A five year projection separates assumptions from outputs and resolves into a profit and loss, a cash flow and usually a balance sheet. The first 24 months should be monthly and defensible. Beyond that it is a direction of travel, and it should be presented as one.
What a five year projection contains
Three to five tabs, in a fixed order: assumptions, revenue, costs, then a profit and loss and a cash flow that resolves into runway. A balance sheet, capex, tax and super come in as the model matures. Every number on the output tabs should trace back to something on the assumptions tab, and the test of that is whether you can say out loud where each one came from.[1]
The rule that keeps a model usable is the separation of inputs from outputs. Assumptions live in one place, are labelled, and are the only cells anyone types into. Everything else is calculated. A projection with hardcoded numbers buried in output formulas cannot be stress-tested, which means it cannot be used for the thing a five year view is actually for.
A single five year line is not a plan on its own, which is why it is usually paired with a base, downside and upside scenario set.
Granularity: monthly early, annual later
Model the first 12 to 24 months monthly, then move to quarterly or annual. That is not a shortcut, it matches how much conviction you honestly have. The cost of monthly detail in year four is not just effort, it is false precision that invites the reader to interrogate a number nobody should be defending.
The other half of the granularity question is history. The most fluid models show how things have actually happened over the last period, then how those same lines grow into the future, rather than starting from a blank sheet at today's date.[2] If the forecast shows five revenue types with specific growth rates and the P&L shows something else entirely, the model has already failed its first check.
The metrics a five year horizon needs
A long projection introduces measures that a twelve month forecast can get away with ignoring. Worth knowing that only some of them are defined by the accounting standards: the statutory statements follow the AASB accounting standards, and ASIC's guidance for preparers of financial reports covers who has to prepare and lodge them. EBITDA is not one of them, which is why two companies can quote it and mean slightly different things.
| What it measures | What to watch for over five years | |
|---|---|---|
| EBITDA | Earnings before interest, tax, depreciation and amortisation | Not defined by the accounting standards, so check what has been excluded before comparing it to anything. A five year EBITDA line that turns positive on schedule every time is usually a plug. |
| CAGR | Compound annual growth rate across the period | It smooths away the shape of the curve, which is the part that matters. A flat two years followed by a spike gives the same CAGR as steady growth and tells a completely different story. |
| Burn rate | Net cash consumed per month | Watch whether it is gross or net, and whether it moves with the hiring plan. A flat burn line across five years means costs were never phased. |
| Runway | Months of cash left at current burn | The one number the board will find first. It should be visible on the cash tab without anyone doing arithmetic. |
How far it can honestly go
This is the part most templates will not tell you. At early stage the biggest modelling mistake is forecasting out too far: twelve months you should have high conviction on, twelve to twenty-four with line of sight, and beyond that accuracy is a bonus rather than an expectation.[3] A five year projection is still worth building, but years three to five are a direction of travel and should be labelled as one.
The failure to avoid is the top-down wish. Capturing half a percent of a ten billion dollar market by year three is not a model, it is a hope, and experienced readers treat it as a signal that the founder does not know what drives growth in their own business.[4]Build from the unit up. If the year five number cannot be assembled from things you can name, it does not belong in the file.
The discipline that keeps the near years honest is grading the forecast against actuals each period, which is also the only part of this you can measure.
Who owns the long horizon
A five year view usually exists because someone outside the business asked for it: an investor, a lender, a board planning an exit. Once a company is profitable and growing it reaches the point where a liquidity event becomes a realistic conversation rather than an aspiration,[5] and the projection is the document that conversation runs on. Driving that event, whether an IPO, an acquisition, a raise or an exit, is a job that demands commercial judgement, financial modelling and investor negotiation in one person.[6]
Which is why the five year plan often changes who you need rather than the other way around. I have seen a CFO of five years who was genuinely strong on governance and controls, and had brought a business back to profitability, part ways by mutual agreement because the skill set was not aligned with the next phase of growth.[7] That is not a failure of the person. It is what happens when the plan and the incumbent are pointed at different decades. When the business pivots into new markets, finance often operates like a startup again, needing re-evaluated processes, a different hiring strategy and reporting that scales so investors can keep trusting it through rapid change.[8]
Common questions
What should a 5 year projection include?
A stated assumptions layer, revenue modelled by stream, a costed hiring plan and cost base, then a profit and loss and cash flow that resolve into a runway line. A balance sheet, capex, tax and super come in as the model matures. Every output number should trace back to a named assumption, and the practical test is whether you can say out loud where each one came from.
How accurate can a 5 year projection be?
Only the front of it. At early stage you should have high conviction on the first twelve months, line of sight on months twelve to twenty-four, and beyond that accuracy is a bonus rather than an expectation. It is still worth building all five years, because the exercise forces the shape of the business into the open. Label years three to five for what they are, so nobody defends them line by line.
Should the projection be monthly or annual?
Monthly for the first 12 to 24 months, then quarterly or annual. That matches the conviction you actually have. Monthly detail in year four is false precision, and it invites the reader to interrogate numbers nobody in the business should be defending. It also helps to line the forecast up against historical actuals rather than starting from a blank sheet at today's date.
Who should build a startup's 5 year projection?
Someone who can defend it, which in practice means whoever will be in the room when an investor, lender or acquirer asks where a number came from. Founders build the first version and that is fine. The plan tends to outgrow them at the point where a liquidity event becomes a realistic conversation, because driving that outcome demands commercial judgement, financial modelling and investor negotiation from the same person.
References
- Michael Batko, co-founder and CEO of Hourglass, on the minimum structural test for a model, in Story Recruitment's panel on what investors look for in founder built financial models.
- Luke Rix, co-founder and CEO of KC Ventures, on lining a forecast up against historicals. From the same Story Recruitment panel on founder built financial models.
- Daniel Ross, advisor at Triple Bubble and Euphemia, on forecast horizon at early stage. From the same Story Recruitment panel on founder built financial models.
- Michael Batko on top-down market sizing as a credibility killer. From the same Story Recruitment panel on founder built financial models.
- Where the stage thresholds actually sit: after achieving profitability and experiencing growth, a company reaches a stage where it can genuinely consider one, and sustained profitability with growth is a relatively rare position for a business to be in.
- From a first CFO brief Tom Hunter wrote: the incoming CFO would be responsible for driving the company's future liquidity event, whether an IPO, acquisition, capital raise or exit, a role demanding strong commercial skills, financial modelling, forecasting and investor negotiation. The first-CFO specialism behind briefs like that one is discussed in an interview with Tom Hunter on the Honest Wealth Builders podcast.
- A situation Tom Hunter has worked on: a CFO of about five years had excelled in governance and financial controls and brought the business back to profitability, but a mutual decision was made to part ways because the skill set was not aligned with the next phase of growth.
- What I consistently see: when businesses move into new markets, finance often operates like a startup again, requiring a re-evaluation of processes, different hiring strategies and scaled reporting so that investors can trust them through rapid change.