Glossary

Actuals are the financial results a company has genuinely recorded. They form the historical factual basis of corporate steering: what really happened, measured in euros?

Actuals serve three functions. First, as the basis for variance analysis (plan vs. actual, budget vs. actual). Second, as the starting point for forecasts and rolling forecasts, which build on current actuals. Third, as training and validation data for forecasting models.

A practical problem: The quality of steering depends on how quickly and reliably actuals are available. If the monthly close takes 15 working days, management is working with stale data. A fast close (under 5 working days) is therefore a central efficiency topic in controlling. The sooner actuals are available, the earlier the forecast can be updated and corrective measures initiated.

Beyond Budgeting is a management philosophy built on 12 principles (6 leadership, 6 process principles) that seeks to replace the classic annual budget as the central steering instrument. It has been developed by the Beyond Budgeting Round Table (BBRT) since 1998.

The process principles cover relative rather than fixed targets, rolling forecasts instead of annual budgets, demand-driven resource allocation and an event-based rather than calendar-based rhythm. The leadership principles call for decentralised responsibility, autonomous profit centres and transparency.

Important: Beyond Budgeting is neither a tool nor a software category. It is an organisational model that requires cultural change. A planning platform can support elements of Beyond Budgeting - rolling forecasts, relative targets, scenario capability - but it does not replace the organisational change process.

Widely discussed in German-speaking markets, rarely implemented in full. According to the WHU Controller Panel 2024, the leadership principles are more firmly established in many companies than expected. Full implementation of the process principles often fails on cultural resistance and IT infrastructure. Only 21 % of companies rate their own adaptability as high.

Bottom-up planning means individual business units or departments build their own detailed plans, which are then aggregated upwards. The advantage lies in operational grounding: whoever plans close to the business delivers more realistic assumptions about volumes, costs and capacity.

In practice, bottom-up rarely works in isolation. Most industrial groups run a hybrid model in which strategic targets are set top-down and underpinned bottom-up with operational detail. The challenge lies in reconciliation: when the aggregated bottom-up numbers diverge from the top-down targets, iterative alignment loops between headquarters and units are needed.

Driver-based planning models make this reconciliation easier, because changes to individual levers become visible immediately in the overall result. That accelerates the counterflow process instead of drawing it out over weeks of spreadsheet consolidation.

A budget is the detailed financial implementation of operational planning for a defined period, typically one fiscal year. It translates strategic and mid-term goals into concrete financial figures: revenues, costs, results and investments, broken down by cost centre, product or business unit.

An important distinction: budget is not the same as planning. The budget is an output of planning, the first detailed pass with financial figures for year 1. Planning additionally covers the strategic level (5-15 years) and mid-term planning (3-5 years). Using “planning” as a synonym for “budgeting” narrows the discussion and gives up the strategic steering claim.

In large companies the budget process typically ties up 3-5 months of controller capacity (Q3 through December). A common pattern: too much detailed planning across too many levels, too few decisions taken during the process.

Budget variance measures the difference between a planned budget value and the actual result. It is reported by cost type, cost centre or business unit and decomposed into volume, price and efficiency components.

What matters is the causal layer, not the variance itself. A favourable variance in material costs may result from lower purchase prices or from lower production volume. The two call for different responses. Without decomposition the number remains ambiguous.

In practice, insights from budget variances feed into the next forecasting round and improve forecast accuracy step by step. The precondition is a planning model built on the same drivers as the variance analysis.

A budget vs. actual comparison sets planned values (budget or forecast) against the results genuinely achieved. Variances are classified as favourable or unfavourable and reported in absolute or relative terms.

The variance figure alone explains nothing. Only the connection to a variance analysis shows whether a difference arises from volume, price or mix effects. In driver-based planning, every variance can be traced back to the input values that caused it. A budget vs. actual comparison at account level, by contrast, delivers only the symptom, not the cause.

In volatile markets many companies complement the classic budget comparison with a running forecast, so the reference value does not stay unchanged for twelve months. That does not change the principle: compare, analyse, act.

Business partnering describes the controller’s role as strategic advisor and decision-support partner to the business functions, going beyond pure reporting and data delivery. The business partner knows the business model, translates financial figures into operational recommendations and is permanently involved in management decisions.

Distinction from internal consulting: A business partner is not deployed project by project but structurally embedded in the business function. They build relationships, understand the operational context, and their interpretation of the numbers becomes the basis for decisions.

According to the WHU Controller Panel 2025, business partner is the most strongly developed controller role at an average of 5.7 (Likert scale 1-7). At the same time the scorekeeper role is declining (average 4.2). The problem: the transformation often fails because controllers tie up too much capacity in routine work - data collection, alignment loops, manual aggregation. Anyone spending 80 % of their time producing reports cannot be a business partner.

According to CMR 2025 (Schäffer/Wallrabe/Wöst), business partnering has “mostly fallen short of its original objective in implementation”.

The Betriebswirtschaftliche Auswertung (BWA) is a standardised monthly report produced from financial accounting, and the closest German equivalent to monthly management accounts. It shows revenue, cost categories and a preliminary result for the reporting month plus a year-to-date view. In German mid-market companies it is usually prepared by the tax advisor and is the monthly look at the numbers for many managing directors.

Distinction from planning: The BWA is purely historical. It contains no expectation, no assumptions and no forecast. The standard format is an income statement only, with no balance sheet and no liquidity outlook. It therefore cannot answer two central questions by construction: where will the year end, and will the cash last?

A practical problem: Because the BWA arrives reliably, it gets treated as a steering instrument. It is a measuring instrument. It only becomes steerable once its actuals feed a projection that varies drivers rather than cost categories, which is why the usual next step is a year-end forecast built on the current BWA figures.

Capital budgeting is the process of evaluating, prioritising and steering capital expenditure (CAPEX). Which projects, facilities or acquisitions get funded? In what order? With what expected return?

In industrial groups, capital budgeting commits substantial financial resources. Typical CAPEX ratios in machinery or the automotive industry run at 4-8 % of revenue. Bad investments have effects for years, because depreciation burdens the P&L and tied-up capital reduces cash flow.

Planning relevance: Investments connect operational and financial planning. Every investment decision has effects on the balance sheet (fixed assets, debt), the P&L (depreciation) and cash flow (payment). An integrated planning model represents these chains of effect and shows immediately how an additional investment project changes the financial metrics.

CAPEX (capital expenditures) are outlays for long-term assets: machinery, buildings, technology, acquisitions. Unlike OPEX (ongoing operating expenses), CAPEX does not hit the income statement immediately but is depreciated over the useful life of the asset.

Balance sheet effect: CAPEX increases fixed assets, and depreciation burdens the P&L over several years. As a result the earnings effect differs from the cash flow effect: the payment happens immediately (or in milestones), while the P&L charge is spread out.

