Financial forecasting is one of the most important disciplines in accounting, FP&A, and business management. A well-designed forecast helps organizations estimate future revenue, expenses, profitability, cash flow, staffing requirements, capital needs, and other key performance indicators.

But forecasting is not simply about taking last year’s numbers and adding a percentage.

A professional forecast combines historical performance, operational drivers, management assumptions, current business conditions, and analytical methods to develop a reasonable estimate of what may happen in the future.

A projection goes one step further by allowing management to evaluate alternative assumptions and “what-if” scenarios.

The objective is not to predict the future with perfect accuracy. That is impossible. The objective is to create a reasonable, transparent, and measurable view of the future that improves decision-making.

This guide explains how to calculate a forecast or projection, the most common forecasting methods, practical formulas, examples, and best practices for building a reliable forecasting process.

What Is Financial Forecasting?

Financial forecasting is the process of estimating an organization’s future financial and operational performance using historical information, current trends, known business conditions, and assumptions about the future.

A forecast may estimate:

Revenue
Volume
Expenses
EBITDA
Operating income
Net income
Accounts receivable
Working capital
Cash flow
Capital expenditures
Staffing requirements
Inventory
Debt requirements

For example, a company may forecast monthly revenue for the next 12 months based on historical sales, current customer activity, pricing changes, and expected market conditions.

Forecasting is particularly important in FP&A because it creates a bridge between actual performance and management’s expectations for the future.

Forecast vs. Projection: What Is the Difference?

The terms “forecast” and “projection” are sometimes used interchangeably, but they can serve different management purposes.

Forecast

A forecast represents management’s best current estimate of future performance based on the information available at the time.

For example:

“Based on current volume, pricing, contracts, and operating trends, we expect full-year revenue to be approximately $50 million.”

A forecast should change when the underlying information changes.

Projection

A projection generally evaluates a future outcome under a defined set of assumptions.

For example:

“If revenue increases by 8% and operating expenses increase by 4%, what would EBITDA be?”

Projections are therefore particularly useful for:

Business planning
Strategic planning
Investment analysis
New product analysis
Capital planning
Scenario analysis
Financing requirements

A useful way to think about the distinction is:

Forecast = What we currently expect to happen

Projection = What could happen under specified assumptions

The original article appropriately emphasizes that projections can incorporate hypothetical assumptions and “what-if” scenarios.

Why Forecasting Matters

A strong forecasting process supports four major areas of management.

1. Decision-Making

Management can use forecasts to determine whether it should:

Hire additional employees
Increase inventory
Reduce expenses
Increase production
Expand capacity
Adjust pricing
Delay capital expenditures
Seek additional financing

2. Financial Planning

Forecasts help estimate future:

Revenue
Expenses
Profitability
Cash requirements
Working capital
Capital expenditures

3. Risk Management

Forecasting allows management to identify potential problems before they occur.

For example:

A company may forecast strong revenue growth but discover that accounts receivable will increase faster than cash collections.

The revenue forecast may therefore look positive while the cash-flow forecast identifies a potential liquidity problem.

4. Performance Management

Forecasts create a benchmark against which actual performance can be measured.

Management can calculate:

Actual − Forecast = Forecast Variance

The variance is then investigated to determine whether the difference resulted from:

Volume
Price
Mix
Timing
Productivity
Market conditions
Assumptions that were incorrect

The original source similarly emphasizes comparing actual results with forecasts and using variance analysis to understand deviations.

The Basic Forecasting Process

A professional forecasting process can be organized into seven steps:

1. Define the objective

2. Collect and validate historical data

3. Identify the business drivers

4. Select the forecasting methodology

5. Develop assumptions

6. Calculate the forecast

7. Compare actual results and update the forecast

This process is more reliable than simply applying a growth percentage to historical financial statements.

Step 1: Define What You Are Forecasting

The first question should be:

What exactly are we trying to forecast?

Examples include:

Monthly revenue
Annual EBITDA
Patient volume
Product sales
Labor expense
Cash balance
Accounts receivable
Capital expenditures
Inventory requirements

The forecast period must also be defined.

For example:

Next month
Next quarter
Remaining fiscal year
Next 12 months
Three-year strategic plan

The forecast methodology should match the objective and time horizon.

Step 2: Collect and Validate Historical Data

Historical data is usually the starting point for forecasting.

