Financial Modelling: Building Models That Test Your Plans Before You Commit

What financial models are for

A financial model is a structured representation of a business, project or transaction in numbers. It turns assumptions about customers, prices, costs, investments and financing into projected profit, cash flow and balance sheet, and lets you see what happens when those assumptions change.

Models support many decisions: raising funds, evaluating investments, planning growth, pricing, assessing acquisitions, applying for loans and setting budgets. A good model does not predict the future perfectly; it makes the consequences of choices and uncertainties visible. This guide explains how our financial modelling service builds models people can trust.

Types of financial models

The type determines structure and level of detail. A model for a bank loan focuses on cash flows and debt service; a startup model focuses on growth, burn and runway.

• Three-statement operating models: integrated profit and loss, balance sheet and cash flow
• Startup and fundraising models with unit economics and runway
• Budget and forecast models for annual planning
• Project and investment models for new plants, products or locations
• Valuation models such as DCF
• Transaction models for acquisitions, mergers and restructuring
• Pricing and unit economics models
• Loan and debt capacity models for lenders

Why three statements matter

Profit and cash are different. A business can be profitable yet run out of cash because customers pay slowly, inventory builds up or capital investment is heavy. A three-statement model links profit and loss, balance sheet and cash flow, so every assumption flows correctly through all three.

The balance sheet acts as a check: if it does not balance, something is wrong. The cash flow statement shows funding needs clearly, which is why lenders and investors expect integrated models.

Driver-based assumptions

Good models are built on drivers, the underlying activities that cause results, rather than arbitrary growth percentages. Revenue might be driven by marketing spend, conversion rates, average order value and retention; or by sales headcount, productivity and win rates; or by capacity, utilisation and price.

Driver-based models are easier to explain, challenge and update. When a manager asks “what if conversion improves by one point?”, the model answers directly. They also connect financial plans to operational targets that teams can act on. Our budgeting & forecasting guide explains drivers in planning.

Structure: assumptions, calculations, outputs

Separating these layers makes models easier to audit and update. Colour conventions, for example one colour for inputs and another for formulas, help users understand what can be changed.

Modelling revenue

Revenue is usually the most important and most uncertain part of a model. Approaches vary by business: customers multiplied by spend for consumer businesses; pipeline, win rates and deal sizes for B2B; capacity and utilisation for services and manufacturing; subscribers, churn and pricing for subscription businesses.

Revenue assumptions should be grounded in evidence: historical performance, market research, pipeline data and realistic ramp-up periods. Market sizing from research helps test whether growth assumptions are plausible.

Modelling costs, working capital and capex

Costs are split into variable costs that move with volume, such as materials and commissions, and fixed or step costs such as salaries, rent and systems. Headcount plans often drive a large share of costs and should be modelled by role and timing.

Working capital assumptions, receivable, inventory and payable days, determine how much cash growth consumes. Capital expenditure, depreciation and financing, including loan drawdowns, repayments and interest, complete the picture. Taxes, including GST timing effects where material, should be reflected sensibly.

Scenarios and sensitivity analysis

The future is uncertain, so a single forecast can mislead. Scenarios model coherent alternative futures, such as base, upside and downside cases, with different combinations of assumptions. Sensitivity analysis changes one assumption at a time to show which drivers matter most.

Together they answer key questions: How much funding do we need in the downside case? Which assumption, if wrong, hurts most? How much buffer should we keep? Investors and lenders increasingly expect this analysis.

Startup and fundraising models

Investors look for a model that shows how the business grows, what drives that growth, how much capital is needed and when, and how unit economics improve with scale. Key outputs include revenue growth, gross margin, burn rate, runway, customer acquisition cost, lifetime value and the use of funds.

A credible fundraising model is ambitious but defensible: assumptions linked to evidence, sensible hiring plans and realistic timelines. It should also align with the business valuation narrative and survive detailed questioning during due diligence.

Project and investment models

Project models evaluate specific investments: a new factory, product line, branch or technology system. They focus on incremental cash flows, investment costs, ramp-up, operating costs, financing and returns measured by net present value, internal rate of return and payback period.

For loan-funded projects, lenders examine debt service coverage, the ability of cash flows to cover interest and repayments, so models must show these clearly and test them under stress.

Best practices

Industry standards such as the FAST modelling standard offer useful conventions for building transparent spreadsheets.

• Start with the decision the model must support
• Keep it as simple as possible, but no simpler
• Put all assumptions in one clearly labelled place
• Avoid hard-coded numbers inside formulas
• Use consistent time periods and formulas across rows
• Build in checks and make them visible
• Document sources, logic and limitations
• Have someone else review the model before it is used

Common modelling errors

Most of these are prevented by good structure and independent review.

