Italian CFOs look at the same data. They use the same ERP systems. They close their months with the same manual reconciliations. Yet in 2025, their organizations find themselves at radically different points on the financial maturity curve: some respond to market shocks in hours, others in weeks. The gap does not depend on team quality. It depends on architecture.

The KPMG Digital Finance Maturity 2025 survey of Italian companies with revenues between 20 and 500 million euros draws a clear picture: 59% of organizations still operate in the first two stages of financial maturity, with predominantly manual processes or basic ERP. Less than one fifth has reached advanced stages. Only 12% already operates with continuous financial intelligence.

The question is not where others stand. The question is: what does it cost, in euros, to stay where you are?

The maturity model: where Italian finance stands

The Finance AI Maturity Model is structured across five stages, each with precise operational characteristics and a quantifiable residual gap. The Italian distribution in 2025 shows concentration in the first two levels, with a thin but rapidly growing tail toward advanced levels.

• Stage 1 – Manual (31%): Excel-based processes, non-integrated data, manual close cycles. Main gap: data integration and elimination of manual reconciliation.
• Stage 2 – Basic ERP (28%): ERP operational, monthly reporting, no forecasting. Main gap: structured forecasting and finance-operations integration.
• Stage 3 – Integrated BI (24%): working dashboards and BI, forecasting still manual. Main gap: continuous intelligence and automated scenario planning.
• Stage 4 – Analytics (5%): automated forecasting and structured scenario planning. Main gap: AI layer, governance and continuous decision systems.
• Stage 5 – AI-Native (12%): continuous financial intelligence and integrated AI governance. Main gap: perimeter expansion, multi-entity and advanced governance.

The most relevant data point is not the static snapshot. It is the dynamic: organizations at levels 4-5 are building structurally superior decision-making capabilities, with response times to market shocks three to five times faster. The advantage does not reset: it accumulates quarter by quarter.

Stages 1-2: the real cost of manual processes

In manufacturing and distribution SMEs operating with traditional accounting systems and Excel models, the cost of information delay does not appear in any P&L line. It appears, instead, in the margin that is never defended.

Vedrai Observatory defines this phenomenon the Economic Visibility Gap: the distance, in time and signal quality, between what happens in operations and what becomes visible in finance. For a company with a 10% EBITDA margin and 50 million in revenue, a visibility gap of 12 days generates, on comparable benchmarks, an expected loss in the order of 400,000 to 800,000 euros per year.

The average cycle between an operational event and analysis available to management lasts between 8 and 15 days in manufacturing SMEs. In enterprises with multiple ERPs, it often exceeds 20 days. Every day of latency is a day when a corrective action is not taken, or is taken on the wrong data.

For a company with 50 million in revenue, a 12-day visibility gap is worth between 400,000 and 800,000 euros per year. Not in any P&L line, but in margin that was never defended.

The cost is not only operational. In multi-ERP companies, according to PwC Finance Effectiveness Benchmark 2025, the cost-to-serve of finance averages roughly double compared to organizations with integrated architecture. You pay twice: once for data that arrives late, once for the team that produces it.

Stage 3: why BI alone is not enough

Stage 3 is the most subtle trap in the maturity model. Organizations that have invested in Business Intelligence have working dashboards, automated reports, faster close cycles. They believe they have solved the visibility problem. They have not.

24% of Italian companies find themselves in this condition: more data available, but still no ability to anticipate. BI photographs the past with greater speed and detail. It does not produce intelligence about the future. Forecasting remains manual. Scenario planning is episodic, driven by individual initiative, misaligned with real operational data.

The real competitive leap occurs between stages 3 and 4, when finance stops aggregating data and starts producing decisions. In this transition, response time to market shocks drops from weeks to hours. Scenario planning frequency shifts from quarterly to monthly. The strategic planning cycle shortens by more than a third.

This is where structural advantage forms. A company still at stage 3 when competitors reach stage 4 does not lose a function: it loses the ability to compete at the same speed.

Stages 4-5: the numbers behind AI-native organizations

Organizations operating at stages 4 and 5 share measurable characteristics that go beyond reporting quality. Available benchmarks on comparable samples draw a precise profile.

• Information latency: from 8-20 days per operational event (reactive finance) to real-time updates (continuous intelligence).
• Forecast error: from 15-25% on a quarterly horizon to a significant reduction through integrated forecasting.
• Scenario planning time: from weeks with a manual process to hours through instant propagation.
• Board reporting frequency: from monthly or quarterly to weekly or continuous.
• Planning cycle: from an annual baseline to a reduction of more than 35%.
• Investment payback: approximately 12 months from full deployment.
• Margin recovery: 1-3 percentage points on comparable benchmarks.
• Liquidity buffer: estimated reduction of 15-20%.

The 12-month payback is driven primarily by two factors: the reduction of undetected margin leakage, meaning variances detected only at closing when corrective action is already costly, and working capital optimization through rolling cash forecasting at 3-12 months.

According to EY CFO Imperative 2025, organizations that have completed this transition report a measurable change in the quality of board discussions: fewer meetings dedicated to reconciling numbers, more time comparing scenarios.

Diagnosing your gap: five operational questions

The maturity model is not a descriptive tool. It is a diagnostic tool. Five questions allow you to precisely locate your organization's finance architecture and estimate the residual economic gap.

• How long does it take between a relevant operational event and its visibility in finance?  Less than 24 hours indicates stages 4-5; 1-3 days an advanced stage 3; 8-15 days levels 1-2; over 15 days level 1, with risk of structurally wrong decisions.
• Is forecasting integrated with real operational data or based on static assumptions?  If the forecast is updated manually once a month, the organization is at stage 2-3. Continuous forecasting integrated with operational data is the marker of stage 4.
• How long does it take to produce a what-if analysis on a price or volume change?  Weeks indicate levels 1-2; days stage 3; hours or minutes levels 4-5. Scenario planning time is the most direct KPI to measure the distance between data and decision.
• How many hours per week does the finance team spend extracting, cleaning and reconciling data?  In reactive organizations, 60-70% of the controller's time is dedicated to data wrangling (EY, 2025). In organizations with continuous intelligence, this share drops below 20%.
• Is variance attribution automatic or does it require manual post-hoc analysis?  Automatic identification of variance causes is the marker of stage 4. Its absence means every anomaly is detected late and analyzed at disproportionate cost.

The five answers converge toward an estimate of residual economic gap. A stage 1-2 organization with 50 million in revenue can estimate an Economic Visibility Gap in the order of 400,000 to 800,000 euros per year. A stage 3 organization has reduced this loss, but not eliminated it: margin leakage from variances not detected in real time remains a structural source of erosion.

Competitive advantage does not belong to those who have more data. It belongs to those who have eliminated the time between data and decision.

Organizations that do not complete the transition to continuous financial intelligence in the next two or three years will find themselves operating with an information architecture that is a generation behind. The gap will manifest first in decision quality, then in economic performance. The cost of inertia, measured in accumulated Economic Visibility Gap, is already quantifiable today.

The question for every CFO is not whether to invest, but what the real cost of not doing so is.