Planning relevance: In industrial groups, CAPEX decisions are long-term commitments and capital-intensive. A driver-based planning model has to represent these relationships: investment decision, payment profile, capitalisation, depreciation schedule, effect on EBITDA, EBIT and free cash flow. Planning CAPEX purely as a budget line, without calculating the balance sheet and cash flow effects, gives up the steering claim.

Contingency planning is the structured preparation for events that break the regular planning frame: supply chain failures, regulatory shocks, abrupt demand collapses. Unlike ongoing forecast adjustment, it is about predefined options for action against clearly named risk cases.

Contingency planning is not the same as scenario planning. Scenarios represent alternative development paths and serve strategic orientation. Contingency plans, by contrast, are operational response plans with concrete measures, responsibilities and the thresholds at which they take effect.

The practical value rises when contingency plans build on the same driver models as regular planning. The financial impact of a disruption can then be calculated within hours, instead of having to build a special model once the crisis has hit. According to recent surveys, only 16 % of companies can calculate scenarios in under a day.

Corporate steering describes the complete control loop for goal-oriented company management: formulate strategy, derive plans from it, measure actuals (reporting), update expectations (forecast), take decisions, adjust strategy.

Distinction from controlling: Controlling supplies the information base - figures, analyses, scenarios. Steering is the use of that base for management decisions. Controlling is a function; steering is a process in which controlling, management and the business functions all take part.

Distinction from EPM (enterprise performance management): EPM describes essentially the same function but emphasises the IT architecture and tool layer. Corporate steering emphasises the decision process.

The target picture (Schäffer, CMR 1/2026): decision-makers have to be put in a position to work directly with the data, supported by unified data platforms, shared models and intelligent automation.

Data integration is the consolidation of data from different source systems into a consistent, unified data foundation. In corporate steering that typically means: ERP (financial accounting, controlling), CRM (sales data), HRIS (people data), production systems and external data sources.

In the DACH context this above all means SAP integration. Large industrial groups run SAP as their leading ERP system, often with historically grown data structures, multiple SAP instances and bespoke customising. The planning platform has to represent this complexity, not simplify it. Interfaces that only cover standard fields fail against reality.

According to the FP&A Trends Survey 2023, FP&A teams spend 45 % of their time on data collection and validation. That is primarily an integration problem: data sits in different systems, formats and states of currency. Companies that do not solve it lose analytical capacity to data logistics.

Debt service capability describes whether a company can cover interest and principal payments on its financing out of operating cash flow on a sustained basis. Banks assess it when granting credit and monitor it over the life of the loan, often tied to covenants.

It is calculated as the ratio of cash flow before debt service to the actual debt service of interest plus principal. A value above 1.0 means debt service is covered; lenders typically expect a safety margin above that. What matters is which cash flow measure is used, since definitions differ between institutions.

Planning relevance: The relevant question is never what debt service capability was last year, but how it develops under changed assumptions. A drop in revenue, a rise in interest rates or an additional investment act on the income statement, balance sheet and cash flow at the same time. That can only be answered in an integrated plan where all three statements are linked. Calculated once for the bank file, the metric is already out of date by the next meeting.

Demand planning forecasts the expected sales volume by product, region and period. It is the volume-based counterpart to value-based revenue planning: demand planning asks “How much will we sell?”, revenue planning asks “How much revenue will that generate?”

In manufacturing companies, demand planning is the starting point for the entire operational plan. The planned sales volume drives the production programme, material requirements, capacity planning, headcount planning and logistics. In an integrated model, a change to the demand forecast automatically pulls every downstream plan with it.

Distinction from sales planning: Sales planning defines quotas, territories and the measures needed to hit targets. Demand planning quantifies expected demand. The two belong together but answer different questions: sales planning asks “What will we do?”, demand planning asks “What do we expect as a result?”

In machinery and plant engineering, demand planning is particularly demanding, because long order lead times, project-based manufacturing and low unit volumes limit the statistical basis for forecasting. Here the combination of pipeline data (CRM), market indicators and management judgement is decisive.

Driver-based planning is a planning approach in which financial results are derived from operational drivers such as unit volumes, prices, utilisation rates or headcount. Instead of entering financial figures manually, the model represents cause-and-effect relationships: a change to one driver automatically pulls all dependent positions with it.

This creates two concrete advantages. First, scenario capability: when management needs a what-if scenario, the result is there in minutes, because only the drivers have to be adjusted. Second, traceability: the logic in the model makes it transparent why an outcome measure has changed.

A common fallacy: “driver-based” is often equated with “detailed”. That is wrong. A driver model can work at a high level of aggregation and still be driver-based; what matters is that cause-and-effect logic is embedded, not that every individual line item is modelled. Too much granularity creates the same effort as classic bottom-up planning and undermines the efficiency gain.

In theory driver-based planning is widespread; in practice it is often only partially implemented. Real driver models require discipline in defining drivers and a technical platform that enables real-time calculation.

EBITDA (Earnings Before Interest, Taxes, Depreciation and Amortization) is profit before interest, taxes and depreciation on tangible and intangible assets. As a metric it isolates a company’s operating earning power from its financing structure, tax situation and investment policy.

EBITDA serves as an approximation of operating cash flow and is the most widely used benchmark metric for industry comparisons. The EBITDA margin (EBITDA / revenue) shows how much of revenue remains as operating earnings.

Limitations: EBITDA excludes depreciation and therefore overstates the earning power of capital-intensive companies. A machinery manufacturer with heavy investment and correspondingly high depreciation can report strong EBITDA while EBIT is significantly lower. For steering purposes EBITDA alone is not enough, because it does not show actual capital consumption.

Enterprise Performance Management (EPM) describes the full set of methods, processes and systems used to steer company performance. Core processes: budgeting, forecasting, consolidation, reporting, scenario analysis.

Important for the DACH context: EPM, Controlling and FP&A describe essentially the same function. The terms come from different traditions. EPM emphasises the IT and systems perspective (Gartner, software vendors). Controlling is the German technical term with an academic tradition (ICV, WHU, CMR). FP&A (Financial Planning & Analysis) is the Anglo-American variant, which has spread into German-speaking markets via US software vendors.

For companies, the naming makes no functional difference. What matters is whether the steering loop works: translate strategy into plans, measure actual against plan, update expectations, take decisions. The system landscape (EPM platform, ERP, BI) is a means to that end.

Financial modeling is the construction of quantitative models that represent a company’s financial relationships. It serves budget planning, forecasting, company valuation and scenario analysis. The basis is historical data plus assumptions about future developments.

Model architecture is decisive. Classic models represent individual line items and aggregate them bottom-up. Driver-based models define operational drivers (volume, prices, utilisation, headcount) and derive the financial results from them. The difference: driver-based models are scenario-capable. When a driver changes, the model immediately calculates the impact on P&L, balance sheet and cash flow.

In practice, model quality often falls short of the ambition. Too many line items, too little driver logic, excessive granularity. A good financial model answers management questions in minutes, not days.

