Modeling is a Discipline

A financial model is a critical decision-making tool that allows users to make accurate and informed financial decisions about a company. A financial model needs to serve as a powerful communication tool to clearly and effectively tell the story of a company to executives and stakeholders.

A builder of financial models requires skills in accounting, finance, spreadsheets and overall business knowledge in order to effectively forecast a company's financial statements into the future.

With excellent modeling skills, a professional can:

• Read and properly understand a company’s historical financial statements.
• Ask strategic questions to help frame thinking about the future.
• Perform effective research in order to determine optimal assumptions.
• Challenge conventional thinking and adopt alternate approaches.
• Coordinate the efforts of all contributing professionals.

A strong financial modeler will be able to:

• Demonstrate exceptional Excel skills and keyboard proficiency.
• Design and build models with precision, to be used for optimal communication.
• Incorporate scenario and sensitivity analysis to manage risk.

Mastering the discipline of financial modeling facilitates the communication of outputs and recommendations with poise and confidence.

The Key Attributes of a Model

To achieve the goals set out above, a model needs to be developed with the following attributes:

Flexible: The model contains scenarios with multiple cases (i.e. base case, best case, worst case) for the most critical variables.

Intuitive: The reader can easily follow and understand the model based on the flow of the pages.

Transparent: Every formula can be understood whether the reader clicks into the cell, or looks at a printout on paper.

Dynamic: Any changes to the assumptions should properly flow through the model.

Tells a story: The model is a powerful communication tool that allows for optimal decision making.

Transferable: When a model is well designed and well built, colleagues and clients will be able to use the file as if they built it themselves.

When a model achieves the criteria above, it creates tremendous credibility for the builder and inspired confidence with the reader.

A well-designed model needs to work in two ways:

• Electronically: A model must be simple to review / audit / manipulate in a spreadsheet to ensure it can be used as a tool in the decision-making process.
• In printed format: Many senior executives and other decision makers prefer to review financial analysis on paper or as PDF (rather than reviewing a spreadsheet).

Model Planning

It is critical to properly plan and design a model before it can be built.

Whenever a model becomes filled with errors, it is often because the modeler did not properly devise a model plan. A model built without forethought is often chaotic and requires many additions on a weak foundation.

A model builder needs to first understand how a business operates, which requires the following:

Review historical financial statements: Review management discussion, analysis and financial statement notes.

Learn about the company's industry: Reading industry journals and equity research reports.

Decide on key drivers: Key drivers that are critical success factors for the industry / company.

Understand the purpose of the model: Is the model used for valuation, credit assessment, project evaluation, etc?

Identify required schedules and components: Plan for all specific calculations such as revenues, costs, working capital, etc.

Identify the required schedules and components to be included in the model.

• Schedules should be planned for all specific calculations needed in the model, such as revenues, costs, depreciation, working capital, etc.
• Take some time to map out these components and model flow, the same way you would when creating and outline for a presentation. Create a skeleton of your model.

As part of the planning process, the model builder should also be able to answer the following questions:

• Can a reader of the model understand the operations of the company?
• Have the assumptions been challenged and vetted?
• Which assumptions are key drivers to be tested as scenarios?
• Is there enough historical detail to validate the required amount of forecast detail?
• What is the important output that the users want to know?
• What metric is used to measure results?
• What's the appropriate periodicity for the model (e.g monthly, quarterly, annual)?
• How many time periods should be included in the forecast? Why?
• What is the most logical order to present the data?