How to manage Market Data costs?

As the volume and number of data sources grows, firms are faced with a triple challenge: efficiently servicing the business by easily onboarding data sources, ensuring regulatory compliance on process scrutiny and, managing and controlling data costs – which often means navigating an array of complex commercial models, data licensing and usage agreements.

How Does Market Data Licensing Work?

Understanding and knowing which data license models are the most optimal and how they flex with evolving business needs requires transparency and controls on consumption and usage, backed by an accurate analytics model. Global spending on market data has reached $37.3B in 2023, according to research firm Burton Taylor.

License models are often complex and can be driven by many factors including direct display, use cases, volume, either bulk pricing or itemized, geographical boundaries, internal distribution boundaries etc. Data owners try to capture value – even going to AUM-based pricing.

Tracking Market Data Cost

To start to track market data cost, it helps to distinguish between different data categories. These include real-time data, evaluated prices, end-of-day market data, security master data (product terms and conditions), corporate actions data, ESG data, legal entity data, fundamental data, fund data (fund reference data, flows etc.), regulatory data (tax, sanctions, etc.) and alternative data (e.g. sentiment, geospatial information).

To properly track data costs, you can use a specific application from which we control costs incurred against budgets set. This can catch requests from different consuming application systems and pool them to prevent duplicate requests.

There are various levels of sophistication ranging from a manual process to track costs from different sources and reconcile against vendor invoices to using the pricing models from the data vendors to dynamically source and use publicly available data sources where available. Cost can then be tracked against different consumer levels: individual users, systems, departments, divisions, customers or geographies. Data budgets can be set by department or overall or broken down more granularly.

Price is one factor in selecting data sets but it is often difficult to do a 1-1 comparison, so the content, the ability to finetune content (e.g. by specific selections on identifiers and fields) and technical delivery options/sourcing and integration options are as least as important.

Optimizing Data Cost

Internal cost allocation and usage tracking can be done by tracking just the invoices from the vendors. However, the more granularly you can track how data is used internally and the more you identify areas of underuse or redundant sourcing, the better prepared you are to get the best sourcing strategy.

Capabilities that help with that include catching requests from different consuming applications systems and pooling them to prevent duplicate requests and using the pricing models from the data vendors to dynamically source. There is a lot of variation in how data can be retrieved. In some cases, you can source your data via itemized shopping lists using a “per security model” where you can specify the identifiers and the individual data fields needed, or you can dynamically interact with an API. With other providers, you need to use bulk data products.

Warehousing too much data leads to overly high IT costs. Warehousing in too many places leads to the need for duplicate work and an increase in operational costs. Warehousing in too many places leads to the need to reconcile different data sets and operational risk. Sourcing too much data via different entry points in the firm leads to overly high data costs and a lack of predictability in data consumption, causing cost surprises. Opaque data pricing models will lead to cost surprises and guarantee a lot of confusion down the road.

Firms can be triggered to reevaluate their provider landscape of market and reference data (management) providers through:

• Price increases of a specific vendor
• General need to rationalize provider landscape and/or control cost
• Business expansion into new products or markets
• New regulatory reporting requirements
• IT transformation

However, the more scattered your data sourcing and storage, the costlier it is to change or expand your set of data sources. The ability to quickly and cheaply onboard new data sources and new downstream consuming applications will greatly reduce the cost of change.

The impact of a Data-as-a-Service solution for market and reference data including cost metering and prevention of duplicate sourcing could lead to material cost improvement and a vast improvement in operational efficiency.