What Are Periods?

Periods define the time axis of your financial model. Every data point in the model -- every revenue figure, every cost entry, every balance -- is associated with a specific period. The collection of periods determines how far back your historical data reaches and how far forward your projections extend.

A well-structured set of periods is essential because financial analysis is fundamentally about how numbers change over time: comparing this year to last year, tracking trends across quarters, or projecting growth over the next five years.

Frequency

The Frequency setting determines the granularity of your time axis:

Annual

One period per year. This is the most common choice for long-range strategic models, valuation models, and high-level planning. Each period represents a full fiscal year.

Daily rate calculations: When the model computes daily rates (for depreciation, interest accrual, working capital adjustments, etc.), it divides annual figures by the number of days in the period -- typically 360 or 365.

Quarterly

Four periods per year. Suitable for models that need to capture seasonal patterns or align with quarterly reporting cycles. Each period represents one fiscal quarter.

Monthly

Twelve periods per year. The most granular option, used for detailed operational models, cash flow forecasting, and short-term planning. Each period represents one calendar month.

Important: Higher frequency means more columns in the generated Excel workbook. A 10-year monthly model produces 120 period columns, which can make worksheets wide and harder to navigate.

Period Types

Each period is classified as one of two types:

Actual

Actual periods contain historical data -- real, recorded financial results. These periods correspond to the past. Data for actual periods is typically imported from accounting systems, ERP exports, or other authoritative sources.

In the generated model, actual periods are visually distinguished (often with different formatting) so analysts can immediately tell where historical data ends and projections begin.

Projected

Projected periods contain forecasted values -- estimates of future financial performance. These periods correspond to the future. Projected values are driven by economic indexes, growth assumptions, and formulas rather than imported data.

The boundary between actual and projected periods is one of the most important structural decisions in a financial model. It determines where the model transitions from "what happened" to "what we expect to happen."

Days in Period

Each period has a Days in Period value that the model uses for daily rate calculations. Examples of where this matters:

• Depreciation -- Converting annual depreciation to the correct amount for a partial year or quarter.
• Interest accrual -- Computing interest charges based on the exact number of days.
• Working capital -- Calculating daily cash requirements or receivable turnover.

Days in Period is set automatically based on the frequency selected:

| Frequency | Days in Period | |------------|---------------| | Annual | 365 | | Quarterly | 90 | | Monthly | 30 |

The frequency and days in period always go together -- selecting a frequency automatically sets the corresponding day count. Period names are free-form strings that can be set to any convention you prefer, but the actual analytical parameter used in calculations is the Days in Period value, not the period name.

Period Naming Conventions

Periods are identified by short names that appear as column headers throughout the model. The naming convention depends on the frequency:

• Annual: Year numbers (e.g., "2022", "2023", "2024")
• Quarterly: Year and quarter (e.g., "Q1 2024", "Q2 2024")
• Monthly: Month abbreviation and year (e.g., "Jan-2024", "Feb-2024")

The Specs Wizard generates these names automatically from a reference date. In the manual editor, you can define custom names, but it is strongly recommended to follow a consistent, chronological naming pattern.

Using the Wizard to Generate Periods

The Specs Wizard simplifies period setup by asking for:

1. Frequency -- Annual, Quarterly, or Monthly.
2. First historical year -- The starting year for actual data.
3. Number of historical periods -- How many actual periods to create.
4. Number of projected periods -- How many projected periods to create.

From these inputs, the wizard generates a complete, correctly ordered set of periods with appropriate names, types, and day counts. You can then fine-tune individual periods in the editor if needed.

How Periods Connect to Other Sections

• Database -- The "Period" column mapping tells the engine which column in your data file identifies the period. Period names in the data must match the period names defined here.
• Indexes -- Each economic index has a value for every period. The index worksheet has one column per period.
• Scenarios -- Scenarios inherit the period structure. Each scenario's index values span all defined periods.
• Accounts -- Account data is organized by period. In the generated model, each account row has one value column per period.

Best Practices

1. Start with the right frequency. Changing frequency later means reconfiguring periods from scratch. Think about your reporting needs before choosing.
2. Include enough history. At least 2-3 years of actual data gives the model meaningful trend context. More history is better for trend analysis but adds workbook width.
3. Be realistic about projections. Five years of projected periods is standard for most models. Very long projection horizons (10+ years) add complexity and imply false precision.
4. Keep period names consistent. The engine matches imported data to periods by name. Inconsistent naming causes import failures.
5. Verify the actual/projected boundary. The transition point between actual and projected periods drives how the model treats data. Double-check that your most recent completed period is marked as Actual.