Your AI Makes You Productive. It Doesn't Make You Profitable.

NAB's July 2026 Business Pulse found a pattern that should make every AI-using business owner pause. Among the 42 per cent of Australian SMEs now using AI, 58 per cent report productivity gains. Only 9 per cent report profitability gains. Seven per cent report revenue increases.

That is a six-to-one ratio of faster to richer. If AI were hitting the bottom line the way the productivity numbers suggest it should, those figures would be closer together. They are not. And the gap points to something specific about how most Australian SMEs are deploying these tools.

The NAB data maps where SMEs are applying AI: 51 per cent in marketing and sales, 39 per cent in operations and logistics, 25 per cent in customer service. Marketing is the standout adoption category — generating social posts, drafting email copy, creating ad content. Productivity? Absolutely. You can produce a week of LinkedIn posts in twenty minutes.

But marketing copy does not hit the P&L the way a faster quoting workflow or a tighter billing cycle does. The 9 per cent seeing profitability gains are embedding AI in the revenue chain — in the quoting, scheduling, invoicing, and collections workflows where time literally equals money.

For a trades business billing at $85 an hour loaded, shaving 30 minutes off every quote does not just feel productive — it frees a half-hour of billable capacity. That is $42.50 recovered per quote. Ten quotes a week, fifty weeks a year: $21,250 in recovered capacity. For an accounting firm, the same logic applies. We wrote recently about firms closing the books seven and a half days faster with AI. Faster close means less WIP, faster billing, and better realisation rates. That is a profitability story, not a productivity story.

Only 22 per cent of AI-using SMEs in the NAB survey describe results as transformational. The other 78 per cent see incremental benefits. A separate analysis from Deloitte Australia, reported by eCommerceNews in May 2026, found only 12 per cent of organisations say AI is genuinely transforming their business. The core finding: staff use AI to draft emails and text, but most businesses have not changed the underlying workflows that drive sales, reporting, and administration.

This is what separates the 9 per cent from the 58 per cent. Productivity is what happens when you bolt AI onto an existing process — you do the same thing faster. Profitability is what happens when you change the process itself, when AI handles an entire workflow end to end rather than one step in the middle.

A trades business that uses ChatGPT to write marketing emails is productive. A trades business that uses AI to handle inbound calls, schedule jobs by technician location, and generate quotes from job photos is profitable. We wrote about how missed calls alone cost the average trades business $66,000 a year. That is a revenue capture problem, not a content creation problem. The AI that solves it lives in the revenue chain, not in the marketing stack.

NAB's data also shows workforce readiness at 3.8 out of 10. That number reinforces a point we covered recently: productivity collapses after three AI tools. More tools do not close the gap. A better-placed tool does.

Pick one workflow that sits on your revenue chain. Not marketing. Not internal communications. The workflow where time delays directly cost you money — quoting, scheduling, invoicing, or collections. Embed AI there. Measure the output in dollars, not hours saved. If it does not move the P&L within 90 days, you have picked the wrong workflow.

The 9 per cent are not smarter or better-resourced. They aimed AI at the cash register instead of the noticeboard.

Key takeaways

Related notes

Want to know which workflow would hit your P&L first?

A 30-minute call can identify the highest-value AI deployment point in your business — the one workflow change most likely to move profitability, not just productivity. Book a call to find out.