In business, few words sound more reassuring than net zero. For years, it’s been the promise pinned to corporate strategies, investor reports, and brand campaigns. But beneath the slogans and offset pledges lies a quieter, more technical truth: most organisations don’t actually know how accurate their carbon data is.
As Owen, co-founder of Fynlo AI, put it, “Fleet managers want to do the right thing. They’re not trying to fudge the numbers. But the data is fragmented, the variables are endless, and most businesses have no real way to validate whether their carbon reporting is anywhere near correct.”
That problem isn’t unique to transport or logistics. Across industries, carbon reporting is often a best guess built on outdated assumptions. It’s an approach that might once have been tolerated as a compliance exercise, but it’s rapidly becoming a credibility issue.
For companies with fleets of vehicles, the problem starts at the tailpipe. Every litre of fuel burnt is a data point, but the factors influencing it are vast – payload, terrain, driver behaviour, traffic conditions, maintenance, weather. Traditional reporting models take a rough average and multiply it by mileage.
It’s quick, it’s easy, and it’s wrong.
“Accuracy is everything in this space,” Owen explained. “Fleet managers want to do the right thing. They’re not trying to fudge the numbers. But the data is fragmented, the variables are endless, and most businesses have no real way to validate whether their carbon reporting is anywhere near correct.”
Those inaccuracies cascade upwards. Fleet managers make replacement decisions based on incomplete insights. Boards set sustainability targets that may not reflect reality. Investors publish ESG reports that rest on assumptions rather than verified data. And when regulators or partners ask for evidence, the industry is left hoping that “close enough” is still good enough.
Owen summed it up simply: “If you can’t measure it accurately, you can’t manage it effectively.”
The shift towards verifiable data isn’t just an environmental issue, it’s an operational one. For freight, transport, and logistics companies, fuel efficiency is margin efficiency. A system that helps you understand real emissions doesn’t just improve reporting—it can reveal inefficiencies that save money.
More broadly, as regulators tighten frameworks around Scope 1 and Scope 3 emissions, and as customers begin to ask harder questions of their suppliers, accuracy becomes strategic. Credibility in sustainability reporting is fast becoming a competitive advantage.
The same is true for public trust. A growing number of organisations have made public net-zero pledges without the infrastructure to measure whether they’re on track. The result is scepticism; and in some cases, accusations of greenwashing. Getting the data right isn’t a branding exercise, it’s an ethical obligation.
Technically, the tools to improve accuracy already exist. Advances in AI modelling, sensor data, and predictive analytics can now account for real-world variability in a way that static spreadsheets never could.
Many businesses are still trapped between aspiration and action, unsure where to start or how deep to go. That’s where structured approaches, like an AI Design Sprint, can help: by taking a complex problem like emissions accuracy and breaking it down into solvable, testable parts.
“The value of the sprint wasn’t just speed,” said Owen. “It forced us to get outside our heads. We went from a list of 20 different ‘maybe’ ideas to one concrete product direction that made sense both technically and commercially.”
But the mindset shift is the real hurdle. Organisations must move away from seeing carbon reporting as an obligation and start seeing it as intelligence. That is data that can inform better decisions, shape procurement strategy, and prove genuine progress.
Owen and his team at Fynlo talk about accuracy the way others talk about innovation or growth. For them, it’s not a marketing term, it’s the foundation for meaningful action. “We’re not saying stop using internal combustion engines tomorrow,” he said. “We’re saying use them better. Measure them better. Because if you don’t know what’s really happening, how can you improve?”
That mindset captures the bigger shift now unfolding across industries. As AI gives us the tools to model and measure with greater precision, the question isn’t whether we can achieve net zero, it’s whether we’ll have the data discipline to prove it.