Financial modelling is where finance stops being a pen-and-paper exercise and becomes an engineering discipline. Two students can arrive at the same enterprise value with very different models — one clean, auditable, and defensible; the other a tangle of hardcoded numbers, circular references, and formulas nobody can trace. Markers can tell the difference at a glance, and the grade difference is bigger than most students expect.
This guide takes you through what a financial modelling assignment actually tests, the Excel conventions that separate professional models from amateur ones, the three-statement linkage that underpins nearly every DCF, a full worked five-year DCF valuation with terminal value and sensitivity, and the presentation moves that lift a competent model into a top-grade one.
What a Financial Modelling Assignment Actually Tests
A financial modelling assignment tests three connected skills. The mistake most students make is treating this as a finance assignment done in Excel. It is not — the modelling craft carries its own marks.
- Model construction — Can you build a transparent, well-structured Excel model that separates inputs from calculations from outputs, uses consistent formulas, and can be audited by someone who did not build it?
- Financial content — Can you forecast revenues, model working capital, compute free cash flow, apply an appropriate discount rate, and value the business using DCF and comparable methods?
- Sensitivity & interpretation — Can you show how the valuation changes as key assumptions change, identify which drivers matter most, and defend a valuation range rather than a false-precision point estimate?
Read the brief for what "the model" means: "Build a DCF model" is a valuation model with a five-to-ten-year explicit forecast and a terminal value. "Build a three-statement model" is a linked income statement, balance sheet, and cash flow statement — usually as the foundation for DCF. "Build an LBO model" is private-equity style and needs debt schedules and returns analysis. Structure differs; identify the model type before opening Excel.
Excel Best Practice — The Standard Professional Model Follows
The single biggest gap between undergraduate models and professional ones is structural discipline. These conventions are what investment banks, consultancies and asset managers use, and what markers look for even when the brief does not spell them out.
1. Separate Inputs, Calculations, and Outputs
Every good model has three logical zones. Inputs (assumptions the user changes) sit in a clearly labelled input section, typically at the top or on a dedicated tab. Calculations reference those inputs by cell reference — never hardcoded. Outputs (the valuation, sensitivity tables, summary) are presented separately. Mixing these three is what makes models unauditable.
2. Colour-Code Cells by Type
The industry convention: blue font for hardcoded input values, black font for calculations and formulas, green font for cross-tab references. This single convention makes it possible to scan a sheet and immediately see what the user is meant to change vs what should never be touched. Markers who model professionally will notice and reward this. Markers who don't will still find it easier to trace your logic.
3. No Hardcoded Numbers Inside Formulas
A formula like =B5*1.08 is a bug waiting to happen. If growth changes, someone has to hunt through every formula to update the 1.08. A formula like =B5*(1+$C$2) — where C2 is a labelled input — is auditable, changeable, and correct. The one exception is universal constants (days in a year, months in a quarter) — everything else lives in the input section.
4. One Formula Per Row
If your DCF model has a "Revenue" row, every cell in that row (across all forecast years) should use the same formula, copied across. Different formulas in different columns of the same row is the modelling equivalent of a broken structural beam — the model looks consistent but behaves inconsistently. Anyone auditing the model will assume year 2's formula is like year 1's; if it isn't, they will spot it and mark you down.
5. Build Error Checks
Balance sheets that balance. Cash flow statements where opening cash + net change = closing cash. Total sources = total uses in an LBO model. Add these as explicit check rows in the model — =IF(A=B,"OK","ERROR"). When something breaks (and something always breaks), the error check tells you where. In a submitted assignment, a visible "All checks OK" row signals that you understand what a model needs to prove about itself.
