Why Dana Is the Logical Next Step for Nocloz

Since the launch of Nocloz, our conviction has been simple: Finance teams should not spend most of their time producing, reconciling, and reworking numbers. They should be able to focus their energy on understanding performance, explaining variances, informing decisions, and supporting the business.

This conviction first led us to build a Financial Business Intelligence platform designed for Finance teams: a solution capable of connecting accounting, analytical, payroll, and operational data, transforming it according to robust business rules, and then presenting it in a reliable, readable, and governed environment.

With Dana, our AI Financial Analyst, we are taking a new step forward.

But Dana is not an artificial break in our trajectory. It is the logical continuation of what we have been building from the start.

AI does not create trust. It requires it.

Since the arrival of generative AI, many companies have been asking how to use it in their financial processes. It is a legitimate question. The promises are powerful: saving time, querying data in natural language, automating certain analyses, detecting anomalies faster, and producing performance commentary in seconds.

But in Finance, one reality quickly becomes clear: AI is only useful if the data it relies on is reliable.

A model can produce a convincing answer. It can structure an analysis, write a comment, or suggest an explanation. But if the input data is incomplete, inconsistent, poorly adjusted, or misunderstood, the output will remain fragile.

In a financial environment, this risk is significant.

A poorly explained variance, an unadjusted data point, an incorrectly consolidated entity, a management rule applied differently from one month to the next: all of this can lead to the wrong conclusion. And the more fluent AI becomes in the way it responds, the greater the danger if the underlying data foundation is not under control.

That is why we believe financial AI does not start with the model. It starts with the data.

Before you can question the numbers, you need to make them usable

In many organizations, financial data already exists. It lives in ERPs, accounting software, payroll tools, Excel files, operational systems, sometimes across several countries, several entities, several charts of accounts, or several analytical models.

So the problem is not only about accessing the data.

The real challenge is making it usable.

That means collecting it, structuring it, harmonizing it, reworking it, connecting it to the right reference framework, applying the company’s own management rules, handling specific accounting entries, securing analytical mappings, normalizing certain aggregates, and maintaining that logic over time.

This is precisely the role of the Nocloz Builder.

Builder creates the reliable foundation. It transforms scattered financial material into a governed, coherent, and traceable base. It allows Finance to remain in control of the rules, adjustments, and definitions. This is not simply about organizing data to create dashboards. It is about building a foundation robust enough to support analysis, steering, and decision-making.

This is the step that makes Dana possible.

Dana is not “plugged into” raw data

Dana was not designed as an AI layer placed on top of accounting exports or Excel files.

She relies on the structuring work carried out in Nocloz. She queries a financial environment that has been prepared, governed, and designed for analysis. This fundamentally changes the nature of the answers she can produce.

Dana can explain a revenue variation, analyze a margin, explore a cost variance, identify an anomaly, help understand a trend, or guide the user toward the likely causes of a financial movement.

But she does not do this from raw and unstable data.

She does it from a foundation where Finance rules have already been applied. Entities are harmonized. Mappings are structured. Adjustments are taken into account. Aggregates carry business meaning. Users work from the same numbers, with the same definitions.

This is where the essential difference lies.

Dana does not replace the work of making data reliable. She benefits from it.

From producing numbers to augmented analysis

Historically, a large share of Finance teams’ time has been absorbed by producing numbers: retrieving data, checking it, reworking it, consolidating it, formatting it, explaining it for the first time, and then answering successive questions from management, operational teams, or investors.

Nocloz has already significantly reduced this workload by automating an important part of financial preparation and reporting.

Dana takes this logic further.

Once the data is reliable, the challenge is no longer only to display it. The challenge is to allow everyone to ask the right questions and quickly obtain clear, contextualized, and actionable analysis.

Why is this margin decreasing over this period?

Which entity explains the main variance?

Which cost item is drifting compared with budget?

Which elements are recurring, and which are exceptional?

Where should we investigate first?

These questions already exist in every organization. The difference with Dana is that teams can start getting answers immediately, without waiting for the next extraction, the next adjustment, or the next back-and-forth with Finance.

AI serving Finance, not bypassing it

We do not believe in financial AI that bypasses Finance.

On the contrary, we believe AI can only be truly useful to the business if it relies on the framework defined by Finance teams: their rules, their definitions, their understanding of the business, and their standards of rigor.

Dana was designed with this logic in mind.

She makes financial analysis capabilities more widely available across the organization, while maintaining a shared framework. Executives, operational teams, and Finance teams can explore performance from the same data foundation, with consistent rules and an aligned understanding.

The objective is not to replace financial judgment.

The objective is to make it more accessible, faster, and more scalable.

Finance keeps governance. Dana accelerates access to answers.

Builder creates the foundation. Boardroom makes it visible. Dana makes it actionable.

The Nocloz trajectory now rests on three complementary layers.

Builder creates the reliable foundation: data is collected, structured, harmonized, and governed.

Boardroom makes this information visible: teams can monitor performance, share dashboards, and steer the business from a common source of truth.

Dana makes this information actionable: users can question their data, understand variances, investigate anomalies, and accelerate decisions.

These three layers are not independent. They reinforce one another.

Without Builder, Dana would lack reliability.

Without Boardroom, analyses would be less visible and less widely shared.

Without Dana, part of the value would remain locked inside dashboards, instead of being directly available in everyday conversations and decisions.

The future of financial analysis will be conversational, but it must remain governed

We are convinced that financial analysis will become increasingly conversational.

Users will no longer want only to consult dashboards. They will want to ask questions, explore hypotheses, request explanations, compare periods, understand the causes of a variance, and obtain possible courses of action.

But in Finance, this evolution cannot come at the expense of rigor.

The future will not simply belong to AI that answers quickly. It will belong to AI that answers quickly based on reliable data, with controlled rules, within a framework governed by Finance.

That is exactly Dana’s ambition.

To move teams from a world where Finance produces the numbers, answers questions one by one, and absorbs complexity alone, to a world where financial analysis becomes more widely and more rapidly available, while still relying on a trusted foundation.

Dana is therefore much more than a new feature.

She is the next step for Nocloz: transforming reliable financial data into useful answers, at the moment decisions are made.