By Stephen Mulholland, RVP EMEA, Fivetran
The financial services sector has made impressive strides in implementing AI in recent years. Indeed, according to recent research by Fivetran, 51 percent of FinServs have already fully implemented generative AI, far surpassing other industries.
This trend underscores the sector’s recognition of AI’s capacity to surface vital insights at unprecedented speed – augmenting human workers’ ability to drive operational excellence and innovation. However, realising AI’s full potential and maintaining a competitive edge demands concerted efforts.
AI hopes and hurdles
There is great optimism surrounding AI in the finance sector, and numerous use cases already demonstrate the value it can bring to operations. For example, wealth management firms are harnessing generative AI to tailor investment recommendations and enhance customer satisfaction and loyalty, while insurance companies are expediting claims triaging with AI.
Despite its relatively recent emergence, generative AI is also well-trusted, with nearly eight in 10 organisations finding its outputs credible. It is then no surprise that organisations are investing, on average, 13 percent of their global annual revenue into AI programmes.
Yet financial services organisations still face substantial challenges in maximising the opportunities offered by AI technology. Fivetran’s survey highlighted that data quality issues and inefficient utilisation of data scientist resources are two major factors hindering progress. In fact, financial services organisations are losing, on average six percent of their global annual revenues due to underperforming AI programmes running on bad data.
The root cause of AI setbacks
To uncover the source of the problem, a more nuanced approach is required. Fivetran’s survey shows that people in technical job roles are much less trusting of AI outputs than those in non-technical roles, and the reason could well be because they see, more closely, how inefficient processes introduce challenges in the path of data. For example, over half of FinServs see their outdated IT systems as the main barrier to adoption.
When financial services organisations rely on legacy technology and bad data processes, more than just AI potential suffers. Inaccessible, siloed data creates delays, leading to out-of-date insights, unreliable analytics outputs, compliance challenges and missed opportunities. Since the vast majority of organisations still lean on manual practices, it’s sadly no surprise that data scientists are spending 67 percent of their time preparing data instead of actually building models. Put simply, organisations are leaving money on the table and placing unnecessary burdens on their data talent.
The crucial role of automating data flows
With the opportunity cost of bad data practices so high, financial services organisations that have not yet strengthened their data movement foundations must now do so. And unlike AI programmes that require human intervention, data movement can be fully automated today – bringing a host of benefits to financial services organisations.
Automating data movement enables organisations to seamlessly and securely centralise disparate data sources from all corners of the organisation, creating a single source of truth in the data. With this in place, teams can conduct analysis using the freshest data and feed it to AI models with the reassurance that outputs will be reliable.
Organisations replacing their outdated IT infrastructures with a modern data stack can leverage this automation to super-charge their decision-making and ensure that they retain full visibility and control over their data pipelines. With more regulation expected to be introduced around the use of AI, the ability to uphold and demonstrate good data governance practices will be paramount – particularly in the financial sector.
Laying the groundwork for innovation
Success with AI depends on organisations’ ability to create unfettered access to clean, fresh and reliable data. If financial services organisations want to remain at the forefront of AI adoption, they must ensure their programmes are robust enough to hold up against increased data volumes and the scrutiny of regulators.
By embracing automation and empowering data workers to focus on more value-added jobs, FinServs will not only see greater returns on their AI investments, but also better collaboration and innovation across all areas of the organisation.
