Ten years in the for-purpose sector: what's actually changed
The model the last decade was built on is running out of road — and the AI wave is exposing exactly how thinly capitalised the sector has been on the things that drive productivity.
In 2016, I was a year into the for-purpose sector. The ACNC's first Australian Charities Report had just landed: a $142.8 billion sector built on government funding, donations, and a long tail of small organisations doing extraordinary work with very little.
The 12th edition was released a few weeks ago. The sector now turns over $239 billion. Revenue grew 7.5% in a year. We employ 1.6 million Australians — 11% of the workforce — and mobilise 3.9 million volunteers.
The headlines look like growth. Spend a decade inside the sector and you see something else: the model the last ten years was built on is running out of road, and the AI wave is exposing exactly how thinly capitalised most of the sector has been on the things that actually drive productivity.
Ten years in, three shifts are the ones I'd put in front of any board right now.
1. The single-payer model is breaking — fee-for-service revenue is now a survival strategy
Government funding climbed from 43% to 49.4% of total sector revenue between the first and 12th ACNC reports. Fee-for-service revenue went the other way: from 49.7% to 34.6%. In dollar terms, around $118 billion of the sector's revenue now comes from government. Expenses grew 8.6% in a year — faster than revenue. And 0.6% of charities now generate 57% of total revenue. A sector this concentrated on one payer is a sector with very little strategic optionality.
The fiscal backdrop is the part most boards are still underweighting. CEDA's analysis of the 2026–27 Budget has net debt rising from $556 billion (18.8% of GDP) in 2025–26 to a forecast peak of $767.8 billion (21.9% of GDP) by 2029–30, with gross debt heading past $1.05 trillion by June 2027. The Budget delivered $63.8 billion in savings, and the largest single saving was the NDIS — growth slashed from around 10% per year to 2%, removing roughly 160,000 people from projected participation and stripping $37.8 billion from the forward estimates, with longer-term reductions of up to $184.9 billion over the next decade.
Read that as the signal it is. Federal government is now in a structural position where growth in social spending has to be contained, not expanded. State governments face the same pressure. The era of "if we deliver, the contract grows" is over for most of the sector. Government will remain a major customer; it will not be a growing one.
That is why the fee-for-service revenue story is the most important strategic story in the sector right now. Social Traders' FY25 results hit $304 million in spend with certified social enterprises — a record, up 18% year-on-year, and $1.4 billion cumulatively over eight years. Powered by Trade, Driven by Purpose (Social Traders and PRF, November 2025) shows 84% of social enterprise revenue now comes from trade, and Commonwealth social procurement at scale could deliver 44,000 jobs and $4.27 billion in savings by 2030. Minderoo's Unlocking Generosity (Feb 2026) estimates $7–12 billion in additional individual giving could be unlocked by 2030, and Philanthropy Australia is tracking the $2.6 trillion intergenerational wealth transfer starting to move through the economy now.
Fee-for-service, social procurement, mission-aligned trade, structured major giving and impact investment are no longer the "innovation portfolio." They are the second and third revenue lines that determine whether organisations are still here in their current form in 2035.
2. The productivity mandate is real — doing more with less is the operating model
If government can't fund growth and fee-for-service revenue takes years to build, the maths only closes one way: the sector has to get materially more productive at delivering its existing mission. That is the strategic case for AI and digital, and it is a very different case to "innovation as a side project."
The size of the prize is already visible. Infoxchange's 2025 Digital Technology in the Not-for-Profit Sector report shows cloud adoption has gone from 24% in 2015 to 87% in 2025, and 67% of NFPs now use generative AI. Cycle time on grant writing, case notes, board papers, donor segmentation, service triage and back-office reporting is collapsing in the organisations using these tools well. The productivity dividend is real — the question is who captures it.
The honest answer right now: very few organisations are converting AI adoption into measurable performance gains. Glean's Work AI Index (June 2026) found 90% of Australian digital workers are using AI at work, but only 10% said it had significantly improved organisational performance. Microsoft's 2026 Work Trend Index tells the same story from a different angle: 63% of Australians say they're producing work they couldn't have a year ago, but only 28% say their organisation has a clear AI strategy and policy. AI already automates an estimated 27% of digital workers' output and is forecast to hit 34% within a year.
