Inside Atmoz's decision to rebuild its climate data platform from scratch for AI

Most software companies spent the last two years bolting AI features onto products that were already working. Atmoz did the opposite. In October 2025, the Stockholm-based climate data company paused development of its existing platform entirely and began rebuilding it from the ground up.

The story made headlines this month when Impact Loop reported on the decision↗, noting that Atmoz had halved its development team in the process and quoting Chief Commercial Officer Jessica Guter, who called it "a difficult decision to make when you have a product that works and customers flowing in." This article sets out the reasoning behind it, what the company built instead, and what it means for businesses that rely on climate data to make decisions rather than simply report a number.

Why pause a growing product

Atmoz helps companies turn accounting and operational data into climate data. By late 2025, the business was growing quickly, but so were the expectations of its customers. They were larger, more complex, and asking for a level of traceability, integration and decision support that the existing platform had not been built for.

At the same time, the pace of progress in generative and agentic AI meant that a problem the climate data industry had struggled with for years, the lack of detailed and reliable activity data, could finally be solved in a way that used to be too costly and too labour intensive.

Rather than adding AI on top of the existing product, Atmoz chose to stop and rebuild. Eight months later, in June 2026, the company launched a new platform in which AI is not a feature. It is the engine that runs the product.

AI changed what was possible, not just what was efficient

"We realised we had reached a point where AI was no longer just about making existing workflows more efficient," says Axel Ehrling, Chief Product Officer at Atmoz. "For the first time, we could let AI agents carry out large parts of the work that used to require manual analysis, data collection and consulting support. At that point, the question stopped being how to add AI to the product, and became whether the product was still built for this new reality."

That shift in thinking is what separates this rebuild from a typical AI feature launch. Instead of layering new functions onto existing technology, Atmoz built an entirely new architecture in which AI agents work directly on identifying, classifying, analysing and quality assuring emissions data.

What a smaller team actually means

Impact Loop's coverage of the rebuild focused on one striking detail: Atmoz cut its development team in half.

The reduction was a direct consequence of the new architecture. Work that previously required teams of engineers and analysts, manually building integrations, cleaning data and checking outputs, is now carried out by AI agents within the platform itself. A smaller team now focuses on designing, training and supervising those agents, rather than performing the underlying data work by hand.

For a company handling data that customers use for audits, procurement and investment decisions, this is an important shift. Atmoz's answer is that trust is built through traceability and quality control, not through headcount. Every output the AI agents produce can be traced back to its source, which is what allows a leaner team to support a growing and more complex customer base.

Part of a broader shift in climate tech

Atmoz is not alone in making this kind of move. Other sustainability-focused software companies are rebuilding significant parts of their operations around AI rather than adding it incrementally. What sets Atmoz apart is the scope of the change: a full platform rebuild, a new architecture built around AI agents, and a shift in target customer, all completed within eight months.

The pattern points to a wider trend in climate technology. As the tools available to automate data collection and analysis improve, companies that built their platforms before generative and agentic AI existed are increasingly deciding that incremental updates will not close the gap. Starting again, while disruptive, may be the more competitive path.

From generic emission factors to activity data

Much of the climate calculation industry has historically relied on template values and generic emission factors. That approach works for compliance reporting, but it limits what businesses can actually do with the data, particularly when it comes to operational and financial decisions.

The new Atmoz platform uses AI to automatically convert financial and operational data into detailed activity data with higher quality and full traceability. In practice, this means:

• Automated data collection and classification.
• Higher quality, more consistent emissions data.
• Full traceability back to the original source.
• The ability to analyse emissions at a far more granular level than before.

"Climate data has long been something companies collected in order to report," says Ehrling. "We are now seeing companies want to use that same data to make decisions about purchasing, suppliers, investments and risk management. For that to be possible, the data needs to hold a completely different standard of quality than the market has historically accepted."

Why the market is asking for AI, but buying on results

Atmoz ran a limited soft launch of the new platform in spring 2026 to gather feedback ahead of a wider rollout. Interest came from companies across industry, real estate, retail, technology and services, and in almost every case, the conversation started with what the new AI based capabilities made possible.

"A year ago, the market was cautious about AI. Today, it is the opposite. AI is often the reason companies get in touch in the first place," says Jessica Guter, Chief Commercial Officer at Atmoz.

"But at the same time, many companies are asking hard questions about how AI is actually being used. No one can afford hallucinations or inaccurate figures when the data underpins decisions, procurement and audits. The first thing companies want to understand is how we guarantee quality, and that is where trust is won or lost. AI is what starts the conversation, but it is the results and the traceability that convince people."

Guter also points to a broader shift in how businesses use climate data. "A few years ago, these conversations were about sustainability reporting. Today, they increasingly focus on business risk, supply chains, cost and competitiveness. Climate data is becoming part of financial governance rather than a separate sustainability topic."

A strategic choice, not just a product decision

The move to AI was not only a technical rebuild. It also marked a change in Atmoz's strategy and target audience. Rather than focusing on general reporting, the company has shifted its attention to larger organisations that use emissions data as a tool for governance and decision making.

"We chose to pause a growing product in order to build for the market we saw emerging," says Fredrik Bentlinger, founder and CEO of Atmoz. "It was a significant decision, but we saw that AI created the opportunity to deliver a completely different level of quality and value for our customers than had previously been possible."

Atmoz began onboarding new customers on the platform from summer 2026.

About Atmoz

Atmoz is an AI driven platform for climate data and climate calculations. Headquartered in Stockholm, the company helps more than 200 customers measure, reduce and remove carbon emissions.