data processing. This means handling calculations even when some fields of transaction data are missing. In these situations, the calculations should adapt accordingly to produce a less detailed result.
Ensuring compliance without compromises
Let’s talk compliance. Banks and financial institutions are under heavy regulatory scrutiny. Therefore, they need strict compliance procedures. These procedures are also relevant for transaction-based digital carbon accounting.
Fully-compliant digital carbon accounting is founded upon consent management. When the user submits their unique and traceable consent to an issuer, they permit the issuer to forward their data into the carbon footprint calculator. The data should be used in a form and scope that’s clearly specified upon consent.
GDPR requires issuers to anonymise customer data used in third-party digital carbon accounting engines. Identifying users based on their carbon footprint data mustn’t be possible. In case of a data leakage, for example, no data can be linked back to any individual user.
Tokenisation of user data is one approach to anonymisation. Each user is assigned an anonymised user ID. Transaction data is then connected to this anonymised ID instead of identifiable user information. This way, data can be shared outside the bank environment to a third-party calculation service.
The PCI-DSS standard regulates the handling of payment card data. It requires that issuers maintain card numbers or identifiers within a secure environment that fulfils the PCI-DSS compliance requirements. Those may relate to network security and card data encryption, for example.
My Carbon Action helps you and your customers become more sustainable
My Carbon Action is more than a carbon footprint calculator. Banks and issuers can integrate it into their services to help their customers cut and compensate for their CO2 emissions.
Learn more about My Carbon Action and make sure you are among the future thought leaders in banking!
Watch this webinar recording to learn how you can help your customers live more sustainably, and why you should do it.