Is the Data Lake the Corporate Excalibur?
Does the data warehouse still exist? The concept of the traditional data warehouse throws up a vision of batch data, painstakingly transformed into one fixed and slowly evolving enterprise data model, which only satisfied a relatively small number of requirements. The traditional bank relied heavily on the data warehouse, but, as banks evolve, so must the data platforms that service their needs for advanced analytics, digital, and AI. This next-generation data platform is known as the data lake.
The original concept of the data warehouse was devised by IBM which it called the ‘information warehouse’. The information warehouse was recommended for organizations so they could leverage their burgeoning data archives to help them gain business advantage over their competitors. However, the implementation of data warehousing required huge investments both in teams and infrastructure. Early claims of ROI were aggressive and only a few organizations, which very skilled internal teams, were able to reap these benefits – many projects failed.
One of the main reasons for project failure was data integration. Internal teams were tasked with developing interfaces to complex operational systems that they were unfamiliar with. The operational systems were often times poorly documented and understanding the meaning of the data was not a trivial process. Even when technical data integration was completed, changes in data semantics and operational schemas caused integrations to fail. Worse yet, due to misunderstanding the business context of the data, it was used incorrectly leading to inaccurate business decisions.
The data lake is a step-change in data platform capabilities for a bank. Batch data integrations are moved to real-time. Fixed data models are replaced with nimble data structures and formats. Relational databases give way to different data storage and processing techniques such as NoSQL, in-memory, and graph which are more appropriate for different digital and AI workloads. Also, given the dynamic nature of data lakes, they are an ideal fit for cloud deployment. Cloud platforms also allow for dynamic scaling of storage and compute, which wasn’t possible in traditional data warehouse architectures.
While we realize that the data lake can provide the underlying capabilities for the next generation digital bank, a better data platform itself is not enough. What banks truly require is banking specific data capabilities on top of the new platforms. They require real-time, productized integrations with the most valuable data systems in the bank such as core banking and payments systems. They require all the data to be secure, fully documented, and traceable to meet the ever stricter data regulations. And they require the capability to prepare and optimize data for AI, machine learning, and advanced analytics.
Fred Cook, Chief Information Officer at BlueShore Financial says: “Temenos Data Lake is mission critical for us in executing our unique strategy and differentiating our value proposition. The ability to integrate and combine many types of data is essential to gaining full business insights. The use of Temenos Data Lake and Analytics has enabled us to surgically execute our strategy, by defining customer micro-segments and tailoring our products and services specifically for these segments. This has led to an increase of 138% of assets under management over 8 years. Temenos Data Lake and the integrated data analytics and engineering tools will help us continue this aggressive growth strategy.”
Temenos Data Lake can harness real-time, structured and unstructured data as well as curate, ingest and blend large volumes of data at scale. With Temenos Data Lake banks can now implement a single, governed, data hub for their downstream systems be it analytics, AI, AML, Temenos analytics, Temenos reporting. This allows banks to be able to store and process all the data they need to power smarter applications from a single source and achieve higher performance at a very low TCO, as opposed to building data lakes in-house with the high development and integration risks that come with it. Banks can also power AI-based applications to tailor recommendations to each customer and offer contextualized advice.
If you want to easily retrieve your organization’s Excalibur from that vast expanse of data – use Temenos’ Data Lake.