AI tools are multiplying faster than most businesses can evaluate themIntuit Mailchimp’s Anthony Capano says the problem is not adoption. It is knowing what to skip.
What’s happening: Australian small businesses are being bombarded with AI tools, frameworks and adoption pressure from every direction. The conversation has focused almost entirely on what to adopt. The more useful question, according to practitioners and researchers, is what to ignore.
Why this matters: Knowing which AI to deploy, which to defer and which to skip entirely is becoming the skill that separates businesses that gain real efficiency from those that add complexity without return.
It has been a big week for AI announcements. NVIDIA’s GPU Technology Conference in San Jose unveiled a new generation of AI agents that run locally on business hardware, free from cloud subscriptions and token costs.
AI startups globally have already attracted over $220 billion in funding in the first three months of 2026, nearly matching the total raised across all of 2025, according to BestBrokers analysis of Crunchbase data. And the term SaaS-pocalypse has entered mainstream business vocabulary as AI agents begin automating tasks that software subscriptions once commanded.
For Australian small business owners trying to make sense of all of it, the noise level is genuinely overwhelming. And that noise is the problem.
COSBOA’s 2025 Small Business Perspectives Report shows that while around 30 per cent of small businesses use AI for day-to-day tasks, only 14 per cent have been able to integrate AI into core operations or services. Separately, Deloitte Access Economics research commissioned by Amazon found that while two thirds of Australian SMBs are using AI, just 5 per cent are fully enabled to realise its potential benefits.
The gap between using AI and actually benefiting from it is wide, and it is not closing as fast as the volume of available tools might suggest. Research developed by OpenAI in partnership with COSBOA suggests AI adoption could help small business productivity grow by 7.1 per cent in the next five years, outpacing productivity growth among larger corporations. The opportunity is real. But so is the risk of chasing it in the wrong direction.
Most businesses are stuck in the middle
Anthony Capano, Regional Director APAC at Intuit Mailchimp, works with mid-market businesses across Australia and New Zealand and says the pattern he sees most consistently is not ignorance of AI but fragmentation in how it is being pursued.
“Most teams are stuck in partial AI mode, running fragmented pilots, testing isolated tools and layering yet another platform into an already busy stack,” Capano says. “What’s missing is a clear roadmap that ties AI to measurable business outcomes.”
The data from Intuit Mailchimp’s own Marketing Equaliser report reinforces that picture. About half of mid-market businesses globally have marketing teams of fewer than ten people. A third of B2B marketers are juggling more than seven platforms in their stack. Adding more tools to that environment without a clear strategic rationale does not create efficiency. It creates complexity.
For small businesses, the pressure to keep up is compounded by the pace of announcement. Each week brings new tools, new frameworks and new claims about what AI will do next. Blackpearl Group CEO Nick Lissette captured the dynamic plainly in recent commentary, describing the current moment as an age of great uncertainty where something new will emerge every day and that volatility is the new normal.
Where AI actually earns its place
The business conversation around AI has been dominated by a single framing: how do we adopt more, faster? COSBOA has consistently advocated for practical support to help small businesses increase their AI capability, and the federal government’s National AI Plan reflects that direction. But adoption as an end in itself is not a strategy.
Capano argues the more useful starting point is identifying where AI will not make a meaningful difference and removing those options from consideration before they consume time, budget and attention. “Before implementing AI at scale, look across your organisation and campaigns to identify where AI can make the biggest difference,” he says. “Once you have a long list of needs, narrow it down to two or three high-impact business problems where improvement would meaningfully influence growth. That becomes your starting point.”
The implication is direct. Most of the AI tools currently being marketed to small businesses are not relevant to those two or three problems. Ignoring them is not falling behind. It is staying focused.
The Blackpearl CEO’s recent commentary on AI agents is instructive here. His argument was not that every business should rush to build or adopt AI agents. It was that the businesses best positioned to benefit are the ones with strong, structured, proprietary data foundations. Intelligence without knowledge, he argued, is noise.
The same principle applies at the small business level. AI tools perform best when they are applied to processes where the business already has clean data, clear workflows and measurable outcomes. Applied to messy, undefined or low-priority processes, they add noise rather than signal.
Capano points to email and SMS marketing as the category where AI creates the most immediate and measurable return for smaller businesses. “For mid-market brands, email and SMS are often the most controllable and cost-effective channels. This is where AI can have an immediate and measurable impact,” he says. AI can refine personalisation based on previous engagement, optimise send times and improve subject line performance. These are narrow, well-defined applications with clear metrics.
That is precisely the kind of AI worth adopting. Narrow scope. Clear baseline. Measurable outcome. Everything else can wait.
What to ignore and why
The NVIDIA GTC announcements this week are a useful illustration of the point. Local AI agents running on business hardware, fine-tuning open models for specific use cases and robotics systems for manufacturers are genuinely significant developments. But for most small businesses, they are not yet relevant. The hardware is not priced for SME budgets and the implementation complexity requires technical resources most small teams do not have.
The same applies to the broader AI agent category generating commentary around the SaaS-pocalypse narrative. AI agents automating narrow software functions are a real and accelerating trend. But for a small business using three or four software tools and not paying enterprise SaaS prices, the disruption to large-scale software vendors is not a near-term operational concern. It is background noise.
COSBOA, in collaboration with the Australian Cyber Security Centre, has also flagged that AI adoption introduces real data privacy and security risks that small businesses are often not equipped to manage, including the risk of uploading sensitive customer or staff information into AI platforms without proper governance in place. Adopting AI tools without understanding those risks is not a competitive advantage. It is a liability.
Capano’s framework for navigating the current environment comes back to a principle that applies well beyond marketing. Start with the real problem. Understand the baseline. Match the tool to the outcome. Measure the result. Scale what works.
“AI tools become more useful the more you use them. The goal is not to roll out AI everywhere all at once. It’s to start by proving value in priority areas, then scaling those wins across channels and segments,” he says.
COSBOA has made the same point from a policy perspective, noting that the implementation phase, not the plan itself, will determine whether small businesses can benefit fully from AI. Having access to tools is not the same as using them well. For Australian small business owners navigating an AI landscape that is expanding faster than anyone can fully track, the most useful reframe right now is this: the businesses that will gain a genuine edge in 2026 are not the ones that adopted the most AI tools. They are the ones that were disciplined enough to ignore the ones that did not matter. “ANZ’s mid-market marketers have already shown they’re willing to adopt early,” Capano says. “The next step is to turn that early momentum into a durable edge by balancing ambition with discipline, and making AI an equaliser in 2026 and beyond.”
The noise is not going away. Learning to tune it out is the skill worth developing.
Anthony Capano is Regional Director APAC at Intuit Mailchimp. Data cited from COSBOA’s 2025 Small Business Perspectives Report, Deloitte Access Economics research commissioned by Amazon, OpenAI and COSBOA research, Intuit Mailchimp’s Marketing Equaliser report, and BestBrokers analysis of Crunchbase data as of March 2026.
Keep up to date with our stories on LinkedIn, Twitter, Facebook and Instagram.
