DIRECTORY RESEARCH
Australian AI consulting landscape: 100-provider snapshot
This dataset describes the services that 100 consultancies publish for Australian buyers. It separates service coverage from quality and labels the evidence behind selected OpenAI, Anthropic / Claude and Snowflake relationships.
Coverage by comparison category
Categories overlap because one listing may offer several services. Percentages use all 100 published consultancies as the denominator.
| Category | Count | Share |
|---|---|---|
| Strategy & readiness | 78 | 78% |
| Implementation | 94 | 94% |
| Training & adoption | 33 | 33% |
| Governance | 42 | 42% |
Published selected-platform relationships
Partnership evidence can overlap. Counts combine vendor-directory entries and explicit operator claims; each profile labels its evidence type.
| Platform | Count | Share |
|---|---|---|
| OpenAI | 19 | 19% |
| Anthropic / Claude | 10 | 10% |
| Snowflake | 13 | 13% |
Evidence coverage
- 2 of 100 listings have a comparable published price or starting price recorded; missing prices remain unknown.
- 100 of 100 listings have a sourced, visually reviewed identifying image.
- The underlying profiles contain 143 operator or vendor source links.
- Listing evidence was checked from 2026-09-19 to 2026-09-19; the snapshot was compiled 2026-09-19.
Method
Each operator appears once. We include consultancies and independent practitioners with an Australian base or stated Australian delivery, and count a category only when an operator page publishes the relevant consulting service. Selected-platform relationships require either a vendor-directory entry or an explicit operator claim; model use or a displayed logo alone is not enough.
Limits
Published capability is not proof of delivery quality, current capacity or a particular team’s credentials. Global partner status may not apply to an Australian engagement. The inventory is selected rather than exhaustive, and most providers do not publish directly comparable project prices.
This page is generated from the same reviewed listing records used by the directory. Counts update with the published dataset; they are descriptive, not quality scores.