A flexible budget adjusts planned costs and revenues to the activity volume actually achieved. Unlike a static budget, which rests on a fixed volume hypothesis, it scales with real output.

Its main benefit is separating volume variance from efficiency variance. If revenue comes in 10 % above plan and costs 8 % above, the static budget shows an unfavourable cost variance. The flexible budget scales planned costs up to the higher volume and reveals that efficiency was in fact better than planned.

Flexible budgets require costs to be split into fixed and variable components. In industrial companies with complex cost structures that calls for a planning model representing variable cost rates per driver.

Forecast accuracy measures how close a financial forecast comes to the later actual result. The most common metric is MAPE (mean absolute percentage error), which reports the average percentage deviation across several periods or line items.

Forecast accuracy is not an end in itself but a quality indicator for the entire planning process. Low accuracy can point to outdated assumptions, missing driver logic or updates that are too infrequent. High accuracy in stable periods says little if the model fails at market breaks.

Companies that measure their accuracy systematically start to see patterns: which business units forecast reliably and which do not? Where does the cause lie in the data, where in the methodology? This meta-analysis is often more valuable than the metric itself.

The forecast cycle describes the recurring sequence of data collection, model update, alignment and management review that leads to an updated financial forecast. It can be organised monthly, quarterly or on a rolling basis.

In practice, time spent on data collection and consolidation dominates. According to an FP&A survey, 45 % of planning departments spend most of their time gathering data, and 51 % need five or more working days for a single forecast run. That leaves little room for analysis and steering impulses.

The shorter the cycle, the higher the demand for automation. Companies moving from a quarterly to a monthly rhythm need integrated data flows from ERP and feeder systems, so the extra effort does not scale linearly with frequency.

The forecast horizon defines how far a financial forecast reaches into the future. Three tiers are common: short-term (up to 6 months, operational), mid-term (12 to 18 months, tactical) and long-term (3 to 5 years, strategic).

The horizon determines the methodology. Short-term forecasts rely on current operational data and granular drivers. Mid-term forecasts work with more aggregated assumptions and scenarios. Long-term forecasts overlap with strategic planning (mid-term planning, MTP) and draw on macroeconomic indicators and market models.

In German group planning, the operational budget typically covers one fiscal year, the MTP three to five years and strategic planning five to fifteen years. A rolling forecast can shift the horizon by continuously appending new months at the end instead of cutting off at fiscal year-end.

An FP&A platform is a specialised software solution that automates and scales financial planning, forecasting and analysis processes. It connects financial data with operational data from ERP, CRM and HR systems into a unified data model and enables driver-based modeling, scenario simulation and collaborative planning.

Positioning in the tool landscape:

• An FP&A platform is not an ERP system; it does not replace SAP but sits on top of it as a planning and simulation layer.
• It is not a BI tool; reporting on past data is not the focus. The emphasis is on planning, forecasting and simulating future developments.
• It is not an Excel replacement; it solves the problems that arise when planning processes hit the limits of spreadsheets (version conflicts, missing driver logic, no real-time data).

Modern FP&A platforms support the transformation of FP&A into a proactively advisory function: less time spent on data collection and validation, more capacity for analysis and decision support.

FP&A software is the umbrella term for applications that support financial planning, budgeting, forecasting and analysis. The spectrum ranges from specialised point solutions (consolidation only, reporting only) to integrated platforms covering the entire planning process.

Distinction from an FP&A platform: An FP&A platform is a specific architectural variant of FP&A software. It provides a unified data model, driver-based modeling and scenario simulation on one platform. Not every piece of FP&A software is a platform: a pure reporting tool or an Excel add-in counts as FP&A software but not as an FP&A platform.

Key evaluation criteria: integration capability with existing ERP systems, scalability for large data volumes, usability for business users (not just IT), and the ability to represent driver-based models natively.

The FP&A team (in German-speaking markets: the controlling department) is the organisational unit responsible for budgeting, forecasting, financial modeling and analysis. It forms the interface between the CFO and operational functions and translates financial data into a basis for decisions.

In large industrial groups the team is typically split into central controlling and decentralised business unit controllers. Central controlling owns group planning, consolidation and methodology. Decentralised controllers work directly with the business functions and act as business partners.

According to the WHU Controller Panel 2025 the balance of roles is shifting: the business partner role (average 5.7) is gaining importance, the scorekeeper role (average 4.2) is losing it. At the same time routine work continues to absorb much of the capacity. 45 % of FP&A team time goes into data collection and validation (FP&A Trends Survey 2023). Without reducing that share, the role ambition cannot be met.

FP&A transformation describes the shift of the financial planning function from a backward-looking reporting role to a proactive strategic advisor to management. In German-speaking markets the same process is discussed as “Wandel im Controlling”.

The transformation spans three dimensions. Technology: replacing manual, Excel-based processes with planning platforms offering real-time data and scenario capability. Processes: moving from rigid annual cycles to rolling, event-driven planning processes. Role: from number provider to business partner who actively shapes decisions.

The WHU Controller Panel 2025 shows the ambition is there but implementation lags. Business partnering is the most strongly demanded role, yet the capacity is missing as long as routine tasks absorb most of the working time. Transformation therefore often starts with automating data flows and standard reports, to create room for analysis and advice.

Headcount planning covers the analysis, forecasting and planning of personnel requirements by number, location, function and timing. It is a core component of workforce planning and at the same time a critical input to financial planning, since personnel costs make up 40-60 % of total costs in many companies.

Headcount planning is one of the most common entry points into driver-based planning. The driver logic: headcount x average salary x employer contributions = personnel costs. Added to this are variables such as attrition, hiring timelines, salary rounds and location factors. A good driver model calculates these relationships automatically and delivers the financial impact of every headcount change immediately.

In practice, headcount planning is often split between HR, controlling and the business functions, with Excel as the connective tissue. That produces version breaks and reconciliation effort. Integration on a shared platform is therefore often a first concrete step towards integrated planning.

Integrated Business Planning (IBP) is the end-to-end linkage of strategic planning, operational planning and financial planning in one consistent steering model. The core idea: decisions taken at one planning level automatically carry consequences through to the others.

In practice that means: when an investment decision is taken, balance sheet and cash flow effects have to become visible in the model automatically. When sales adjusts its demand plan, production capacity, personnel costs and the income statement follow.

IBP is the interplay of three dimensions:

• Vertical: strategic goals to mid-term planning to operational annual planning
• Horizontal: across functions (sales, operations, HR, finance)
• Financial: P&L, balance sheet and cash flow consistently linked

The most common weakness: companies call their planning “integrated” but link P&L and balance sheet only manually via Excel. Real integration requires driver logic that represents chains of effect, not just more data points.

Integrated planning is the consistent linkage of P&L, balance sheet and cash flow statement in one continuous model. In addition: the connection of operational planning (volumes, capacity, headcount) with financial planning.

