Zero-based budgeting funds every cost line from a fresh justification each cycle instead of rolling last year forward. It surfaces spending nobody would approve today. It is also slow, and in a growing company it only works if someone owns the argument with each budget holder rather than emailing out a spreadsheet.
How zero-based budgeting works
Start with the arithmetic, which is the easy half: take the money available for the period, list every cost the business wants to carry, and allocate against each one until the remaining balance is zero. Nothing is carried forward automatically. A line that existed last quarter starts the new quarter at nil and has to earn its funding on the strength of what it will do next, not on the fact it was approved before.
That is the whole difference from the incremental method most businesses actually run, where last year is the starting point and the conversation is about the increase. Neither method changes what you are required to record or lodge; the ATO's business records rules and business.gov.au's guidance on business accounting apply the same way whichever budgeting approach you pick. Zero-based budgeting is a management discipline, not an accounting treatment.
| Incremental budgeting | Zero-based budgeting | |
|---|---|---|
| Starting point | Last period's actuals, plus or minus a percentage. | Nil. Every line starts unfunded and has to be argued for on what it will do next period. |
| Software and subscriptions | Renewals roll over because the line already exists and nobody owns the review. | Each tool needs a named owner and a current use case, which is where dormant licences surface. |
| Headcount | Existing roles are assumed; the debate is about additions. | The whole team plan is rebuilt against next year's work, so misallocated roles show up as well as missing ones. |
| Effort | Days. It is largely a maths exercise on last year. | Weeks, and it consumes management attention, which is the real reason most businesses run it selectively. |
Building your first zero-based budget
Start with the money, not the costs. Fix the revenue or funding you can genuinely count on for the period, then list every outgoing at the level you can actually make a decision at: payroll by team, software by tool, contractors by engagement, not one line called operating expenses. Separate the costs you are contractually locked into from the ones that are a choice this quarter, because only the second group is really in scope.
Then work down the discretionary list and force a sentence out of the owner of each line about what it buys. This is where the method pays for itself. Subscription waste is the obvious one: paying for multiple software tools long after the employees who used them have left, platforms with no recent logins, idle licences quietly draining cash every month.[1] None of that would survive a fresh justification, and none of it gets caught by adding five percent to last year.
Keep going until allocations equal available funds. If they do not balance, the choice is explicit: cut a line, or accept that you are funding it from reserves and say so out loud.
If the answer is reserves, size that deliberately. Here is how much cash reserve a business actually needs and where to hold it.
Templates, spreadsheets and tools
For a first pass a spreadsheet is fine and usually better. The value of the method is in the conversation with each budget holder, and a sheet you built yourself forces you to see the structure of your own cost base. Dedicated budgeting and planning software earns its place later, when you have enough cost centres that consolidating them by hand is the thing eating the month.
Do not expect the tool to do the judgement. AI can handle 80 to 90 percent of basic finance tasks such as checking, reconciliation, analysis, forecasting or modelling, but the last 10 percent needs a qualified human who understands what the challenge is and what good looks like as an outcome.[2] A zero-based budget is almost entirely that last 10 percent, because the output is a set of decisions about what the business will stop doing.
What it costs you
The honest downside is time and goodwill. Rebuilding from zero every cycle is slow, it asks every manager to defend spending they already consider settled, and done badly it turns into an annual ritual that produces the same answer as last year with more meetings. Most Australian scale-ups I work with run it selectively: full zero-based treatment on the discretionary cost base, incremental on the contracted lines.
The discipline problem is real and it does not disappear at scale. Jorrick Chivers described managing spend under a soft salary cap in sport, where the temptation is to overspend incrementally and it takes discipline to pump the brakes and walk away from a deal rather than justify stretching the limit one more time.[3] That is the same failure a zero-based budget is built to prevent.
Who owns it as you scale
A zero-based budget is only as good as the person willing to have the conversation. Founder-led finance does not scale at all, and it eventually runs its course as the business grows, leaving a scrambled financial model and cash conversations that are more of a guess.[4] A founder can chase down idle licences once. They will not do it four quarters running while also running the company.
Where it lands depends on which finance role you actually have. The biggest difference between a financial controller and a head of finance is that the controller focuses on controls, compliance and reporting, while the head of finance is broader and covers the forward-looking work: financial modelling, FP&A, budgeting and forecasting.[5] Zero-based budgeting sits squarely in the second group. Hire a controller and hand them this, and you will get a tidy spreadsheet and no reallocation.
The upside when it is owned properly is not subtle. Changes put in by one finance manager we placed are now saving an international SaaS business an additional $20,000 to $30,000 a month.[6] That is one person, looking hard at a cost base nobody had rebuilt from zero in years.
The sequencing question underneath this is how a startup finance team should be structured as it grows.
Common questions
What is zero-based budgeting?
Zero-based budgeting is a method where every cost line starts each cycle at nil and has to be justified again before it is funded, rather than being carried forward from last period with an adjustment. You keep allocating available funds against justified costs until the balance reaches zero and nothing is left unassigned. It is a management discipline rather than an accounting treatment, so it does not change what you record or lodge.
How is zero-based budgeting different from normal budgeting?
Normal, or incremental, budgeting takes last period as the starting point and argues about the change. Zero-based budgeting deletes the starting point. The practical consequence is that incremental budgeting protects existing spending by default and zero-based budgeting exposes it, which is why dormant software licences, stale contractor arrangements and roles that no longer match the work tend to surface only under the second method.
Is zero-based budgeting worth the effort for a startup?
Selectively, yes. Running it across the entire cost base every quarter consumes more management attention than most scale-ups can spare. The version that works is full zero-based treatment on the discretionary cost base, where a real decision exists, and incremental treatment on contracted and statutory lines where it does not. The first pass is usually the one that pays, because it clears years of accumulated spend nobody would approve today.
Who should own the budget in a growing company?
Whoever owns the forward-looking finance work, which in a scale-up means a head of finance or a CFO rather than a financial controller. A controller is focused on controls, compliance and reporting; the budget is a planning and reallocation exercise that requires arguing with each budget holder about what their spend buys. Founders can run it once, but founder-led finance does not scale, and the discipline lapses first.
References
- What I consistently see: common subscription waste includes paying for multiple SaaS tools long after employees have left, platforms with no recent logins, and idle licences draining cash monthly.
- Tom Hunter, interviewed on the limits of AI in a finance function: AI can handle 80-90% of basic finance tasks like checking, reconciliation, analysis, forecasting or modelling, but the last 10% requires a qualified human who understands what the challenge is and what good looks like for the outcome.
- Jorrick Chivers on The CFO Track, on managing spending under a soft salary cap in sport: the temptation is to incrementally overspend, and it takes discipline to pump the brakes and walk away from deals when limits are stretched too far, rather than justifying the spend.
- My position on it: it does not scale at all, and eventually runs its course as a business grows, leading to a scrambled financial model and cash conversations that are more of a guess.
- Tom Hunter, interviewed on how finance roles differ: the biggest difference between a Financial Controller and a Head of Finance is that a Financial Controller focuses more on financial controls, compliance and reporting, while a Head of Finance is broader, covering forward-looking tasks like financial modelling, FP&A, budgeting and forecasting.
- From a placement I worked on: changes implemented by a Finance Manager placed by Story Recruitment are now saving an international SaaS business an additional $20-30k per month.