In the CFO’s office, STACKS raises €19.6 million and bets on the battle for the infrastructure layer

The transformation of the finance function has historically been structured around two complementary dynamics. On one side, the modernization of ERP systems, designed to unify accounting, procurement, treasury, and reporting within a single integrated backbone. On the other, the rise of FP&A platforms, offering more granular forecasting, collaborative modeling, and planning processes gradually moving away from Excel.

In recent months, however, a third front has begun to emerge. It is more technical and often more decisive for the operational reality of finance teams, because it focuses less on planning and more on execution. It does not necessarily replace the ERP, but instead seeks to take control of what makes it usable on a daily basis: transactional data, its consistency, and the ability to automate workflows in a heterogeneous environment.

The problem no one truly “owns”: fragmented finance

Within most organizations, finance rarely operates within a single system. Transactions are spread across ERP systems, spreadsheets, vertical tools, data lakes, legacy platforms, and integrations built incrementally over time. This fragmentation creates a series of workarounds. Teams spend a significant amount of time reconciling and reclassifying information, and a large portion of the effort goes not into analysis but into reconstructing a shared version of the truth.

It is not a glamorous issue, but it is the prerequisite for everything else. As long as the underlying data is not stabilized, AI remains a surface-level assistant, largely confined to text generation, drafting support, or high-level summaries. In an environment shaped by financial close and audit requirements, the priorities are different: reliability, traceability, and reproducibility.

The “infrastructure layer”: the new object of desire

Stacks presents itself as an AI platform for accounting and corporate finance teams, built around the idea of creating a data layer directly connected to financial systems in order to produce a single, coherent, and usable financial view. This layer aims to normalize data and make it immediately actionable for automation. The ambition is no longer to plan better, but to execute more cleanly.

Why agentification attracts investors, and sometimes worries teams

Stacks emphasizes the use of “agents” capable of automating workflows across the finance stack. This is where the topic becomes strategic. In an accounting context, an agent cannot rely on approximation. It must produce outcomes that are verifiable, documented, and compatible with internal control requirements and often with external audit standards.

The traction of this segment stems from a simple convergence. On one side, finance departments face increasing pressure: faster closing cycles, higher reporting quality, and a stronger need to explain performance. On the other side, the technological building blocks are finally becoming available: more standardized integrations, more mature data infrastructure, models capable of handling context, and a growing demand for automation that goes beyond traditional RPA.

The tension point lies in governance. Companies can accept a copilot that “suggests.” They are far less comfortable with a system that “acts” without guardrails. This is what separates product messaging from real-world adoption: the question is not whether the agent is impressive, but whether it is controllable.

The real question: who will control the value chain above the ERP

The battle over the infrastructure layer is not purely technical. It touches on data ownership and orchestration power.

If a platform becomes the operational entry point for finance teams, it can gradually capture functions that were previously considered native to the ERP or handled in Excel. Conversely, large ERP vendors have clear incentives to reabsorb these use cases by integrating more AI and automation into their own suites. Between these two dynamics, a grey zone is emerging: the layer that connects to everything, unifies data, explains it, and triggers actions.

Stacks as a symptom of the CFO shifting from production to arbitration

Perhaps the most interesting consequence is the redistribution of time. When finance teams spend fewer hours simply making the numbers “hold together,” they can discuss them differently. In practice, AI in the CFO’s office does not primarily replace people; it reduces certain manual tasks. This, in turn, implies a shift in governance and supervision.

A trajectory financed within twelve months

London-based Stacks develops an AI platform for accounting and corporate finance teams designed to unify data scattered across ERP systems, spreadsheets, and legacy platforms, and to automate workflows through agents. The company reports more than thirty customers, including Future plc and Epidemic Sound. It was founded by Albert Malikov, formerly of Uber and Plaid.

Stacks has announced a €19.6 million Series A led by Lightspeed, with participation from EQT Ventures, General Catalyst, and S16VC, twelve months after a $12 million seed round. In total, the company says it has raised $35 million to date.