The Case for a Single Input Page
The appeal of a single input page lies in its simplicity and centrality. Having all inputs in one place enables a quick overview of all the assumptions driving the model, and this consolidated view facilitates easier review and adjustments, streamlining the modeling process.
As Timothy R. Mayes, author of "Financial Analysis with Microsoft Excel", says, "Consolidating all model inputs on a single worksheet can enhance the clarity of your model and make it much easier to modify your assumptions."
Such an approach can significantly benefit sensitivity analysis. With all variables in one location, adjusting inputs to evaluate different scenarios becomes more streamlined. It also aids in documenting changes, which is crucial when multiple analysts work on the same model.
The Case for Spreading Inputs Throughout the Model
On the other hand, spreading inputs throughout the model may provide more contextual relevance. Inputs located in the sections where they're directly used can enhance the understanding of the model's structure and the interrelations between its components.
John Tjia, in his book "Building Financial Models," argues, "Embedding inputs within their respective sections of the model helps to maintain a clear line of sight between an assumption and its impact, aiding in understanding and troubleshooting the model."
Distributing inputs can also aid in sanity-checking the consistency of assumptions. Locating inputs close to their corresponding calculations makes it easier to check that the assumptions align with the resulting outputs visually.
Striking a Balance
While both approaches have merits, it often boils down to the model's complexity and the analyst's preference. Some prefer a hybrid approach: key assumptions and drivers are placed on a single input sheet for easy manipulation and high-level analysis, while more granular, module-specific assumptions are embedded within their respective sections.
This approach strikes a balance between maintaining an overview and ensuring contextual relevance. As Danielle Stein Fairhurst, one of my fellow Master of Modeling inductees, states "A well-organized model will gather its inputs in a central location, but it's also helpful to have some context around where and how these inputs are used."





