The AI controller: what AI in Smartbooks does for your finance team

The AI controller steps in exactly where the role is under pressure: reports look backward while management wants to look ahead. There is more data than ever, but every question means an export, a pivot table and an afternoon of work. Budget deviations across departments only surface once the damage is done. And meanwhile, the objective watchdog guarding the numbers competes for the same hours as the strategic business partner management actually asks for. The watchdog usually wins.

That's exactly where the AI controller in Smartbooks comes in: not a chatbot giving generic answers, but AI that works directly on your own numbers, down to transaction level. In this article, we show per pain point what that means in practice.

Checking: from IC mismatches to incorrect bookings

This is the watchdog role, and it's exactly the part that automates best.

Intercompany reconciliation that goes beyond matching

Most consolidation software can detect intercompany transactions. Smartbooks AI goes a step further: it doesn't just find the mismatches, it also suggests next steps. Is a counter-booking missing at a subsidiary? Is there a currency difference in play? The AI analyzes the difference and tells you what needs to happen to resolve it, instead of leaving you to dig through journal entries across multiple entities.

Anomaly detection in your data

Incorrect bookings often only surface when it's too late. The AI controller scans your data for deviations: entries that don't fit the pattern of a GL account, amounts that stand out, or cost lines that suddenly double. Budget deviations across departments and entities also surface early, not just at the quarterly review. Practical checks become simple too, such as verifying whether all invoices were actually sent. You catch errors before the close, not after.

Analyzing: answers without building a report first

This is where the shift happens from looking backward to working data-driven.

Ad-hoc insights down to transaction level

“What were the marketing costs per entity in Q2, and why are they rising?” A question like that used to mean building a report or running an export. Now you just ask. The AI digs through your numbers, drills down to transaction level where needed, and comes back with a substantiated answer. Have a question but no report yet? That's no longer a blocker. The distance between data and advice shrinks from afternoons to minutes.

Asking questions at consolidated level

Because Smartbooks has transactions available at consolidated level, you can ask questions about your entire organization in one go. How much revenue does customer X generate across all entities combined? Which suppliers are used by multiple subsidiaries? Where you previously had to line up exports from multiple administrations side by side, the AI controller gives you the group-wide answer directly.

Planning: budgets, forecasts and scenarios

From describing what happened to calculating what's ahead.

Help with budgets and forecasts

Budgeting remains human work, but the AI controller thinks along with you. It helps set up your planning and business models, recognizes trends and patterns in your historical data, and suggests where your budget or forecast likely needs adjusting. You decide the assumptions and fill in the numbers; the AI makes sure you never start from a blank sheet.

Calculating investments and loans against your projection

Is management considering an investment or new financing? The AI controller works through ad-hoc questions based on your existing projection. What does a half-million loan do to your liquidity? Is that machine purchase justified given expected cash flows? You get a calculation on your own numbers, not on assumptions from a generic model. This is exactly the terrain where the controller makes the difference as a business partner.

The right metrics for your organization

Which KPIs actually matter for your business? The AI controller helps determine the metrics that are relevant for your organization, fitting your industry, business model and the questions your management asks. You then set those metrics up directly in Smartbooks, so they appear in every report.

Producing: reports and documents

Creating new reports

Setting up a new report no longer has to go through the configurator alone. Describe what you want to see, for example a P&L per cost center compared against budget, and the AI controller builds the report for you. You can always fine-tune it afterwards.

PowerPoints, Excel models and Word documents

Numbers are only useful when they land with the right people. The AI controller uses your Smartbooks data to create board presentations, Excel models for further analysis, or Word documents such as a monthly report. No more manually retyping figures, and no more errors from outdated versions.

How does the AI controller work?

The AI controller works best from within Smartbooks itself. There, the AI has the full context of your organization at hand: your entities, structure and dimensions, combined with prompts that Smartbooks has tuned specifically for finance work. You don't have to explain how your administration is set up; the AI already knows. Prefer working from your own AI environment? Your Smartbooks data can also be accessed through an MCP connection with Claude.

From watchdog to business partner

The AI controller doesn't replace the controller, but takes over the searching, the manual work and the building. In doing so, it also resolves the squeeze: the watchdog role is largely automated, freeing up time for the role management actually asks for, that of strategic business partner. The controller stays in charge: reviewing, adjusting and deciding. The difference is speed and depth. Questions that used to take a day are answered in minutes. Differences that used to surface at the audit come to light at month-end close.

Frequently asked questions about the AI controller

What is an AI controller?
An AI controller is AI functionality that supports the work of a financial controller: from intercompany reconciliation and anomaly detection to forecasting and reporting. In Smartbooks, the AI works directly on your own financial data, down to transaction level.

Will AI replace the financial controller?
No. The AI takes over repetitive searching and building work, but the controller remains responsible for judgment, interpretation and decision-making. AI makes the profession faster and more analytical, not obsolete.

How does AI help the controller become more of a business partner?
By automating the control work: reconciliations, flagging deviations, building reports. The time freed up can go to analysis, scenarios and strategic advice to management.

What data does the Smartbooks AI controller work on?
The data in your Smartbooks environment: general ledger, transactions, budgets, projections and consolidated figures across all your entities. The AI looks at your actual numbers, not at generic benchmarks.

Want to see the AI controller in action?

Book a demo and ask your first question to your own numbers.