Carbon accounting finds its feet

The first step towards net zero is calculating how much carbon we emit today and where. Carbon accounting start-ups are rising to the challenge.

Calculating the cost of greenhouse gas emissions is hard, especially for companies. Do you start from the moment your factory lights are switched on or when your employees get into their cars to commute to work? Do your responsibilities end with the shipment of the goods from their factory gates, or does it matter how much of the end product is wasted or recycled?

These aren’t just philosophical questions. Figuring out how to measure carbon costs is crucial to controlling climate change. It’s also necessary in order to comply with a tightening net of rules from governments, investors, and stock exchanges, not to mention the changing attitudes of consumers. This is also an essential first step to reducing emissions and, in due course, to achieving the hallowed “net zero” status. 

Companies have the raw data for their production facilities and supplier relationships, but have struggled to harness technology, automation, data analytics and climate science to turn this into meaningful information necessary to guide them through decarbonisation. This is where a new breed of carbon accounting start-ups can help.

“There has been a culture of spreadsheet-driven carbon accounting, with companies getting a number, putting it in a report once a year and not doing anything with it. Organisations did not have a solution for how to decarbonise,” says Mauro Cozzi, cofounder of Emitwise, a carbon accounting start-up whose backers include former Uber CEO Ryan Graves.

Indirect emissions from a company’s value chain – known as “Scope 3” emissions – are particularly tough to calculate.1 They cover the carbon footprint of everything from business travel and employee commutes to company investments and purchased goods and services. For companies including Amazon, Nike, Ford and Apple, reported Scope 3 accounts for well over half of their total greenhouse gas emissions, indicating a significant lack of sustainability across their supply chains (see chart).

Start-ups to the rescue

However, it is a difficult process.

“Companies can have data on the thousands of suppliers, hundreds of facilities, millions of raw materials that go into their products,” says Cozzi. “If you want to get precise enough to manage something, you cannot do it at human speed.”

This is where technologies like machine learning, data science and software engineering can be leveraged, to put carbon accounting principles into code, automate it at scale, and adjust to the ever-evolving climate science.

Carbon accounting start-ups, including Emitwise, collectively raised nearly USD300 million in 2021 – six times the level from 2020.2

Start-ups are bringing two capabilities to the market. First, they are using automation and data-sharing widgets to gather and harmonise disparate data. Second, they are working out the real emissions impact of companies’ business activities – their carbon footprint – covering both activities, like miles travelled by a transport fleet, and financial data, such as calculating carbon emissions related to buying specific products, based on the GHG protocol, the world’s leading greenhouse gas accounting methodology.

“The principle is that the data lives somewhere in the business. It might be in a nice and structured database, or it might be in a randomly placed spreadsheet on a desktop,” says Cozzi.

We make it as easy as possible to pull [together] the data regardless of where it lives. If it is a structured system, we can go and plug in through an application programming interface (API). If it is messy, unstructured data, then we offer a simple way for somebody to upload it.

Not all oranges are equal

Next, the numbers need to be turned into credible estimates. “The data from your business will tell us ‘I bought ten oranges last week from a supplier in Florida’. We have the brain that says your purchase of an orange from Florida on average produces 10kg of carbon, whereas an orange from Chile might produce 20kg,” says Cozzi.

That knowledge comes from looking at all the factors that have gone into producing each orange – including agricultural techniques, fertilisers and transportation methods used – information which may need to be sourced from third parties. The platform also provides visual dashboards that can help companies track progress over time and identify emission intensities in their supplier network. 

PlanA.Earth, a Berlin-based carbon and ESG reporting start-up, pulls in existing data from companies, whether from data warehouses or enterprise resource planning (ERP), and works out the emissions profile using climate science tools. For Danish fashion brand GANNI, for example, PlanA.Earth has calculated the total carbon footprint, with a specific focus on the materials it purchased, such as leather, cotton and nylon, and their respective carbon intensity. Through the partnership, GANNI was able to set a target of cutting carbon emissions by 45 per cent by 2025, adopt a tailored strategy to achieve this and fully automate its data input, emissions calculations and external reporting.

San Francisco-based Watershed, a spinout of fintech company Stripe by the people who built its carbon management tools, offers a ‘carbon data engine’ that quantifies emissions down to the line item, including more contemporary dynamics like remote work, cloud computing and cryptocurrencies.

It might seem surprising that start-ups are signing up corporations at such a rate, rather than companies doing the work themselves. But companies might be too close to realise the value of data resources they are already sitting on and may need more climate and data science insights to turn numbers into realistic quantified figures that can feed into more sustainable practices.

“You need to take a white canvas and build from there,” says Lubomila Jordanova, founder and CEO of PlanA.Earth. “There is a lot of science you need to embed in the product, a lot of unique knowledge.” Existing in-house accounting and data departments are rarely able to do this alone, she explains, due to intricacies related to geography, industry and the intensity of emissions. 

As outsiders, start-ups can also help companies realise the value of data they are sitting on but might not see as relevant to decarbonisation, she adds. “Automotive companies have data on suppliers in supplier tracking systems, for instance. It is our job to identify what data points are needed and then connect our software to them”.

Companies might also need external experts to help them work together, including with competitors, such as sharing data and collaborating across the supply chain. “To achieve Scope 3 decarbonisation, you need to plug everything in, and it is really hard to get competitors to collaborate on this,” says Cozzi. This openness will be essential if companies are to truly get a handle on their full climate impact. And if regulators start to make Scope 3 reporting mandatory – as the state of California did in a recent landmark bill for large companies – they may have the choice foisted on them. Getting ahead of the game could prove to be very rewarding.

[1] Scope 1 emissions are those that a company makes directly—for example from the company vehicles to the boilers in its buildings. Scope 2 are the emissions it makes indirectly, from the generation of purchased electricity, steam, heating and cooling it consumes. 


Please confirm your profile
Please confirm your profile to continue
Confirm your selection
By clicking on “Continue”, you acknowledge that you will be redirected to the local website you selected for services available in your region. Please consult the legal notice for detailed local legal requirements applicable to your country. Or you may pursue your current visit by clicking on the “Cancel” button.