Fast Multidimensional Financial Analysis for Retail Banking

About the Customer

A leading North American bank operating at global scale

One of the largest banking institutions in Canada, managing over US $1.5 trillion in total assets and serving more than 12 million customers across retail, commercial and wealth-management businesses. With a history spanning over 200 years, the organization operates a broad network of branches, digital platforms and financial-services teams across Canada, the United States and international markets.

Challenges

Legacy architecture preventing cloud-scale financial analysis

One of the world’s largest retail banks struggled with a 5-hour adjustment process related to periodic earnings reports on Essbase.  They had tried to move the corresponding analyses into Redshift.  However, Redshift could not provide interactive levels of performance with Power BI, or support Excel connections without painful data extracts. This resulted in:

Severe performance limitations: Power BI was not performant when pointed directly at Redshift. Performance was too slow for analysts to run their business. A lot of work had gone into getting data into Redshift, but the bank still couldn’t make use of that data at the required speed.

Unsustainable Power BI imports: When direct-connect performance wasn’t acceptable, teams attempted Power BI imports. Two issues quickly became apparent:

Building and managing imports for specific queries created too much overhead and technical debt for data engineering.
Imports had to be aggregated to such a high level that they were losing valuable detail.

Constraints on multidimensional financial workflows: Finance teams run complex, ad hoc, multidimensional queries daily. Without a robust semantic model, they were unable to generate enterprise views of the bank’s financials at the granularity required.

Inability to scale performance for 100s of analysts: Existing stack could not deliver fast, consistent performance across large user groups, limiting the bank’s ability to expand financial reporting use cases.

Buisness Goals

Building a modern, high-performance foundation for enterprise financial analytics

The bank had already invested significantly to get their financial data into Redshift and they needed a way to actually make use of their existing stack while finding a modern alternative to Essbase. They aimed to:

Achieve fast, reliable performance for Power BI and Excel on Redshift.

Eliminate dependence on imports and reduce technical overhead.

Support advanced, multidimensional financial hierarchies.

Enable deeper-grain YoY and end-of-period analysis.

Scalability to handle hundreds of users across the enterprise.

How Kyvos Helped

Delivered the high-speed, multidimensional financial analytics they needed

The bank wanted to simplify and accelerate queries. They had looked at a dozen options before finding Kyvos. However, no other solution supported the complex data hierarchies found within financial analysis. Only Kyvos semantic layer met their benchmark, enabling fast, detailed financial analysis.

Reduced a 5-hour Essbase adjustment process to 2 seconds: Kyvos accelerated critical periodic earnings adjustment workflows from hours to seconds, with both Power BI and Excel. It natively supports the complex data hierarchies required for financial analysis. Kyvos also supports writeback to a source system as a part of an adjustment process, keeping both the system of analysis and system of record in synch.

Cloud-native semantic models: Kyvos provided multidimensional semantic models that finance teams could rely on. Automated processing and scalable data models reduced runtime by nearly half. The models were easier to build and manage.

Advanced hierarchy support: Supported highly specific, multi-level hierarchies used by the CFO’s office. Alternate rollups and complex reporting structures were handled with ease.

Faster time intelligence analysis: Enabled year-over-year comparisons at much deeper grain levels, including deposit and loan-level detail. Also, support for fast adjustments for end of period scenarios created new visibility into performance patterns that finance leaders previously could not access.

Enterprise-wide scalability: Delivered consistent performance and ease of use required to support hundreds of concurrent users across teams.

Impact

Accelerated performance and efficiency across financial analytics

2-second financial adjustments, down from 5 hours

Deep-grain YoY financial analysis enabled

100s of users enabled

We evaluated 12 other platforms (on advanced hierarchy support).  I don’t think anyone (outside of Kyvos) even understood what we were asking.

Financial Reporting Lead, one of the world’s largest retail banks

We have a CAGR report that used to take us 1 full day to work on, with Kyvos we have accurate numbers and the report is ready in 5-6 mins. Internally, we are having a hard time believing this.

Senior Manager, Financial Reporting & Analytics