The productivity gap is not a technology gap. It is a leadership, redesign and discipline gap. The organisations that will do more with less in the next five years are the ones treating AI as an operating-model change, not a tool rollout: deciding which processes to redesign, which roles to redefine, where human judgement is the differentiator, and how to govern the outputs. Everyone else is paying for licences while shipping work they can't fully explain.
3. The AI wave has exposed a decade of underinvestment in capability
This is the change I think matters most, and the one most boards still haven't internalised. The reason the sector is struggling to convert AI into productivity is the same reason it has struggled to convert almost every wave of technology into productivity: it has been chronically under-capitalised on the things that actually compound — leadership, systems, data, evaluation, governance and digital literacy.
The Paying What It Takes report (Centre for Social Impact, SVA, Philanthropy Australia) named the dynamic years ago: most funders won't fund indirect costs, so charities chronically underinvest in the very things that make them effective. The ABS now puts private non-profit R&D spend at just $1.69 billion — 0.06% of GDP, unchanged in real terms. Infoxchange's data confirms the operating-layer consequence: 67% of NFPs use AI, only 14% have an AI policy, only 23% have a cyber security plan. The technology arrived; the capability to govern and exploit it didn't.
The funder shift is the most important thing happening in the sector right now. Paul Ramsay Foundation's 2025 Annual Review reports $320M in grants in FY25, $28M in impact investments, and $1.476B distributed since 2016 across 175 partners. The structural move is Outcome 4: "Conditions support thriving, and capacity and capability is enhanced across the for-purpose sector." The investment packages underneath it — Democratising Data and AI, Capability for Systems Change — treat sector capability as core infrastructure, not overhead. PRF's Endowment Impact Fund has now committed 17.2% of its corpus and leveraged $2 billion alongside it: about $10.90 of co-investment for every $1 committed.
The wider market is moving with it. The CSI / Impact Investing Australia benchmark puts the Australian impact investing market at $157 billion — eight times its 2020 size. The Productivity Commission's Future Foundations for Giving and the NFP Sector Development Blueprint both reach the same conclusion: the long-term resilience of the sector depends on capability, data and trust — not just service contracts. Capability used to be the line item that got cut. The current generation of funders, frameworks and policy is starting to treat it as the precondition for everything else.
How to position for the next 10 years
The growth numbers in the new ACNC report are real. They are also a distraction. The next decade will reward a different set of decisions to the last one, and the 2026 data points clearly at what they are.
For senior for-purpose decision-makers, three moves matter now.
1. Engineer fee-for-service revenue as a board-level capital allocation decision. A 49% government-funded sector inside a federal balance sheet heading past $1 trillion in gross debt is structurally exposed. The next decade belongs to organisations that deliberately build a second and third revenue line — trade, social procurement, fee-for-service, mission-aligned investment, structured major giving. This is not a fundraising team problem; it is a capital allocation decision. The Social Traders, Minderoo and Productivity Commission numbers all point to where the new revenue is sitting. Designing for it takes 18–36 months. Starting in 2027 is starting late.
2. Treat AI and digital as your productivity strategy, not your innovation strategy. "More with less" is now the operating mandate of the sector. Sixty-seven per cent of NFPs already have AI in the building; only a fraction are turning it into measurable performance gains. The competitive edge in the next five years will come from boards that can answer four questions confidently: which processes are we redesigning around AI, which roles are we redefining, what data is AI touching, and how are we governing the outputs? Productivity will not come from another seat licence. It will come from operating-model discipline.
3. Self-fund the capability gap before the next wave makes it bigger. The AI wave has exposed how thinly capitalised most of the sector is on leadership literacy, data foundations, evaluation, cyber and governance. The funder shift toward capability is real, but it will reward organisations that already have the infrastructure to absorb the investment well. Boards should be asking, every year, what proportion of unrestricted revenue is being reinvested in the organisation's own capability — and whether the answer is closer to a top-quartile professional services firm than a top-quartile charity. If the sector doesn't fund its own capability, the next wave — agentic AI, payments-by-outcome, real-time data sharing — will expose the same gap, only wider.
The organisations that will matter in 2035 are not the biggest ones in the 2026 report. They are the ones whose 2026 board minutes show those three conversations already on the agenda.
This article was first published on Linkedin on 1 July 2026, here.