Distinction from “more data”: Integrated planning means steering logic, not data volume. The linkage comes from driver models that represent chains of effect: an investment decision automatically pulls balance sheet and cash flow effects with it. More data points without that logic do not make planning more integrated, they make it more complex.

In practice many companies plan P&L and balance sheet separately or link them only manually via Excel. Rieg/Sindl/Tran (Controller Magazin 4/2025) argue that genuine integration has to connect three dimensions: operational planning, strategic planning and risk management. That is rarely achieved in practice.

Intercompany elimination is a sub-step of group consolidation: internal group transactions are removed from the consolidated financial statements so that the group does not book revenue with itself. Affected items include intercompany revenue, receivables and payables, dividends, loans and transfer prices.

The principle is simple: what entity A sells to entity B is not external group revenue. In practice the execution is laborious, because intercompany balances have to be reconciled. Differences arise from timing lags in bookings, differing exchange rates or valuation differences.

For planning the rule is: if the sub-plans of individual entities are simply added up without eliminating intercompany effects, the total plan is systematically overstated. A planning model that automates elimination avoids this error.

Key performance indicators (KPIs) are quantifiable metrics that make progress against strategic and operational goals measurable. They differ from general metrics through their direct link to company objectives: not everything that can be measured is a KPI. Something is only a KPI if it triggers a steering decision.

Financial KPIs measure economic performance: revenue growth, EBITDA margin, free cash flow, ROCE. Non-financial KPIs measure operational performance drivers: customer satisfaction, employee turnover, delivery reliability, time to market.

SMART criteria (specific, measurable, achievable, relevant, time-bound) are a widely used quality framework for KPI definition. In practice the most common problem is not the definition but the quantity: too many KPIs dilute focus. A management report with 50 metrics informs, but it does not steer.

Long-range planning (LRP) covers financial and strategic planning over a horizon of typically 5 to 15 years. It addresses questions of capital structure, investment scenarios, growth paths and portfolio decisions.

LRP is not the same as strategic planning, even though the two are often mentioned in the same breath. Strategic planning defines direction and priorities. LRP quantifies their financial consequences over a long time horizon. In the planning architecture of German industrial groups, mid-term planning (3-5 years) forms the bridge between the operational budget (1 year) and LRP.

The longer the horizon, the less useful detailed plans become. LRP therefore works at a higher level of aggregation and leans more on drivers and assumptions than on individual line items. The ability to vary key drivers in isolation becomes the decisive planning characteristic here.

Machine learning (ML) in FP&A refers to the use of algorithms that recognise patterns in historical data and derive forecasts, anomaly detection or automated data classification from them. Typical application areas: revenue forecasts based on historical sales data, detection of outliers in cost positions, automatic categorisation of accounting entries.

A sober assessment: ML in FP&A is currently more promise than practice. Forecast quality depends on the quality and quantity of historical data. In many companies that data is fragmented, inconsistent or too short-dated to support robust patterns. ML can relieve the controller of data preparation work, but it does not replace professional judgement on planning assumptions.

The most common fallacy: AI makes the controller redundant. The opposite is true. ML models deliver statistical forecasts, not steering decisions. Judging whether a forecast deviation is structural or random remains a human task. ML shifts the work from collecting data to interpreting it.

A management report is an internal reporting document that prepares the relevant financial and performance data for decision-makers on a regular basis (monthly or quarterly). Typical content: financial results (P&L, plus balance sheet and cash flow highlights where relevant), KPI development, plan-actual variances with explanation, forecast update and recommended actions.

A good management report answers three questions: where do we stand? Why are we off plan? What do we recommend? The most common weakness: too many numbers, too little interpretation. A report containing 30 pages of tables but no recommendation is a data package, not a steering instrument.

Producing management reports absorbs considerable capacity in many controlling departments. Automating data preparation (actuals import, variance calculation, standard commentary) is one of the fastest ways to free up controller capacity for analysis and business partnering.

Mid-term planning (MTP) covers a planning horizon of three to five years and forms the bridge between strategic planning and operational annual planning. It translates strategic goals into quantified financial targets: revenue growth, margin development, investment volumes, headcount growth.

In the planning architecture of DACH industrial groups, MTP is the central steering framework. It sets the corridor within which operational annual planning and the budget process take place. The group planning process typically starts in Q2 with strategic target setting, followed by the MTP round, before detailed operational planning begins in Q3.

The most common weakness: MTP and operational planning are methodologically decoupled. MTP works with aggregated top-down assumptions in PowerPoint or Excel, operational planning with granular bottom-up figures in the ERP environment. The continuity is missing. Driver-based models can close this gap by applying the same driver logic at different levels of aggregation: MTP varies market growth and market share, operational planning fills in the underlying volumes and prices.

Milestone-based budgeting ties the release of funds to predefined project goals rather than calendar periods. Budget only becomes available once an agreed milestone has been reached and documented.

The approach suits companies with a high share of project business, such as machinery and plant engineering or make-to-order manufacturing. There, costs and revenues are coupled to project progress, not to quarters. Calendar-based budgeting regularly produces misleading variances in such settings, because cash flows and delivery of work fall apart in time.

Implementation requires clear milestone definitions, robust progress measurement and a planning model that represents project logic. In groups with mixed business (project and series production) both budgeting forms often exist in parallel.

Model granularity describes the level of detail at which a financial model processes data, from group level down to the individual cost centre, from product segment down to SKU.

Choosing the right granularity is a steering decision, not a quality attribute. More detail does not automatically mean better planning. On the contrary: many companies build driver-based models that become so granular they create the same effort as classic bottom-up planning, without the hoped-for gain in speed.

A good driver model can work at a high level of aggregation and still be driver-based: it represents cause and effect, not every individual line item. The right question is not “How detailed can we model?” but “What level of detail creates steering value?”

Factors for the right granularity: which decisions should the model support? Which data is reliably and currently available? How often does the model need updating?

Model versioning is the systematic tracking of different versions of a financial model. Every change to structure, logic or parameters is documented, so earlier versions can be restored and compared at any time.

The benefit goes beyond mere traceability. When several controllers work on one model, versioning prevents changes being overwritten or errors going unnoticed. In audit situations it documents who made which adjustment and when.

Distinction from version control in planning: Model versioning concerns model structure and logic. Version control in planning concerns the plan data and plan states (for example budget v1, v2, v3).

The term describes the shift from rigid annual budgeting to a continuous, modelable planning approach. The core is not abolishing the budget but complementing it with rolling forecasts, scenario capability and driver-based logic.

Modern planning is not a technology topic but a change of method. Instead of rolling forward hundreds of individual line items, driver models represent causal relationships: how does a price change affect margin, volume and cash flow? Simulations make these relationships visible in minutes rather than weeks.

For DACH industrial groups with SAP-shaped processes this means concretely: the ERP data supplies the actuals basis, the planning model supplies the steering logic on top. Modern planning replaces neither ERP nor BI, it adds to both the ability to calculate decisions in advance.