• Hard-coded numbers hidden in formulas
• Circular references and inconsistent formulas
• Balance sheets that do not balance
• Ignoring working capital and its cash impact
• Hockey-stick growth with no supporting drivers
• No scenarios or sensitivities
• Models so complex that nobody else can use them

An example: a model for expansion funding

A manufacturing company seeks a term loan to add a production line. The bank asks for a projected financial model. The model links capacity, utilisation and prices to revenue; material, labour and overhead costs to volume; working capital to sales; and the new loan’s drawdown, interest and repayments to the cash flow.

Scenarios test slower ramp-up and lower prices, and sensitivity analysis shows debt service coverage under each. The bank receives a clear, documented model with summary outputs, and the company’s management gains a tool to track actual performance against plan after the investment.

Unit economics in models

For many businesses, especially startups and subscription companies, unit economics are the heart of the model: what it costs to acquire a customer, how much margin they generate over time and how quickly acquisition costs are recovered. Modelling these explicitly shows whether growth creates value or simply consumes cash.

Unit economics should be built from real data where possible: marketing spend and conversions, average revenue, gross margins and retention curves.

Presenting model outputs

Decision-makers rarely want to explore spreadsheets. A summary dashboard showing key outputs, such as revenue, EBITDA, cash, funding needs, returns and scenario comparisons, with simple charts, communicates results quickly. Detailed calculations remain available for those who need them.

Clear outputs also make models easier to present to boards, investors and lenders, and reduce misunderstandings about what the numbers mean.

Version control and governance

Models evolve through many versions, and confusion over which version is current can lead to wrong decisions. Simple governance helps: clear version naming, a change log, a single master file, restricted editing of key assumptions and review before major uses such as fundraising or board approval.

Linking models to actual results

A model becomes more valuable when actual results are compared with projections regularly. Tracking variances shows which assumptions were optimistic or conservative, improving future forecasts and building credibility with investors and lenders who see that the business understands its numbers.

Modelling for multiple entities

Groups with several companies or business units may need models for each, consolidated into a group view with inter-company eliminations. Consistent structures and assumptions across entities make consolidation reliable and comparisons meaningful.

Tax and compliance in models

Models should reflect tax sensibly: income tax on profits, timing of advance tax payments, GST effects on cash flow where material, TDS on certain receipts and payments, and the treatment of losses. Simplifications are fine when clearly stated, but material tax effects should not be ignored, especially in cash flow forecasts and investment evaluations.

Getting a model built

Share the purpose, timeline, historical financials and your view of the key drivers. A first version can be built quickly for discussion, then refined as assumptions are tested. We hand over models with documentation and training, so your team can update and use them confidently.

Read our Virtual CFO guide, or talk to our modelling team.

Frequently asked questions

• Can you model scenarios for our board?

  Yes, with clear base, upside and downside cases and their implications for cash and funding.

• Can models be built in Google Sheets?

  Yes, where collaboration matters more than advanced features; we follow the same structure and checks.

• Can you model debt repayments and interest?

  Yes. Models include loan drawdowns, repayment schedules, interest and coverage ratios that lenders review.

• Do models include charts and dashboards?

  Yes. We include summary outputs and charts for presentations to boards, investors and lenders.

• What is the FAST standard?

  A set of modelling conventions promoting models that are flexible, appropriate, structured and transparent.

• Can models include multiple currencies?

  Yes. Models can handle multiple currencies with explicit exchange rate assumptions and sensitivity to currency movements.

• Can you model the impact of a new product or market?

  Yes. We build incremental models showing investment, ramp-up, costs, cash needs and returns for new initiatives.

• What is a three-statement financial model?

  A model that links projected profit and loss, balance sheet and cash flow so every assumption flows consistently through all three.

• Why do investors want a financial model?

  To understand how the business grows, how much capital it needs, when it becomes profitable and how sensitive results are to key assumptions.

• How long does it take to build a model?

  Simple models can be built in one to two weeks; complex operating, project or transaction models may take three to six weeks.

• Do you use Excel or specialist software?

  Most models are built in Excel or Google Sheets for transparency and ease of use; specialist tools are used where appropriate.

• What is sensitivity analysis?

  Changing one assumption at a time to see how much it affects results, revealing which drivers matter most.

• Can you review a model we already have?

  Yes. We review structure, formulas, assumptions and checks, and suggest corrections and improvements.

• What information do you need?

  Historical financials, current management accounts, business plans, key operational data and your views on growth and costs.

• Can the model be used for budgeting later?

  Yes. Well-built models can be adapted for annual budgets and rolling forecasts.

• Do you build models for bank loans?

  Yes, including debt service coverage analysis and the outputs lenders typically request.

• Will our team be able to update the model?

  Yes. We design models for usability, document them and train your team.

Still curious? Ask the assistant Talk to a specialist →

Explore our Accounts & Finance services