Advanced analytics ripe for no-code, low-code tools to enter the field

Last year, my colleague Rebecca Natale wrote a great article on the “no-code movement,” which its supporters are saying has the potential to democratize software development. No longer do you have to be an expert coder to build software, even full-stack apps or enterprise platforms; low- or no-code providers supply tools with which any user can build software visually, dragging and dropping pre-figured components and widgets into their apps. Genesis, a start-up on the roster of specialist VC Illuminate Financial, is one example of a vendor that is offering this kind of solution now, with its microservices development framework that uses pre-built software to create new applications. Reb quotes figures that say the low-code application market is set to reach nearly $50 billion by 2026.

I can see why low code is seductive to financial services, because it seems to offer a solution to problems that have appeared intractable. If anyone can build platforms, organizations no longer need worry so much about attracting developer talent, and business and IT can be united in one individual. It could enable firms to develop their own tools more easily without involving external vendors on the build. And I think there is one area in particular where more modular approaches will be attractive in 2021: advanced analytics.

Any reader of this magazine surely knows all too well how banks and asset managers generate and collect vast amounts of data; and how they fail utterly to make any real use of it in advanced analytics. Consultancy Element22, in conjunction with Greenwich Associates and UBS, published a benchmark study in which they found that most buy-side firms they surveyed are eager to leverage alternative data to beat the competition, but only a minority were anywhere near being able to do so. The study was published in late 2019, but Element22’s London-based partner, Mark Davies, says those patterns are still relevant, and there is no reason to think that the situation is much better at banks.   

So is it likely that, as a kind of corollary to the low-code movement, we will see more self-service providers emerging in the next few years, specifically in advanced analytics? Davies seems to think so; he says the technology from vendors in this space is maturing.

Analytics tools emerged about five years ago, he says, but were mainly targeted at data scientists who were wizards in languages like R or Python. The scientists had to then build the analytics capabilities from the ground up. 

Providers in this space are now lowering the bar in terms of the expertise needed to build analytics capabilities, he adds, which has meant more offerings in self-service tools that give users capabilities that they couldn’t write themselves. 

Apteo, for example, says co-founder and CEO Shanif Danani, wants to “give every person the tools to become a data scientist,” and believes its Predictive Insights is a no-code tool for identifying the segments of users’ data with the biggest impact on their KPIs. It wants to use artificial intelligence to create predictions from machine-learning models built from their data.

So, yes, I think we’ll see more of these providers emerge this year. However, not to be the raincloud at the picnic, I suspect these tools in some cases will be wastes of time and money for many firms. It’s a boring fact of life that financial firms wanting to take advantage of self-service tools for advanced analytics will have to get their data management right first.

If a firm wants to use a self-service visualization tool for a single task, that is one thing. But if it wants to scale it up for enterprise-wide use cases, the data will have to be sourced from siloes in legacy systems and prepared for use. There’ll no doubt be gaps and inconsistencies in this data, some of it will be stale, and there will be security concerns and controls around its use. Companies should have data governance policies and plans in place before they can really tap the benefits of advanced analytics for customer insight, or getting the best prices for the front office, or beating competitors to investment opportunities. 

These old-fashioned data governance concerns are perennial, no matter what shiny new toys may be available to financial services for analytics and machine learning.