Relevant information may include:

Financial Data

Revenue
Expenses
Gross margin
EBITDA
Accounts receivable
Accounts payable
Cash flow

Operational Data

Units sold
Customers
Patient volume
Production volume
Labor hours
Capacity utilization
Average selling price

External Data

Inflation
Interest rates
Economic growth
Industry trends
Competitor activity
Regulatory changes

The original article identifies sales, market trends, and operational information as important forecasting inputs.

However, historical data should not automatically be assumed to be representative of the future.

Before using it, investigate:

One-time revenue
One-time expenses
Acquisitions
Divestitures
Major contract changes
Accounting changes
Unusual market conditions
Significant operational disruptions

This process is often called normalization.

Step 3: Identify the Drivers of the Forecast

One of the biggest differences between a basic forecast and a professional FP&A forecast is the use of drivers.

Instead of forecasting revenue simply as:

Prior-Year Revenue × 105%

a driver-based model asks:

What actually generates revenue?

For example:

Revenue = Units Sold × Average Selling Price

If a business expects:

100,000 units
Average selling price of $50

Then:

Forecast Revenue = 100,000 × $50 = $5,000,000

This approach is more transparent because management can separately change the assumptions for volume and price.

Revenue Forecasting Example

Suppose a company generated $4.5 million of revenue last year.

Management expects:

Volume to increase by 8%
Average price to increase by 3%

Instead of simply forecasting revenue at 11% growth, calculate the drivers separately.

Assume last year’s revenue represents:

100,000 units × $45 = $4,500,000

Forecast volume:

100,000 × 1.08 = 108,000 units

Forecast price:

$45 × 1.03 = $46.35

Therefore:

108,000 × $46.35 = $5,005,800

Forecast revenue is approximately

$5.006 million

This produces an expected increase of approximately 11.24%, rather than simply adding 8% + 3%.

This is a good example of why driver-based forecasting is superior to applying arbitrary growth rates.

Step 4: Choose a Forecasting Method

There is no single forecasting method that works for every business.

The appropriate methodology depends on:

Data availability
Forecast horizon
Business volatility
Seasonality
Complexity
Management requirements

Common methods include the following.

Method 1: Percentage-Growth Forecast

The simplest method is to apply an expected growth rate to historical results.

Formula

Forecast = Prior Period × (1 + Growth Rate)

Example:

Prior-year revenue = $10 million

Expected growth = 6%

$10,000,000 × 1.06 = $10,600,000

This method is simple and useful for preliminary planning, but it can be weak when the business has significant changes in volume, pricing, mix, or operating structure.

Method 2: Moving Average

A moving average uses recent historical periods to calculate an expected future value.

Formula

Forecast = Sum of Selected Historical Periods ÷ Number of Periods

Suppose monthly sales were:

January: $90,000
February: $100,000
March: $110,000

A three-month moving average would be:

($90,000 + $100,000 + $110,000) ÷ 3 = $100,000

The forecast for April would therefore be $100,000.

Moving averages are useful when management wants to reduce the impact of short-term fluctuations.

The original source identifies moving averages as a time-series method used to smooth historical data and reduce random fluctuations.

Method 3: Exponential Smoothing

Exponential smoothing gives greater weight to recent observations.

This can be useful when recent performance is more representative of future conditions than older historical data.

Conceptually:

New Forecast = α(Actual Result) + (1 − α)(Previous Forecast)

where α represents the smoothing factor.

A higher α places greater emphasis on recent actual performance.

Exponential smoothing is particularly useful for recurring monthly forecasts where recent changes in demand are important.

Method 4: Regression Analysis

Regression analysis estimates the relationship between a dependent variable and one or more independent variables.

For example:

Revenue = a + b(Advertising Expense)

A more sophisticated model might include:

Revenue = a + b₁(Volume) + b₂(Price) + b₃(Marketing Spend) + b₄(Seasonality)

Regression can be useful when historical data demonstrates measurable relationships between business drivers and financial outcomes.

The original article identifies regression as a method for analyzing relationships between dependent and independent variables.

Method 5: Driver-Based Forecasting

For FP&A purposes, driver-based forecasting is often one of the most practical approaches.

Instead of forecasting every account independently, financial results are linked to operational drivers.

Examples include:

Revenue

Volume × Price

Labor Expense

Employees × Hours × Hourly Rate

Sales Commission

Revenue × Commission Rate

Interest Expense

Average Debt × Interest Rate

Inventory

Units Required × Cost per Unit

Accounts Receivable

Revenue × Collection Assumption

This creates a model where changes in operational assumptions automatically flow through the financial statements.