The Three-Statement Model — What Links to What
Most valuation assignments are built on a three-statement model as the foundation. The three statements — income statement, balance sheet, and cash flow statement — are not independent. They lock together through specific linkages that must be built into the model.
| Linkage | From | To |
|---|---|---|
| Net income | Income statement (bottom line) | Retained earnings on balance sheet + start of cash flow statement |
| Depreciation & amortisation | Income statement (expense) + cash flow (add-back) | Balance sheet (reduces PP&E carrying value) |
| Capex | Cash flow (investing outflow) | Balance sheet (increases PP&E) |
| Working capital changes | Balance sheet (period-on-period change in current assets/liabilities) | Cash flow (operating adjustment) |
| Ending cash | Cash flow statement (opening + net change) | Balance sheet (cash line) |
When these linkages are built correctly, the balance sheet balances automatically — no manual plugging. When they are not, the balance sheet will not balance, and you will spend hours hunting the error. This is exactly why the error-check row matters.
A Worked Example — 5-Year DCF Valuation
The most common financial modelling assignment builds a DCF valuation for a company. Here is how the full end-to-end calculation works. The setup is constructed for demonstration; the method is identical for any real target.
Revenue growth (Years 1–5): 8% per annum
Free cash flow margin: 15% of revenue
WACC (discount rate): 10%
Terminal growth rate: 3%
Forecast horizon: 5 years
Valuation method: DCF with Gordon Growth terminal value
Step 1 — Forecast Revenue and FCF
Year 2: Revenue £583.20m × 15% = FCF £87.48m
Year 3: Revenue £629.86m × 15% = FCF £94.48m
Year 4: Revenue £680.24m × 15% = FCF £102.04m
Year 5: Revenue £734.66m × 15% = FCF £110.20m
In Excel, this would be one row of revenue (growing at the input growth rate), one row of FCF margin (input), and one row multiplying the two. Three formulas that could be copied across five columns — that is the "one formula per row" convention in action.
Step 2 — Discount the Explicit Forecast to Present Value
Year 2: £87.48m ÷ (1.10)2 = £87.48 × 0.8264 = £72.30m
Year 3: £94.48m ÷ (1.10)3 = £94.48 × 0.7513 = £70.98m
Year 4: £102.04m ÷ (1.10)4 = £102.04 × 0.6830 = £69.69m
Year 5: £110.20m ÷ (1.10)5 = £110.20 × 0.6209 = £68.43m
Sum of PVs (Years 1–5) = £355.03m
Step 3 — Compute Terminal Value (Gordon Growth)
The explicit forecast ends at Year 5, but the business is assumed to continue indefinitely. Terminal value captures all cash flows beyond the forecast period, using the Gordon Growth formula:
Where g is the terminal growth rate (assumed perpetual).
TV (at end of Year 5) = £113.51m ÷ (10% − 3%)
= £113.51m ÷ 0.07
= £1,621.51m
PV of TV = £1,621.51m ÷ (1.10)5
= £1,621.51m × 0.6209
= £1,006.83m
Step 4 — Compute Enterprise Value
= £355.03m + £1,006.83m
= £1,361.86m ≈ £1,362m
Interpretation: The DCF values the enterprise at approximately £1,362m. Notice something important — of that total, £1,007m (74%) comes from the terminal value, not the explicit forecast. This is not unusual: in most DCFs, terminal value accounts for 60–80% of enterprise value. It also means the entire valuation is disproportionately sensitive to two assumptions — WACC and terminal growth — that we have essentially guessed. That is why sensitivity analysis is not optional.
Financial modelling assignment due — and Excel isn't your comfort zone?
Our finance specialists build full DCF, LBO, and three-statement models — with clean structure, error checks, and sensitivity analysis — to your university's marking criteria.
Sensitivity Analysis — Where Marks Are Won
A single-point valuation of £1,362m is arithmetic. A sensitivity table showing how the valuation changes across a plausible range of WACC and terminal growth assumptions is analysis. This is the single biggest scoring opportunity most students miss.
