Data report - H1 2026
The state of SMB AI adoption in Australia.
Anonymised operator data on AI adoption across Australian small and medium businesses. Drawn from the XLev AI Readiness Scorecard plus client-engagement fieldwork. Refreshed quarterly. Free to cite and reuse under CC BY 4.0.
Methodology
How this report is assembled.
Two data sources. First: anonymised submissions to the XLev AI Readiness Scorecard - a free 14-question diagnostic that scores SMBs across strategy, tooling, capability and outcomes on a 0-100 scale with a tier. Second: operator-level fieldwork from XLev client engagements and our founder's own 80-staff SMB.
The sample skews toward businesses already considering AI seriously enough to take a structured assessment. The directional signal is reliable; the exact percentages reflect XLev's submission base, not the entire Australian SMB population.
Tier distribution
Where Australian SMBs sit on the AI readiness curve.
61% of respondents sit in the bottom two tiers (Reactive and Exploring). 3% have reached the Leading tier where AI is woven through the entire operation.
| Tier | Share | Description |
|---|---|---|
| Reactive | 28% | AI used occasionally by individuals; no company-level rollout, no governance. |
| Exploring | 33% | AI used regularly by some staff; early Project-style work; minimal governance. |
| Operational | 24% | AI rolled out across at least one function; governance signed off; measurable productivity uplift. |
| Embedded | 12% | Multiple functions; custom builds for high-leverage workflows; named AI lead. |
| Leading | 3% | AI woven through the operation; evidence pack ready for due diligence. |
Workflow adoption
Most-adopted AI workflows.
Share of respondents using each workflow at least weekly. Text-based knowledge work dominates; agent-level and voice work is still early.
- AI-assisted writing76%
- Information synthesis / research61%
- Internal knowledge / Q&A43%
- Customer communication drafting38%
- Code completion / engineering34%
- Image generation24%
- Structured extraction from documents19%
- Custom AI agents11%
- Voice agents8%
Adoption barriers
Top reported barriers to broader AI adoption.
Single most-cited barrier. The dominance of “not knowing where to start” over cost is the most striking finding - the barrier is direction, not budget.
- Not knowing where to start47%
- Governance, data and privacy concerns29%
- Cost11%
- Team capability / change-management bandwidth8%
- Other5%
FAQs
Methodology FAQs.
- Where does this data come from?
- Anonymised submissions to the XLev AI Readiness Scorecard, supplemented by operator-level fieldwork inside XLev client engagements and the XLev founder's own SMB.
- How representative is the sample?
- The sample skews toward businesses already considering AI seriously enough to take a structured assessment. The directional signal is reliable; the exact percentages reflect XLev's submission base, not the entire Australian SMB population.
- When is the next update?
- Quarterly refresh on the same methodology. H2 2026 update lands in November 2026, with subsequent quarterly snapshots in January, April, July and October.
- Can we use this data?
- Yes. Released under Creative Commons Attribution 4.0 International (CC BY 4.0). Cite as 'XLev - State of SMB AI Adoption in Australia, H1 2026' with a link back to https://xlev.ai/data/state-of-smb-ai.
Related reading
Long-form analysis of this data.
For the operator-essay analysis of these numbers, read the companion article.
Read the full analysis →