Monthly forecasting is the practice of updating the financial forecast in every calendar month. Compared with a quarterly forecast, response speed increases because plan deviations become visible earlier and countermeasures take effect sooner.

Monthly forecasting does not replace the annual budget. It complements static annual planning with a continuously updated expectation. The two instruments have different functions: the budget sets targets and the resource framework, the monthly forecast reflects the current expectation.

The value added stands or falls with the effort involved. If every monthly cycle runs through the same manual process as a quarterly forecast, the higher frequency ties up capacity without generating proportionally more steering impact. Automated data integration and driver-based models reduce the marginal effort per cycle.

Multi-entity consolidation is the process of combining the individual financial statements of all group entities into consolidated group accounts. Intercompany transactions are eliminated, currencies translated and differing accounting standards harmonised.

In automotive OEMs and large industrial groups, consolidation is a complex process: dozens to hundreds of entities in different countries, with different currencies, local accounting rules and intercompany interdependencies. The challenge lies not in the methodology (that is standardised) but in data quality and speed.

Planning relevance: Consolidation affects not just actuals but the planning layer too. Anyone planning at group level has to eliminate intercompany effects in the plan as well. Otherwise inflated plan figures emerge that do not reflect genuine group performance. An integrated planning model consolidates automatically instead of bolting this step on manually afterwards.

Objectives & Key Results (OKRs) are a goal-setting framework that pairs ambitious qualitative goals (objectives) with measurable outcomes (key results). Developed at Intel in the 1970s, popularised by Google.

OKRs create transparency and strategic alignment: every unit and team defines goals that contribute to the company strategy. Key results make progress measurable. Typically: 3-5 objectives per unit, each with 2-4 key results.

Link to corporate steering: OKRs can act as a bridge between the strategic and operational levels (Rieg/Sindl/Tran, Controller Magazin 4-6/2025). Strategy is translated into objectives, key results supply measurable milestones, financial planning quantifies the resources. In practice this only works if OKRs and financial planning are connected rather than running in parallel.

An OLAP (online analytical processing) cube is a multidimensional data structure that enables fast analytical queries on large data volumes. Typical dimensions: time, product, region, cost centre, scenario. Users can rotate data along these dimensions (slice, dice, drill-down, roll-up).

OLAP cubes have been a standard building block of corporate steering since the 1990s. Many EPM platforms and data warehouses use OLAP technology at their core. The multidimensional structure mirrors the way controllers analyse data: “Show me revenue by region and quarter, then drill down to product group.”

Modern planning platforms partly replace the rigid cube architecture with more flexible in-memory calculation, but retain the multidimensional concept. For the user the principle stays the same: analyse data from different angles without being limited to predefined reports.

Operational planning translates strategic goals into concrete measures with a planning horizon of one year or less. It covers demand, production, headcount and cost planning at the level of individual business units or cost centres.

The most common fallacy: operational planning has to be detailed. Depth of detail alone does not create steering capability. What matters is whether the plan builds on drivers that can be adjusted when conditions change. A budget with 5,000 individual line items that has to be reworked manually when demand shifts is operationally less useful than a driver model with 30 levers.

In the planning architecture of large groups, operational planning is the bottom layer beneath mid-term planning (3-5 years) and strategic planning (5-15 years). The group planning process typically runs from Q2 to December, with the operational detail happening in the second half of the year.

Periodic planning is the creation of budgets and forecasts at fixed intervals: annually, half-yearly or quarterly. The planning period is fixed, not rolling.

The central difference from a rolling forecast: under periodic planning the remaining planning horizon shrinks with every month. An annual plan created in January has only three months of remaining reach by October. Rolling approaches keep the horizon constant by appending a new period at every update.

Periodic planning is not inherently outdated. For regulatory obligations (annual accounts, budget approval by supervisory bodies) it remains necessary. It becomes problematic when it is the only steering instrument and forecast updates only become possible in the next cycle. According to surveys, 51 % of companies need more than five days for a single forecast run.

The planning cycle describes the recurring sequence of target setting, budgeting, monitoring actuals and revising the forecast. In DACH industrial groups this cycle typically runs from Q2 to December and breaks into phases: strategic target setting, mid-term planning, detailed operational planning, alignment via the counterflow method and plan approval.

Cycle length is a direct indicator of planning maturity. The longer a run takes, the more likely the result is already outdated by the time it is approved. Companies looking to shorten their cycle attack two points: consolidation (automated rather than manual) and planning logic (driver-based rather than line-item-based).

A widespread misconception: the planning cycle can be shortened through better tools alone. In fact, coordination processes, approval stages and data availability matter at least as much. Technology accelerates the calculation work, not the organisation.

A planning model is the structured framework of assumptions, drivers, calculation logic and scenarios on which corporate planning is built. It determines which variables are controllable, how they relate to one another and which outcome measures they influence.

The decisive difference lies between line-item-based and driver-based models. Line-item-based models roll historical values forward (prior year plus X percent). Driver-based models represent causal relationships: volume times price gives revenue, revenue less variable costs gives contribution margin. When a driver changes, the entire model adjusts consistently.

Driver-based does not automatically mean detailed. A good planning model concentrates on the 20-30 levers that explain 80 % of the variation in results. More drivers mean more maintenance effort, not better steering. The art lies in the right granularity for the given planning level.

Planning workflows comprise the defined sequences, roles, responsibilities and approval stages that structure a planning round. They govern who supplies which data and when, who consolidates, who approves and how variances are escalated.

In groups with several business units and regions, planning workflows quickly become complex. A typical budget process runs through three to five iterations between headquarters and units before final plan approval. Process efficiency is measured not by the number of steps but by the time between data input and decision-ready output.

Digitalised workflows replace email-based coordination with defined status paths and automatic notifications. The bigger lever, however, often lies in reducing the iteration loops themselves, for example through driver-based models that make gaps between top-down targets and bottom-up inputs visible immediately.

Predictive forecasting uses statistical methods and machine learning to derive future developments from historical data. Typical methods are time series analysis, regression models and neural networks.

Algorithmic forecasts complement driver-based planning but do not replace it. Their strength lies in recognising recurring patterns and in reducing subjective bias (anchoring, overconfidence). They fail where structural breaks occur that do not appear in the training data: new business models, regulatory intervention, supply chain disruption.

In practice, predictive forecasting proves its worth as a validation tool. The algorithmic forecast provides a reference value against which management judgement is tested. Where the human assessment diverges significantly, a justification is required. Forecast quality improves without handing planning authority to a model.

Procurement controlling steers and monitors a company’s sourcing performance. Core questions: how are purchase prices developing against budget? Which suppliers deliver reliably, which cause extra costs? How do material price changes affect product costs and therefore margin?

In industrial groups with a high material share (automotive, machinery), procurement influences 40-60 % of total costs. Procurement controlling gives the purchasing function transparency over savings, price trends and contract compliance, and gives financial controlling the input for cost planning.