Step 5: Develop Explicit Assumptions

Forecasts are only as strong as their assumptions.

A professional forecast should document assumptions for:

Volume
Pricing
Market share
Wage increases
Inflation
Headcount
Productivity
Supplier costs
Interest rates
Capital expenditures
Collection rates
Customer behavior

The original article emphasizes documenting assumptions to provide transparency and help stakeholders understand the basis of the projection.

A good assumption should be:

Specific + measurable + supported + dated

Instead of:

“Revenue will increase significantly.”

Use:

“Revenue is forecast to increase 7% because unit volume is expected to increase 5% and average selling price is expected to increase 2% beginning in October.”

The second assumption can be tested and measured.

Step 6: Build the Forecast

Once the drivers and assumptions are established, calculate the forecast.

A basic financial model may contain:

Forecast Component	Calculation
Revenue	Volume × Price
COGS	Revenue × COGS %
Gross Profit	Revenue − COGS
Labor	FTE × Average Compensation
Other Expenses	Driver-based assumptions
EBITDA	Gross Profit − Operating Expenses
Depreciation	Fixed Assets × Depreciation Assumptions
Interest	Average Debt × Interest Rate
Net Income	EBITDA − D&A − Interest − Taxes

This approach creates an integrated forecast rather than a collection of disconnected estimates.

Step 7: Develop Multiple Scenarios

A single forecast can create a false sense of certainty.

Management should consider at least three scenarios:

Base Case

Management’s most reasonable expectation.

Upside Case

Assumes stronger-than-expected performance.

Downside Case

Assumes adverse operating or market conditions.

For example:

Scenario	Revenue Growth	EBITDA Margin
Downside	2%	12%
Base	6%	15%
Upside	10%	18%

Scenario analysis allows management to understand the potential range of outcomes rather than relying on a single number.

Example: Forecasting a Retail Business

Consider a fictional retailer called Trendy Styles.

Last year’s holiday sales were:

November: $30,000

December: $40,000

Management expects:

10% increase in customer traffic
Continued growth in online sales
New product introductions

A simplified projection would be:

November

$30,000 × 1.10 = $33,000

December

$40,000 × 1.10 = $44,000

Total

$33,000 + $44,000 = $77,000

The original example uses this same basic calculation.

However, a stronger forecasting model would go beyond the 10% assumption.

Management could separately forecast:

Store traffic
Conversion rate
Average transaction value
Online orders
Average online order value
Promotional impact

For example:

Store Revenue = Traffic × Conversion Rate × Average Transaction Value

This creates a much more useful operational model.

Forecasting the Income Statement

Once revenue is forecast, expenses can be modeled using appropriate drivers.

Suppose projected revenue is $5 million.

Management expects:

COGS = 60% of Revenue

Therefore:

$5,000,000 × 60% = $3,000,000

Gross profit:

$5,000,000 − $3,000,000 = $2,000,000

Assume operating expenses are $1.4 million.

Then:

EBITDA = $2,000,000 − $1,400,000 = $600,000

Therefore:

EBITDA Margin = $600,000 ÷ $5,000,000 = 12%

This allows management to forecast both absolute profitability and profitability margins.

Forecasting Cash Flow

A financial forecast should not stop with the income statement.

A company can report strong revenue and EBITDA while experiencing a cash shortage.

A basic cash forecast can be structured as:

Beginning Cash

+ Cash Collections

− Operating Payments

− Capital Expenditures

− Debt Payments

+ Financing

= Ending Cash

Working capital assumptions are particularly important.

For example:

Accounts Receivable

If forecast revenue is $10 million and management expects an average collection period of 45 days, the forecast should incorporate the expected timing of collections.

Accounts Payable

Similarly, vendor payment terms should influence the timing of cash disbursements.

This is why a strong FP&A forecast connects the:

Income Statement + Balance Sheet + Cash Flow Statement

Forecast Accuracy: How Should You Measure It?

Forecast performance should be measured systematically.

One simple measure is:

Forecast Error = Actual − Forecast

For example:

Forecast sales = $200,000

Actual sales = $180,000

Forecast error:

$180,000 − $200,000 = ($20,000)

The forecast was $20,000 higher than actual results.

