Blab Tech
CommBank iQ Integrates Snowflake AI to Unlock Core Banking Data Insights
At the Snowflake Summit in San Francisco, CommBank iQ – the joint venture between the Commonwealth Bank of Australia and data‑analytics firm Quantium – revealed that it has begun applying Snowflake’s newly released AI tools, CoWork and CoCo, to its core banking behavioural dataset. The announcement marks the first time the company can harness artificial intelligence directly against its most sensitive data asset.
The core banking dataset, a de‑identified collection of Commonwealth Bank transaction records, has long served as the backbone of CommBank iQ’s consumer‑spending insights. Until now, strict security, governance and risk controls—necessary because the data is classified as the bank’s “crown jewels”—had barred the use of external AI platforms. According to former chief executive Makenna Ralston, the organisation had been “convinced” that an all‑in AI strategy would unlock capacity and broaden the reach of its insights, but the data’s protection requirements ultimately blocked the technology.
Ralston explained that the team had been applying AI to all non‑core activities—strategy work, customer communications and other knowledge tasks—while the core analytics pipeline remained unchanged. The result was a bottleneck: analysts continued to follow lengthy workflows for client briefs, and the firm had to turn away new work because the team was at capacity.
The turning point arrived with Snowflake’s roadmap updates. CoWork, formerly Snowflake Intelligence, is an agent that can answer natural‑language queries against a dataset, while CoCo, formerly Cortex Code, translates natural‑language prompts into SQL, Python, dbt models, Airflow DAGs and machine‑learning pipelines. Ralston said CommBank iQ received early access to both tools and that the new capabilities “met our reality right when we needed it.”
The practical impact surfaced in a test with New South Wales’ tourism data. Within a month of deploying CoWork, a government client asked a question about an unusual spike in transaction growth. Using CoWork in the meeting room, the client lead received an answer in real time, linking the spike to school holidays and a new tourism business. The insight enabled the client to consider funding and partnership opportunities.
“This is a real actionable insight,” Ralston said. “A small question, a curious question that got answered in the room, not a week later.” The example illustrates how the AI tools can democratise access to the core dataset, allowing non‑technical users to query data directly and accelerate the delivery of insights.
Beyond client‑facing queries, CommBank iQ is developing “virtual analysts” on CoCo. These virtual analysts are designed to handle end‑to‑end analytical workflows—from scope definition to delivery—without human intervention. Ralston noted that analysts will focus on setting briefs, making judgment calls and maintaining client relationships, while the virtual analysts perform the routine analytical work. The first production deployment is expected later this month in a single team, with plans to scale.
The integration of Snowflake’s AI tools also addresses the firm’s capacity constraints. By automating routine tasks, the team can handle more client briefs without expanding staff. Ralston said the firm had previously turned away work because the team had “zero headroom.” The new AI capabilities should allow CommBank iQ to meet demand and pursue new opportunities.
Regulatory and governance implications remain a priority. The company continues to operate under a joint‑venture governance structure that includes strict risk and compliance controls. While the AI tools are now part of the core data pipeline, the firm maintains that it upholds the bank’s security posture.
Looking ahead, CommBank iQ plans to apply the CoWork and CoCo capabilities across its other industry verticals, ensuring that the AI‑enabled workflow is consistent across the business. The firm’s CEO, who has moved to London, said the company is “weeks, not months, away from reshaping how we work completely with this Snowflake capability.”
The move positions CommBank iQ at the forefront of AI‑driven analytics in the financial services sector, demonstrating how a large banking institution can safely integrate advanced AI into its most sensitive data assets while improving client service and operational efficiency.
The core banking dataset, a de‑identified collection of Commonwealth Bank transaction records, has long served as the backbone of CommBank iQ’s consumer‑spending insights. Until now, strict security, governance and risk controls—necessary because the data is classified as the bank’s “crown jewels”—had barred the use of external AI platforms. According to former chief executive Makenna Ralston, the organisation had been “convinced” that an all‑in AI strategy would unlock capacity and broaden the reach of its insights, but the data’s protection requirements ultimately blocked the technology.
Ralston explained that the team had been applying AI to all non‑core activities—strategy work, customer communications and other knowledge tasks—while the core analytics pipeline remained unchanged. The result was a bottleneck: analysts continued to follow lengthy workflows for client briefs, and the firm had to turn away new work because the team was at capacity.
The turning point arrived with Snowflake’s roadmap updates. CoWork, formerly Snowflake Intelligence, is an agent that can answer natural‑language queries against a dataset, while CoCo, formerly Cortex Code, translates natural‑language prompts into SQL, Python, dbt models, Airflow DAGs and machine‑learning pipelines. Ralston said CommBank iQ received early access to both tools and that the new capabilities “met our reality right when we needed it.”
The practical impact surfaced in a test with New South Wales’ tourism data. Within a month of deploying CoWork, a government client asked a question about an unusual spike in transaction growth. Using CoWork in the meeting room, the client lead received an answer in real time, linking the spike to school holidays and a new tourism business. The insight enabled the client to consider funding and partnership opportunities.
“This is a real actionable insight,” Ralston said. “A small question, a curious question that got answered in the room, not a week later.” The example illustrates how the AI tools can democratise access to the core dataset, allowing non‑technical users to query data directly and accelerate the delivery of insights.
Beyond client‑facing queries, CommBank iQ is developing “virtual analysts” on CoCo. These virtual analysts are designed to handle end‑to‑end analytical workflows—from scope definition to delivery—without human intervention. Ralston noted that analysts will focus on setting briefs, making judgment calls and maintaining client relationships, while the virtual analysts perform the routine analytical work. The first production deployment is expected later this month in a single team, with plans to scale.
The integration of Snowflake’s AI tools also addresses the firm’s capacity constraints. By automating routine tasks, the team can handle more client briefs without expanding staff. Ralston said the firm had previously turned away work because the team had “zero headroom.” The new AI capabilities should allow CommBank iQ to meet demand and pursue new opportunities.
Regulatory and governance implications remain a priority. The company continues to operate under a joint‑venture governance structure that includes strict risk and compliance controls. While the AI tools are now part of the core data pipeline, the firm maintains that it upholds the bank’s security posture.
Looking ahead, CommBank iQ plans to apply the CoWork and CoCo capabilities across its other industry verticals, ensuring that the AI‑enabled workflow is consistent across the business. The firm’s CEO, who has moved to London, said the company is “weeks, not months, away from reshaping how we work completely with this Snowflake capability.”
The move positions CommBank iQ at the forefront of AI‑driven analytics in the financial services sector, demonstrating how a large banking institution can safely integrate advanced AI into its most sensitive data assets while improving client service and operational efficiency.