Link to simulation: The most interesting steering question in procurement is not retrospective (What did we save?) but forward-looking (What happens to our margins if raw material prices rise by 15 %?). Product cost simulation and should-cost analyses connect procurement data with the financial model: material price changes feed through cost of goods sold into EBIT and cash flow. Driver-based models represent this chain of effects and make it calculable in minutes.

A projected balance sheet is the balance sheet planned for future periods. It is the third statement alongside the planned income statement and planned cash flow, and it shows how assets, tied-up capital and debt develop over the planning horizon.

Why it is often missing: Many companies plan the income statement only. Without a projected balance sheet, exactly the questions asked from outside stay open: how much capital is tied up? How does leverage develop? Will the balance sheet ratios in loan agreements be met? An earnings plan alone cannot answer any of them.

The derivation matters, not the estimate. A robust projected balance sheet does not come from rolling balance sheet items forward, but from linking them to operational planning: revenue and payment terms give receivables, material costs and inventory coverage give stock, capital expenditure and the depreciation schedule give fixed assets. Once those bridges are in place, the projected balance sheet moves automatically whenever an operational assumption changes.

Quarterly planning breaks operational planning into three-month segments with their own milestones and targets. It is often coupled to OKR frameworks or comparable steering logics.

Quarterly planning is not a replacement for annual planning but its concretisation. The annual budget provides the frame, quarterly planning fills it with current assumptions and operational priorities. In volatile markets, quarterly review gains importance because annual plans can already diverge significantly from reality after three months.

For controlling departments, quarterly planning means additional effort that is only justified if the results genuinely feed into steering decisions. Without driver-based models that allow fast recalculation, the quarterly update becomes an end in itself.

Real-time collaboration means several users can work on plans, models or forecasts simultaneously, with changes becoming visible to everyone straight away. In practice: the controller at headquarters sees immediately when the business unit controller at another site adjusts their demand plan.

In corporate steering this solves a concrete problem. Classic planning processes run sequentially: one unit plans, passes the file on, the next unit adds to it. Every handover creates waiting time and version conflicts. Working in parallel on a shared data foundation shortens planning cycles and eliminates the question “Which version is the current one?”

The precondition is a central data model with clear access rights. Real-time collaboration without a role concept produces chaos rather than efficiency.

Real-time reporting describes the provision of financial and operational data in reports without manual preparation or time lag. Data flows directly from ERP, data warehouse and operational systems into the reports.

In practice, “real time” rarely means to the second. The more relevant question is: how current is the data at the moment management has to make a decision? If the monthly report only arrives two weeks after month-end, steering capability is missing. If actuals are reflected in the forecast on a daily basis, response speed increases.

Distinction from BI dashboards: Real-time reporting in an EPM context means more than visualising historical data. It connects current actuals with plan and forecast data, so that variances become visible immediately and can be translated into corrective action.

Resource planning covers the systematic planning and allocation of all operational resources: people, budget, technology, fixed assets and capacity. The goal is to identify bottlenecks early and avoid excess capacity.

Distinction from ERP: Enterprise resource planning (ERP) is a software system that processes transactions (orders, postings, goods movements). Resource planning in a steering context is a planning process that asks: do we have the right resources, in the right quantity, in the right place, to reach our strategic goals?

In practice, resource planning often fails on missing linkage. Headcount is planned in isolation from investments, capacity in isolation from demand. Driver-based models can create that linkage: when the demand plan rises, capacity requirements follow automatically, and personnel requirements follow from those.

Revenue planning covers the forecasting, modeling and steering of revenue development. It typically segments revenue by product, region, customer group and sales channel, forming the top-line basis for the entire financial plan.

In practice, revenue planning consists of two components. The market-driven perspective derives revenue from external factors: market size, market share, price development, economic conditions. The sales-driven perspective builds revenue bottom-up from the pipeline: opportunities, win probabilities, sales quotas.

Good revenue planning connects both perspectives and makes the gaps transparent. If the sales pipeline sits 15 % below the market-based target, that is an early warning indicator, rather than waiting for the actual variance at quarter-end.

Rolling forecasts are a continuous planning approach in which the forecast horizon is pushed forward by one further period at regular intervals, typically monthly or quarterly. The company therefore always looks the same distance into the future, for example always five quarters ahead.

An important distinction: a rolling forecast is not the same as a current year forecast. The current year forecast only looks to 31 December of the current fiscal year and is updated during the year. The rolling forecast detaches from the calendar year and rolls the horizon forward independently of fiscal year-end.

This distinction is not academic: planning only to year-end means losing sight of the following year from October onwards. Rolling forecasts keep the strategic forward view constant and encourage future-oriented steering.

In practice the rolling forecast is best practice but rarely fully implemented. Breaking away from calendar-year thinking often fails on processes and systems designed around fixed fiscal years.

A rolling horizon keeps the planning horizon constant by appending a new period at every update. When a month or quarter ends, the horizon moves forward by the same unit. The forward view therefore always stays the same length, typically 12 to 18 months.

The difference from periodic planning is structural, not just temporal. Under periodic planning the remaining horizon shrinks with every month. By October of an annual plan, only three months of planning reach remain. A rolling horizon avoids this problem but creates a higher update burden, because new periods have to be planned regularly.

That extra effort is only sustainable if planning is based on driver models. Anyone who has to replan hundreds of individual line items every month will soon abandon rolling forecasts. Driver-based models reduce the update to adjusting a few levers, from which the detailed values are derived. The fact that 51 % of companies need more than five days per forecast run shows the scale of the need.

Sales planning is the strategic and operational process of defining sales targets and the steps needed to reach them. It includes revenue quotas per sales rep or region, territory allocation, resource allocation and action planning.

Link to revenue planning and financial planning: Sales planning supplies the operational input for revenue planning. What sales plans as pipeline and quotas becomes the basis for the financial revenue forecast. Conversely, revenue planning sets the frame: if the company targets 10 % growth, sales planning has to translate that goal into concrete quotas and measures.

In many companies, sales planning is the bridge between strategy and operational execution. The most common weakness: sales planning happens in the CRM, financial planning in Excel or the EPM platform. The data is reconciled manually. Integrating both systems eliminates that break.

A scenario in a corporate context describes a possible future state based on a consistent set of assumptions and drivers. Typical scenarios: base case (expected plan), best case (upside potential), worst case (downside risk).

Two distinct uses matter here:

Scenario simulation (operational/financial): calculation-based variation of drivers in the financial model. “What happens to our EBIT if volume comes in 10 % below plan?” Result: concrete financial figures, ideally available in minutes.

Scenario planning (strategic): development of 2-4 consistent pictures of the future across several years, narrative in character, qualitative. “How does our business develop if the Chinese market falls away?” Result: strategic options for action.

A scenario differs from a sensitivity analysis: the sensitivity analysis varies a single variable in isolation. A scenario changes several related variables at once to produce a realistic overall picture.