2.0% 2.5% 3.0% 3.5% 4.0%
WACC 8% 1,650 1,773 1,920 2,100 2,325
WACC 9% 1,408 1,494 1,594 1,713 1,855
WACC 10% 1,227 1,290 1,362 1,445 1,541
WACC 11% 1,087 1,134 1,188 1,248 1,317
WACC 12% 974 1,011 1,052 1,098 1,150
Interpretation: Across a plausible range of assumptions, enterprise value spans from £974m (WACC 12%, g 2%) to £2,325m (WACC 8%, g 4%) — a factor of 2.4×. That range is the honest answer to "what is this business worth?" The £1,362m base case is one point inside that range; it is not the truth. A strong recommendation would quote a valuation range like £1.2–1.5bn as more defensible than a single figure, and explicitly acknowledge which assumptions the valuation is most sensitive to.
In Excel, build this using a two-way Data Table (Data → What-If Analysis → Data Table). Set WACC as the column input, terminal growth as the row input, and reference your EV cell. This gives you the sensitivity grid in one operation — professional and auditable.
2:2 vs First: The Sensitivity Paragraph
Both students arrive at the same base-case £1,362m EV. The gap between grade bands sits in what they do with it.
"The DCF valuation gives an enterprise value of £1,362m using a WACC of 10% and terminal growth of 3%. This is the estimated value of the business. The valuation is based on the free cash flow forecast and the assumed discount rate."
"The base-case DCF valuation is £1,362m, but the sensitivity table shows the valuation ranges from £974m to £2,325m across plausible WACC (8–12%) and terminal growth (2–4%) combinations — a 2.4× spread. Terminal value accounts for 74% of the base-case EV, meaning the valuation is disproportionately sensitive to two assumptions I cannot observe directly. A more defensible recommendation is a valuation range of £1.2–1.5bn, with the following key caveats: reducing WACC by 200bp adds ~£560m to EV; a 100bp shift in terminal growth moves EV by ~£70m. The base case should be read as one point in this range, not as the answer.
Five Mistakes That Cost Students Marks
Hardcoding numbers inside formulas. =Revenue*1.08 instead of =Revenue*(1+$C$2). Not only unauditable but breaks the moment the marker changes the growth assumption to test sensitivity.Fix: Every assumption lives in a labelled input cell. Every formula references it by cell reference. No exceptions except universal constants.
Mixing inputs, calculations, and outputs on the same sheet without visual separation. Makes the model impossible to audit and signals amateur work.Fix: Use colour coding (blue inputs, black formulas, green cross-tab links) and clear section breaks. Even a single-tab model can have three visually distinct zones.
No error checks or balance sheet non-balancing. A three-statement model where the balance sheet does not balance is a broken model, and the difference between "close to balancing" and "balancing" is not judgement — it is a real error you have not found.Fix: Include an explicit check row: =IF(Assets=Liab+Equity,"OK","ERROR: "&(Assets-Liab-Equity)). Fix the error before submission, do not paper over it.
Single-point valuation with no sensitivity analysis. Reports "the value is £1,362m" as if it were a fact rather than a function of assumptions. Loses marks even when the calculation is correct.Fix: Always include at least a two-way sensitivity table. Discuss the valuation range in your write-up, not just the base case.
Ignoring terminal value dominance. Not mentioning that terminal value is often 60–80% of EV signals shallow understanding. It is the single most important sensitivity point in a DCF.Fix: State the terminal value share of EV explicitly. Discuss its assumption dependence (WACC, g). This one paragraph consistently lifts a grade band.
Frequently Asked Questions
What is a financial modelling assignment usually asking me to do?
What Excel conventions should I follow in a professional model?
How do I decide on the WACC and terminal growth rate?
How much of DCF value should the terminal value be?
Should I use a data table or manual sensitivity in Excel?
My deadline is close — what should I prioritise on a financial modelling assignment?
Need Your Financial Modelling Assignment Done Right?
Our expert writers build fully worked financial models — DCF valuations, three-statement models, sensitivity tables — with the clean Excel structure and interpretation that wins marks. To your university's marking criteria and your exact deadline.
Get Expert Help Today →
