Treasury Operations Stuck in Spreadsheets as Agentic AI Offers 90% Forecast Accuracy: EY India

Treasury Teams Trapped in Manual Work

Treasury operations remain stuck in outdated processes, with teams spending 60%-70% of their time on manual tasks and low-value activities, according to a new EY India report titled 'An Agentic AI Adoption Playbook for CFOs and Treasurers'1

1

2

2

. This leaves little bandwidth for strategic work that could drive real business value. Spreadsheet-driven treasury operations continue to dominate despite significant investments in treasury technology, with mature treasury functions managing between 50-100 interconnected spreadsheets covering cash positions, foreign exchange exposure, investments and regulatory reporting1

1

. More than 50% of corporates globally still rely on manual reconciliation, creating inefficiencies and increasing operational risks2

2

.

Forecast Variances Impact Liquidity Management

The reliance on spreadsheets carries a steep cost. Forecast variances in spreadsheet-led treasury environments often exceed 20%, making it harder for companies to manage liquidity effectively1

1

2

2

. This level of inaccuracy undermines the ability of CFOs and treasurers to make timely decisions around cash positioning and liquidity. The fragmented systems and legacy processes that characterize most treasury operations create blind spots that prevent organizations from gaining the visibility and control they need.

Agentic AI Delivers 90% Forecast Accuracy

Agentic AI presents a path forward. AI-enabled treasury models can improve cash forecasting accuracy to as much as 90% across 30-, 60- and 90-day liquidity horizons, allowing companies to make faster and more informed decisions1

1

2

2

. Hemal Shah, Partner at Risk Consulting, EY India, explained that agentic AI can help move treasury operations from a reactive function to a predictive and intelligent operating model2

2

. However, he cautioned that realizing this potential requires strong data foundations, robust governance frameworks and clearly defined workflows.

High-Impact Use Cases for Early Adoption

The report identifies cash forecasting, reconciliation and KYC/AML exception handling as the most promising starting points for agentic AI adoption1

1

2

2

. Cash forecasting offers the greatest potential business impact, while AI agents could handle 70%-80% of routine KYC/AML exception cases with full auditability1

1

. This would enable treasury and risk teams to focus on higher-value activities and more complex tasks that require human judgment.

Workflow Automation Shows Early Results

Workflow transformation must come first. Organizations that have already introduced digital breaks and workflow automation are seeing 80%-90% auto-match rates in reconciliation processes, according to EY India analysis1

1

2

2

. The report makes clear that workflow automation is the critical first step in successful agentic AI adoption, not an afterthought. One of the most common reasons AI initiatives fail to scale in treasury functions is the absence of a reliable and governed data architecture2

2

.

Building the Data Foundation

Companies need to fix their data architecture before deploying agentic AI. EY recommends building a treasury data lake that acts as a single source of truth by bringing together structured and unstructured data from ERP systems, banking platforms, contracts, emails and market information1

1

2

2

. The report also advocates establishing a Treasury Center of Excellence to oversee data lake pipelines, workflow libraries and governance frameworks1

1

2

2

. Organizations that combine workflow automation, trusted data foundations and strong governance will be best positioned to build more resilient treasury operations and realize the benefits of agentic AI.