Scenario analysis is the systematic examination of different future scenarios and their financial impact. It combines the construction of scenarios (which sets of assumptions are plausible?) with quantitative calculation (what do they mean for P&L, balance sheet and cash flow?).

Distinction from related methods:

• Sensitivity analysis varies a single variable in isolation. Scenario analysis changes several related variables at once.
• What-if analysis is often selective and ad hoc. Scenario analysis follows a structured methodology with defined scenarios (typically base case, best case, worst case).
• Scenario planning is strategic and narrative. Scenario analysis is the quantitative part: it calculates the scenarios financially.

In practice, scenario analysis often fails on speed. Only 16 % of companies can calculate scenarios in under a day, and 20 % cannot calculate scenarios at all (FP&A Trends Survey 2023). Fast scenario analysis requires a driver-based planning model in which changes to assumptions feed through to all outcome measures immediately.

Scenario planning is a strategic planning method in which organisations develop several plausible pictures of the future in order to test their strategies against different developments. Unlike a classic financial forecast, which predicts a single expected outcome, scenario planning deliberately explores different possible futures.

Every scenario rests on a coherent narrative: which external factors (market shifts, regulatory intervention, competitive moves) come together, and what strategic implications follow?

Distinction from scenario simulation: Scenario planning is qualitative and strategic; it asks “Which future is plausible?” Scenario simulation is quantitative and operational; it calculates “What does this future mean financially?” The two complement each other: first formulate the strategic scenario, then simulate the financial impact in the model.

Market data: only 16 % of companies can calculate scenarios in less than a day. 20 % cannot calculate scenarios at all.

Self-service analytics is the ability of non-technical users to query data, build reports and run analyses on their own, without depending on IT or data teams. In a planning context: the business unit head can calculate a scenario or analyse a variance themselves.

The goal is decentralising analysis. If every ad hoc question has to detour through controlling, a bottleneck forms. If business functions can access the data themselves, controlling is relieved of data requests and can concentrate on interpretation and recommendation.

The precondition: a data model business users understand without knowing SQL. And an authorisation concept that ensures everyone only sees the data they are allowed to see. Self-service without governance leads to contradictory figures across departments.

Sensitivity analysis isolates a single variable and varies it systematically while all other parameters stay constant. The goal is to quantify the earnings impact of individual drivers: how much does EBIT change if the raw material price rises by 10 %?

Sensitivity analysis is not the same as scenario planning. Scenarios change several variables at once and represent alternative pictures of the future. Sensitivity analyses answer a narrower question: which driver has the greatest influence on the result?

In practice, sensitivity analysis is a precursor to scenario planning. It identifies the drivers that should be varied in scenarios. The precondition is a planning model with explicit driver relationships. In line-item-based models that merely roll totals forward, individual effects cannot be cleanly isolated.

Short-term forecasting covers a period of one week to twelve months and serves operational steering. It draws on current sales, order and production data as well as operational drivers such as utilisation, inventory coverage or order intake.

Short-term forecasts differ methodologically from mid-term and long-term forecasts. The data basis is more granular, the update frequency higher and the uncertainty lower, provided no abrupt market dislocations occur. In exchange, the steering lever is limited to tactical measures: adjusting the production programme, correcting order quantities, shifting personnel capacity.

In a group context, short-term forecasting supplies the input for liquidity planning and working capital management. It is often the first instrument to be professionalised when a company moves to monthly or rolling forecasting.

Simulation in a corporate steering context means calculating the effect of driver changes through the entire financial model in real time. One driver is varied - a drop in volume, a price change, a capacity adjustment - and the model immediately calculates the impact on P&L, balance sheet and cash flow.

Distinction from scenario planning: Simulation is operational and calculation-based; it answers “What happens financially if driver X changes?” Scenario planning is strategic and narrative; it asks “Which future is plausible?” The two complement each other but address different decision situations.

Distinction from sensitivity analysis: A sensitivity analysis typically varies a single variable in isolation. Simulation can change several drivers at once and calculate their interactions.

The decisive factor is speed. If management wants to see a scenario at short notice and the team needs days to prepare it, simulation capability is missing. Only 16 % of companies can calculate scenarios in less than a day.

Single source of truth (SSoT) is a data management strategy in which all company data is held in one central place and consumed from there. Every metric has exactly one authoritative source, and every report draws on the same data foundation.

The problem SSoT solves: in large organisations the same figures exist in different versions. Sales has one revenue number, controlling another, the board sees a third. The differences arise from different data sources, calculation logic or cut-off dates. Every management meeting starts with the question “Which number is right?” instead of “What are we going to do?”

SSoT is not a technology project but a governance decision. The technology (a central planning platform, a data warehouse) is a precondition but not sufficient. What matters is that all areas accept a shared data foundation as binding.

Strategic planning defines a company’s long-term priorities, resource allocation and direction over a horizon of typically 5 to 15 years. It forms the top layer of the planning architecture.

Strategy without quantification remains a statement of intent. The connection to financial planning is made through mid-term planning (3-5 years), which translates strategic goals into measurable financial targets. Operational annual planning is derived from there. This continuity from strategy through to the operational budget is an unsolved problem in many groups, because strategic and operational planning happen in separate systems and processes.

Driver-based models create a shared logic here: the same growth drivers stored in strategic planning feed into mid-term and annual planning. When a strategic assumption changes, the effect becomes visible across all planning levels.

The tech stack describes the totality of software and technologies a company uses for its business processes. In a planning context: ERP, CRM, HR systems, planning platform, BI tools, data warehouse and the interfaces between them.

In the DACH industrial context the stack is typically SAP-heavy. SAP ECC or S/4HANA as the ERP core, SAP BW or BW/4HANA as the data warehouse, SAP SuccessFactors for HR. The planning platform has to fit seamlessly into this landscape, understand SAP data structures and communicate bidirectionally.

The decisive question for any stack is not the number of tools but the quality of integration. A poorly integrated stack with ten tools creates more manual work than a well integrated one with five. Data integration is the bottleneck, not the functionality of individual systems.

Top-down planning means senior management sets overarching targets, which are then broken down to business units, regions and cost centres. The advantage lies in speed and strategic consistency.

The well-known drawback: limited acceptance at the operational level. Targets set without feedback from the units are frequently seen as unrealistic. Most groups therefore work with a counterflow method, which brings top-down targets and bottom-up detailed plans together over several iterations.

The practical challenge lies in the speed of those iterations. If every reconciliation between target and operational proposal takes days, the entire planning process stretches over months. Driver-based models accelerate the counterflow, because shifts in targets become visible in the overall result immediately rather than having to be consolidated manually first.

The Summen- und Saldenliste (SuSa) is the German trial balance: a list of every general ledger account for a period with its debit and credit totals and the resulting balance. It is the complete account-level extract from financial accounting, and every condensed report is derived from it.

