Can the Next Analyst Defend the Assumptions in Your Financial Model and Why a Book of Assumptions Matters
Receiving a financial model without its surrounding context is a common handover problem. The formulas may still work, and the outputs may look perfectly clear, but the incoming analyst may not know:
Why the model was originally prepared.
Which assumptions are fundamental to the forecast.
Where those assumptions came from.
Which methodologies were applied.
What the model is designed, and not designed, to do.
Which assumptions can be changed and who has authority to change them.
This matters because two models can contain very similar inputs, calculations and outputs while serving very different purposes. A model developed to assess potential synergy benefits from an acquisition, for example, should not automatically be treated in the same way as a model prepared for internal budgeting.
There are therefore two elements to a successful financial model handover: good documentation and a capable analyst. The Book of Assumptions provides the context. The analyst must be able to understand that context, challenge it where appropriate and translate new information into sensible changes to the model.
Start With Context Before Testing Individual Assumptions
Before asking an incoming analyst to defend a revenue growth assumption, they should first understand what the model is trying to achieve. The Book of Assumptions should therefore start by documenting the model's purpose and objectives. It should explain the intended use of the model and the decisions it was designed to support.
It should also establish the fundamental conventions governing the workbook, including:
Currency and units of measure.
Forecast timeline and periodicity.
Treatment of inflation.
Sign conventions.
Input, calculation, output and linked-cell formats.
Model checks and controls.
Categorisation of key line items.
Key exclusions, restrictions and limitations.
The limitations section is particularly important. Every financial model simplifies reality in some way. Users should understand those boundaries before relying on its outputs for decision-making. This is also a key part of Projectify’s existing Book of Assumptions framework.
Before delving into the detail of the formulas, the incoming analyst should review the purpose of the model, its setup, what it can reasonably be used for and where its limitations could affect a decision.
Give the Analyst a Map of the Financial Model
A good Book of Assumptions should provide a high-level map of the workbook rather than requiring a new analyst to spend hours trying to understand it by reverse-engineering it.
Such a map would normally distinguish between input, calculation and output tabs and explain the purpose of the main worksheets and how information flows between them. For example, the documentation might explain that operating assumptions are entered in dedicated input tabs, flow through calculation schedules and ultimately feed the financial statements, dashboard and KPIs.
The same applies to model outputs. Important tables, graphs, KPIs and dashboards should be documented together with what they are intended to communicate.
This enables the incoming analyst to understand the architecture of the model before diving into individual cells.
Understand the Source and Quality of Assumptions
A financial model should document not only what assumptions are used, but also where they came from, why they were selected and how much reliance can reasonably be placed on them.
For material assumptions, the Book of Assumptions should record, where relevant:
source of the assumption;
date or period of the source;
person or function responsible for providing it;
rationale for its use;
supporting evidence;
reliability or confidence in the source;
how it was validated or sense-checked; and
when it was last reviewed.
A 10% revenue growth assumption, for example, carries a very different level of credibility depending on whether it is supported by contracted orders, a detailed sales pipeline, historical performance, external market data or management judgement.
It is also useful to distinguish between assumption ownership and model ownership. The analyst may be responsible for incorporating an assumption correctly, but the underlying business assumption may belong to management, sales, operations or another function.
Greater attention should be given to assumptions that are highly judgemental, uncertain or materially affect the model outputs.
Documenting the source, reliability and ownership of assumptions helps the next analyst understand not just what number has been used, but also how much confidence to place in it, who to speak to if it needs to be challenged, and what evidence should be considered before it is changed.
Test One Assumption From Source to Output
Rather than asking the analyst to understand an entire workbook, select one material assumption, such as sales volume growth, and ask the analyst to trace:
Source → Assumption → Calculation → Financial impact → Output
The Book of Assumptions should make that possible without requiring the analyst to spend hours tracing formulas across the workbook
For each major forecasting area, it should explain in plain language:
What inputs are used.
How those inputs are applied in the model.
What calculation methodology is applied.
How those calculations ultimately affect the forecast.
This could cover areas such as revenue, direct costs, operating expenses, staff costs, working capital, receivables, payables, fixed assets, borrowings, equity, reserves and cash flow. Where useful, the underlying calculation can also be expressed in simple words. For example:
Revenue = forecast units sold × average selling price.
The underlying Excel formula could be considerably more complex, but the documentation should allow someone with a reasonable finance background to understand what the model is actually doing. Not every stakeholder has the same level of Excel expertise, and even experienced modellers can find it time-consuming to follow complex formulas across multiple worksheets.
