Speed. Agility. Competitive advantage. We could be talking about the Winter Olympic games – but we’re not. These terms describe the current business landscape, where technological advances kicked into overdrive in 2020 and have yet to slow down. More than ever before, organizations are being pressured to innovate by launching new products powered by data (both internally for business intelligence and externally for customer acquisition and user experience) in order to retain their competitive advantage.
In the past, launching a new data-driven product was challenging for several reasons. First, allocating the necessary IT resources to scope and build such products can discourage teams from even starting. Second, the nuances of identifying data assets, testing them for value, then managing ETL workflows across siloed databases can take months to years. Finally, identifying the necessary budget for data product development can be difficult to justify among competing priorities.
The Rise of Low-Code Development Platforms (LCDPs)
A significant influence in today’s race to produce at the speed of ideation is the rise of low-code development platforms (LCDP); those often utilized by in-house professionals with varying levels of technical expertise. LCDPs allow entities to create and release applications in a fraction of the time the same process takes by traditional methods. LCDPs are also typically cloud-based, have pre-built and customizable applications, and – true to their name – require little-to-no coding to produce functional insights.
LCDPs enable shorter development life cycles at a time when speed to production and business agility are more important than ever. In fact, a recent report says that by 2025, 70% of new applications developed by enterprises will use low-code or no-code technologies – almost triple the percentage from 2020.
Demand for Data through LCDPs
Underpinning most low-code applications are a combination of internal and/or external datasets that provide decision-makers with valuable insights. For example, in the retail industry, the ability to launch low-code dashboards to monitor footfall traffic has enabled targeted decision-making around store placement and advertising. During the Covid-19 pandemic, these dashboards were used by governments to monitor social distancing metrics.
In the financial services sector, there are low-code applications for every use case from minimum question digital applications to customer profile dashboards to due diligence (KYC, AML, KYB) dashboards. Across all of these applications, the ability to enrich internal data with external data in real-time is critical.
Additionally, when the business landscape turned upside down in 2020 and 2021, financial service providers launched portfolio monitoring tools powered by external data. Relying solely on internal data couldn’t come close to gauging the dramatic economic changes that lay outside the bounds of existing statistical models. Financial institutions needed to integrate more rapidly refreshing alternative data to monitor portfolio behavior and the likelihood of delinquency.
As recently as 2019, one survey found that 92% of data analytics professionals were looking to increase their company’s use of external data sources. And that was pre-pandemic. By 2020, leveraging data and analytics ecosystems enabled by an augmented approach was listed as a top trend by leading analysts in the space.
From a strategic standpoint, LCDPs offer easier access to the myriad of external data sources available. Given that companies who incorporate analytics into decision-making are reportedly four times more likely to use multiple data sources – mobile, social and public to name a few – accessing actionable data becomes a business imperative. Low-code methods used to access external data can accelerate its return-on-investment, but that depends on several factors:
Data Curation
There is now a vast number of external data providers for any given insight or data type. As in any industry, not all data providers are equal and there can be substantial differences across datasets as straightforward as tax assessor data (a relatively commoditized dataset that aggregates property information from county offices) or business firmographics (try comparing business revenue data across three different datasets and you’re likely to get three different answers). LCDPs are in a unique position to help their users discover the right external data sources to meet their needs using graphical interfaces (filter sources by a type or search by an attribute), and in certain cases helping them to identify dataset coverage against populations that are relevant to their use case.
Data Compliance
In order to inform accurate analysis and the right business decisions, external data must be heavily vetted. An LCDP may organize a robust influx of data, but companies need to know the specifics around how that data was first gathered by the third-party sources providing it. Any concerns around privacy, ethical collection, biased methodologies or accuracy need to be addressed before the data is used for production use-cases (e.g., business intelligence decisions, customer decisioning, etc.). In this context, tagging and certifications (e.g., subject to GDPR, FCRA, GLBA, etc.) offer substantial value to business users as it enables them to quickly identify any internal compliance or due diligence they should conduct on their side before considering a data source.
Data Access Method
LCDPs offer multiple methods of accessing external data through their interfaces. In the context of large datasets, LCDPs can offer cloud-hosting of full datasets that can be accessed and analyzed on-demand. The example of mobility data for retail analytics refreshed on a regular basis is such an example, where having the full dataset to analyze across a region may be more powerful than only examining a certain area. In other cases, LCDPs also do the heavy-duty work of helping users enrich internal data (or customer data) with relevant insights using a simple API call. For the user, the experience is seamless, but the back-end infrastructure required to process hundreds of thousands to millions of API transactions with just a few lines of code is quite sophisticated.
Ultimately, LCDPs help organizations deploy applications faster by lowering barriers to technology and data access. Having the ability to spin up data-driven applications and shut them down at rapid speed is a powerful tool in the race to be competitive. That being said, it is important for LCDPs to keep up with changes in external data to ensure that organizations align the right data with the right use cases.
About the Author
Prashant P. Reddy, Head of Data Advisory for Demyst. Prashant has 8+ years of experience working with tier-1 and tier-2 clients, leading data-driven workflow transformations. He has also led the due diligence and onboarding of 100s of data products into the Demyst Platform. Prior to joining Demyst, Prashant worked as investment banker at Morgan Stanley where he structured and executed debt transactions for public sector clients. Prashant holds a Master of Public Administration with an emphasis in International Finance and Economic Policy from Columbia University and a Bachelor of Arts in Economics and Political Science from the University of California, Berkeley.
Sign up for the free insideAI News newsletter.
Join us on Twitter: @InsideBigData1 – https://twitter.com/InsideBigData1