Distinction from the BWA: The BWA is a condensed presentation with a fixed structure. The trial balance is the raw layer underneath it, more granular and without any interpretive grouping. When a BWA line cannot be explained, the trial balance holds the accounts it is made of.

Planning relevance: That account-level detail is what makes the trial balance the usual starting point for building an integrated planning model. Income statement and balance sheet structure can be derived from it programmatically and then backed with drivers, whereas a condensed BWA does not carry enough detail. In DATEV environments the trial balance export is therefore regularly the first interface between accounting and planning.

A value driver tree (also: driver tree) is the hierarchical representation of the operational and financial drivers that influence a target measure. At the top sits a steering metric (for example EBIT, ROCE, free cash flow), which is broken down across several levels into its components until operational levers become visible.

Example: EBIT to revenue minus costs, revenue = volume x price, volume = market size x market share. The tree makes visible which operational levers act on which financial metric, and how strongly.

The driver tree is the foundation of driver-based planning. It defines which variables in the planning model are controllable and how they relate. Without a cleanly defined driver tree, driver-based planning remains a label without substance: the model then contains drivers, but not the causal logic connecting them.

A common mistake: too many levels, too many drivers. A driver tree with 200 leaves is not a steering instrument but a database. The art lies in reducing it to the 20-30 drivers that explain 80 % of the variation in results.

A variance is the quantitative difference between a planned value and the actual result. It is the founding concept of management accounting: all steering starts with recognising where reality departs from the plan.

Variances are typically expressed in absolute terms (EUR) and relative terms (%). A variance of EUR 100,000 can be critical against a budget of EUR 1 million and negligible against EUR 100 million. That is why variances need context, not just calculation.

Favourable vs. unfavourable variance: In controlling parlance, a favourable variance (actual better than plan) is not automatically good news. If revenue comes in 20 % above plan but capacity cannot keep up, delivery problems follow. Variances in either direction call for analysis and, where needed, corrective action.

Variance analysis is the systematic comparison of plan, budget or forecast values against actual results. The goal is not merely to establish the difference but to understand its causes: does the variance come from price, volume, mix or efficiency?

A typical decomposition of a revenue variance: price variance (actual vs. planned price at constant volume), volume variance (actual vs. planned volume at constant price), mix variance (shifts between products or regions with different margins).

Variance analysis is a cornerstone of enterprise performance management. It connects the past (What happened?) with the future (Does the forecast need to change?). In practice it loses its value when it arrives too late (the month-end close takes too long) or becomes too granular (explaining every single cost centre without drawing out the overall message).

Version control in planning is the management of multiple iterations of a plan or forecast. Typically the budget runs through several rounds (draft, management approval, final version). Each version is stored as a plan state in its own right, comparable and traceable.

The audit trail documents which changes happened between versions and why. That is not only a compliance requirement but a steering instrument: when management asks why the forecast moved 8 % from Q2 to Q3, the answer has to be available in minutes.

Version control is also the basis for scenario planning. Comparing different scenarios requires cleanly separated plan versions with clear naming and a consistent data foundation.

What-if analyses deliberately change one or more variables in a planning model to observe the impact on outcome measures. Typical questions: what happens to cash flow if sales volume drops by 15 %? What if raw material costs rise at the same time?

Distinction from scenario planning: What-if analyses are selective investigations, often carried out ad hoc. Scenario planning is broader and represents internally consistent pictures of the future built from several linked assumptions. Distinction from sensitivity analysis: that varies only one variable in isolation, whereas what-if analyses can change several at once.

The explanatory power stands or falls with the quality of the underlying model. Only when driver relationships are correctly represented do what-if analyses deliver robust results. In pure roll-forward models they stay superficial, because interactions between variables are not captured.

Workforce planning goes beyond pure headcount planning. It aligns the entire human capital with strategic company goals: which capabilities will we need in three to five years? Where do gaps open up through attrition, retirement or technological change? How do we develop succession for key positions?

Distinction from headcount planning: Headcount planning answers “How many employees do we need, where and when?” Workforce planning additionally asks “With which capabilities, in what organisational structure and with what development path?”

In industrial groups, workforce planning is gaining relevance because several trends are acting at once: skills shortages, demographic change, automation and the shift of capabilities (for example from mechanical to software skills in the automotive sector). These trends require planning beyond the next budget cycle.

Working capital is the difference between current assets (receivables, inventories, cash) and current liabilities (trade payables, short-term provisions). It measures how much capital is tied up in day-to-day operations.

Key steering metrics: DSO (days sales outstanding, the average collection period), DPO (days payable outstanding, the average payment period to suppliers), DIO (days inventory outstanding, inventory coverage). These three combine into the cash conversion cycle: the time from purchasing material to receiving payment.

Planning relevance: Working capital ties up liquidity. In industrial groups with long production cycles and high inventories, working capital can absorb several hundred million euros. Every reduction frees up cash flow without requiring revenue growth. In integrated planning, working capital therefore has to be modelled as a planning area in its own right, not as a residual of balance sheet planning.

A worst-case scenario models the least favourable but still realistic conditions for a company. It serves as a stress test for financial resilience and as a basis for contingency plans.

Realistic is the key word. A worst case that sets every negative factor to an extreme value simultaneously loses its explanatory power. More useful is a combination of probable negative developments: a drop in demand in one core market alongside rising input costs, rather than the simultaneous collapse of all markets.

In planning practice, worst-case scenarios are often presented alongside best case and base case as a corridor. Building them is only worthwhile if the planning model represents driver relationships, so that changed volume automatically feeds through to revenue, variable costs and result. Manually assembled worst cases are error-prone and go out of date quickly.

Extended Planning & Analysis (xP&A) extends the classic FP&A function beyond the finance department. Instead of isolated financial planning, operational planning processes from sales, HR, supply chain and other functions are brought together on a shared platform with a shared data model.

The difference from traditional FP&A: under xP&A, changes in one sub-plan propagate automatically to all linked plans. When HR adjusts headcount planning, personnel costs, capacity and revenue forecasts follow automatically, provided the driver logic is cleanly set up.

According to Gartner (Market Guide for Cloud xP&A Solutions, 2024), more than 60 % of all planning initiatives will carry xP&A requirements. In practice, implementation lags the ambition: data governance and change management are the most common hurdles.

Zero-based budgeting (ZBB) requires every expense item to be justified from scratch in every planning period. The prior-year budget is not the starting point. Every cost centre has to demonstrate why funds are needed and at what level.

ZBB targets cost transparency and the dissolution of historically grown budget cushions. In groups with hundreds of cost centres, positions accumulate over the years that are never questioned because they count as “given”. ZBB breaks this pattern by reversing the burden of proof: it is not the cut that has to be justified, but the requirement.

The effort is considerable. A complete ZBB round for an industrial group ties up months of management time. Many companies therefore apply ZBB selectively, for overhead areas or on a multi-year rhythm, while operational areas continue to plan incrementally. Combining both approaches is more common than the pure doctrine.