Test the Explanation, Not Just the Calculation
A technically capable analyst may be able to identify where an assumption sits in the workbook. The more important question is whether they understand why it is there. During recruitment, ask the candidate to walk you through the model live rather than only submitting a completed workbook. How is it structured? What are the main assumptions, where do they come from, and how do they affect the outputs? Can the candidate explain why a particular approach was used?
For the selected assumption, ask:
Why is this assumption reasonable?
What evidence supports it?
Who owns the underlying business assumption?
When was it last reviewed?
What uncertainty surrounds it?
What evidence would make you reconsider it?
The discussion should go beyond explaining the formulas. Ask the candidate what the results mean for the business. If sales increase, what happens to staffing requirements, working capital and cash flow? Can they explain these implications to a client or manager who has not worked on the model?
This distinguishes Excel competency from financial judgement. It also reinforces an important principle: the model should not become the source of the business assumption simply because the assumption happens to be stored in Excel.
Introduce Conflicting Evidence
Suppose the model assumes a 15% increase in sales volumes, but a new operational update indicates that manufacturing capacity is constrained. A strong analyst might first determine whether the capacity constraint is temporary or structural, establish what period is affected, discuss the issue with the relevant assumption owner, run alternative scenarios and assess the financial impact before recommending any change to the base case.
You can introduce this situation during the live presentation. Ask the candidate to show you what would change in the model and explain the business implications. What would they recommend? What information is missing? Which changes could they make independently, and which would require approval?
The purpose is not to see how quickly they can change a cell, but whether they understand what they are changing and why. That tests both modelling capability and judgement. It also highlights the distinction between updating a model and authorising a change to a business assumption.
The Book of Assumptions Supporting the Financial Model Should be a Live Document
The Book of Assumptions should clearly identify the version of the financial model to which it relates. If the methodology of the model changes, the documentation should change with it. It should remain a live companion to the financial model rather than a document created once and then forgotten.
There is also a practical benefit for the original modeller. Months or years later, even the person who built the model may no longer remember why a particular methodology was chosen. Good documentation provides that institutional memory.
The Analyst Remains Critical
A Book of Assumptions does not compensate for a poorly constructed financial model.
The analyst building or reviewing the model still needs to:
Understand the requirements of the end user and translate them into the model.
Build the model using clear and consistent modelling principles.
Select appropriate methodologies and assumptions.
Sense-check the outputs.
Present and explain the model and its conclusions.
Deal with questions and challenges from stakeholders.
Hand the model over in a way that allows others to use it independently.
The Book of Assumptions is therefore not a substitute for the analyst. It is the final layer of governance around a well-developed and reviewed financial model. The analyst also needs judgement when preparing the documentation. Too little detail leaves important questions unanswered; too much creates a cumbersome manual that nobody uses.
The aim is to document enough that another competent finance professional can understand the model without going through every Excel formula and cell.
Maintain a Decision Trail
For material assumptions, the Book of Assumptions should also retain a clear record of significant changes over time. This could include:
Previous assumption | Revised assumption | Reason for change | Date changed | Forecast impact | Approved by
This becomes particularly important where assumptions evolve during budgeting, forecasting, transactions or scenario analysis. It allows a reviewer to distinguish between a simple model update and a formally approved change in the underlying business assumptions.
Why This Matters Beyond Handover
A well-prepared Book of Assumptions supports:
Model handovers.
Internal review and challenge.
Board and management decision-making.
Model audits.
Transaction and due diligence processes.
External stakeholder review.
Future model updates.
This is where documentation also adds credibility. Centralising the assumptions and methodology makes it easier for reviewers to trace and challenge the reasoning behind the forecasts, which strengthens auditability and accountability.
It also helps the analyst who created or reviewed the model. A transparent model accompanied by clear documentation builds trust with the end users and allows the analyst to hand it over properly rather than creating permanent dependency on the person who originally built it.
A good financial model should survive the person who built it. The spreadsheet provides the calculations. The Book of Assumptions provides the context. The analyst brings the judgement required to understand, challenge and update both.
About the Authors
David Busuttil is the founder of Projectify, a financial modelling business providing Excel models, bespoke financial modelling and web-based finance tools. He is a CPA, CFA and ACMA, with professional experience across financial planning, forecasting, investment appraisal and model development. His work focuses on building practical, transparent financial models with clear assumptions and methodology.
Sergio Ermacora is co-founder of Nexteam, which places finance and accounting professionals, including investment analysts and associates, with investment firms and operating businesses. His background is in investment banking, and his work focuses on assessing candidates’ technical skills, business judgement and ability to explain their work.
Comments