Nearly two-thirds (63%) of data scientists in financial services firms say their organization is not currently able to combine data and analytics in a single environment, according to new research from Alveo, a provider of managed data services for data mastering and analytics.

The survey also found that nearly four out of 10 respondents (38%) experienced “the need to integrate structured and unstructured data” as one of the main challenges their organization faces in “bringing analytics to data and using the combination to drive effective decision-making.”

For financial services firms, closing the gap between data mastering and analytics capabilities is key in deriving insights from an increasingly broad range of data sources. In a financial context, structured data adheres to a predefined data model and includes everything from financial instrument terms and conditions to pricing feeds, while unstructured or semi-structured data does not conform to a pre-set data model and might incorporate earnings call transcripts and social media activity. It also can help gauge scores against environmental, social and corporate governance (ESG) indicators.

Another key challenge highlighted by the research was the issue of ever-expanding data volumes. According to the survey, 39% of data scientists claim “it is difficult for us to manage large data files and scale our infrastructure to the volumes we face” as the main challenge in bringing analytics to data.

“Financial services firms struggle with growing data volumes that are often siloed in data stores and legacy systems, making access difficult,” Mark Hepsworth, CEO of Asset Control, said. “This causes a bottleneck when firms look to get a broader range of data-to-data scientists and decision makers, creating a range of challenges as a result, including lack of integration of meaningful data and analytics.”

In line with this, Alveo’s research shows that many financial services firms across the UK, the U.S. and Asia still struggle with significant issues in integrating different types of data and scaling infrastructure to cope with ever-increasing data volumes.

Efforts to bring analytics and data together also are hampered by inadequate data lineage and poor data quality. Nearly a third of firms (32%) don’t have full transparency of data lineage across their organizations and the wider operational ecosystem, while just under a quarter (22%) list “lack of contextual information such as data lineage that helps us trust the data” among their main data management challenges. This can lead to a lack of trust and redundant data sourcing and verification.

In the search for data quality, a lack of a data catalogue leading to time-consuming data searches or double sourcing is the top issue, referenced by 28% of respondents, followed by the fact that “a proactive focus on data quality is hampered by need for ad hoc incident resolution,” which was identified by 23% of respondents.

“There are tools now available that can help firms identify data quality issues and proactively address them. More specifically, increasingly, there are solutions that enable businesses to explain the value and origin of data, trace data back to its external sources and ensure data lineage,” Hepsworth said. “And thanks to the latest advances in cloud, data processing and analytics, it is now possible to combine analytical and data management capabilities and use the results to maximize market data ROI and enable data scientists and other business users.”