A percentage error can be calculated as:

($20,000 ÷ $200,000) × 100 = 10%

Therefore, actual sales were 10% below the forecast.

However, avoid relying exclusively on a simplistic “accuracy rate.” A forecast should also be evaluated for bias.

If forecasts consistently exceed actual results, the organization may have a systematic upward bias.

If forecasts consistently fall below actual results, management may be forecasting too conservatively.

Variance Analysis: The Missing Link

Forecasting becomes much more valuable when combined with variance analysis.

After actual results become available, compare:

Actual vs. Forecast

Then investigate the drivers.

For example:

Revenue Variance = Actual Revenue − Forecast Revenue

But management should go further:

Volume Variance

Price Variance

Mix Variance

Timing Variance

This transforms forecasting from a static financial exercise into a continuous learning process.

Rolling Forecasts

Traditional annual budgets can become outdated quickly.

A rolling forecast continuously extends the forecast horizon.

For example, a 12-month rolling forecast might always maintain the next 12 months.

After January closes:

Remove January
Add the following January
Update assumptions for the remaining months

The result is a continuously refreshed view of future performance.

The original source specifically recommends rolling forecasts as a way to continually update expectations rather than relying exclusively on a forecast established at the beginning of the fiscal year.

Common Forecasting Mistakes

1. Using Historical Growth Without Understanding the Drivers

Applying 5% growth because revenue grew 5% last year may be inappropriate if the business environment has changed.

2. Ignoring Seasonality

A December forecast should not necessarily be based on an average monthly revenue figure if the business experiences significant seasonal demand.

3. Overly Optimistic Assumptions

Management forecasts can become targets rather than objective estimates.

A forecast should represent what the organization reasonably expects—not simply what management wants to achieve.

4. Ignoring Capacity Constraints

A company cannot forecast unlimited growth if it lacks:

Employees
Production capacity
Inventory
Equipment
Facilities
Working capital

5. Failing to Update the Forecast

A forecast prepared six months ago may no longer reflect current conditions.

6. Excessive Model Complexity

More sophisticated models are not automatically better.

The best forecasting model is one that is:

Accurate + Understandable + Maintainable + Actionable

Technology and Forecasting

Technology can significantly improve forecasting efficiency.

Tools such as:

Microsoft Excel
Microsoft Power BI
ERP systems
FP&A platforms
Statistical forecasting software
Business intelligence systems

can automate data collection, calculations, visualization, and scenario analysis.

The original article also recommends leveraging technology, involving cross-functional teams, documenting the process, and continuously updating forecasts.

However, technology does not eliminate the need for management judgment.

A sophisticated model using poor assumptions can produce a sophisticated-looking but unreliable forecast.

Best Practices for Building a Reliable Forecast

A professional forecasting process should follow these principles:

Use reliable data

Validate the underlying historical information before modeling.

Forecast the drivers

Whenever possible, connect financial results to operational activity.

Document assumptions

Every significant assumption should have a clear rationale.

Use scenarios

Understand the financial consequences of different outcomes.

Integrate the financial statements

Revenue, expenses, balance-sheet accounts, and cash flow should be connected.

Measure forecast accuracy

Compare actual results with previous forecasts.

Analyze variances

Investigate why actual results differed from expectations.

Update continuously

Use rolling forecasts when business conditions change rapidly.

Involve the business

Finance should collaborate with sales, operations, marketing, procurement, and other departments.

Final Takeaway

Calculating a forecast or projection is not simply a matter of applying a percentage to historical financial results.

A reliable forecast begins with a clear objective, uses validated historical information, identifies the operational drivers of performance, establishes explicit assumptions, selects an appropriate forecasting methodology, and continuously compares actual results with expectations.

The most useful forecasting models connect operations to financial results.

Instead of asking only:

“What will revenue be next year?”

a stronger FP&A process asks:

“What volume will we generate, what price will we achieve, what customers will drive that volume, what resources will be required, what costs will those activities generate, and how will those assumptions affect profitability and cash flow?”

That is the difference between a simple financial projection and a driver-based financial forecast.

Forecasting will never eliminate uncertainty. Its purpose is to make uncertainty more manageable.

A well-designed forecast gives management a framework for answering three critical questions:

Where are we likely to go?

What could cause us to deviate from that path?

What actions should we take now?

When forecasting becomes an ongoing process rather than an annual exercise, it becomes one of the most powerful tools available to finance and management teams for planning, performance management, risk management, and strategic decision-making.