Data Analytics Offerings on the Move

As interest in data analytics picks up among credit unions, so have efforts by vendors to hone their offerings in the financial services sector. 

Among recent launches, partnerships and mergers and acquisitions activity, CUES Supplier member Trellance, a credit union service organization based in Tampa, Florida, acquired two data analytic companies, IronSafe and OnApproach, and CUESolutions provider for executive benefits and retirements solutions CUNA Mutual Group, Madison, Wisconsin, formed CUESolutions provider AdvantEdge Analytics as a fintech startup in 2017. In the wider financial services sector, FICO and Equifax introduced the Data Decisions Cloud, described as an end-to-end data and analytics suite for financial institutions “that addresses key needs across risk, marketing, and fraud.”

In addition, CUES Supplier member Temenos has acquired Avoka and its member onboarding software to enhance Temenos Infinity, its digital front office banking product. “What Avoka provides from an analytics perspective is a real-time view of the customer journey,” so credit unions and banks can see where accountholders are running into problems and modify screens and processes promptly, explains Todd Winship, product director/data and analytics for Temenos.

Expanding Scale for Future Development

Trellance’s acquisitions bring together IronSafe’s data extraction capabilities and the M360 platform of OnApproach, which offers real-time availability and continuity of data, with the aim of expanding the customer base for those products and providing greater scale for continued development, says President/CEO Tom Davis.

Portability and continuity of data will be a big concern for credit unions moving forward, Davis says. “Credit unions want to be able to pick the best of breed in their technology providers. Some of our credit unions have a dozen or more of these relationships. If you move from one provider to another, you don’t want to think about losing data and starting from scratch with data analytics and mapping to new data sets.”

Trellance may be best known for its payment services, but in 1999 the company (then CSCU) created a program that analyzed aggregated data to help credit unions maximize the value of their credit card portfolios through a scenario-based tool. Credit unions could model changes to various elements of their offerings, such as launching a balance transfer or introductory rate campaign and forecast the impact on their portfolio. 

Trellance is augmenting its staff with experts across data domains from auto loans to call center interactions, but Davis suggests that the CUSO’s wheelhouse—payments data—“is among the richest of all domains in providing a behavioral view in terms of how members are interacting with their credit unions and other financial service providers.”

Multipurpose Tool

Even though many credit unions are just getting started with data analytics, there is no shortage of use cases for them to adopt and adapt. Among the most common are models “to understand who your profitable members are, what their product and service use looks like, and how you make sure that they continue to be engaged,” says Shazia Manus, chief strategy and business development officer with AdvantEdge Analytics. These applications aim to stem attrition and to improve lead generation so the right offers go to the right members at the right time. 

AdvantageEdge Analytics is also working with credit unions to develop “next-best branch” applications through membership geographic segmentation models. “And there is a lot of interest in robotic process automation, which is more around general resource management, efficiency and channel optimization,” she says. For example, “how can you use digital bots to provide more 24/7 experiences through your delivery channels?”

But first things first, “you need to have the data organized into a single source so that it’s searchable, consumable and usable,” Manus notes. Forming that foundation begins with conversations with business users on their marketing, operation, lending and finance needs, with guidance from IT on the credit union’s various systems, other sources of data and accessibility issues. That information forms the basis for a strategic data analytic plan that lays out an implementation roadmap and for a data governance foundation, which encompasses identification of data assets, data loss prevention, creation of ownership, policies, procedures and training. 

Credit unions should look to their data analytic partners to provide a full range of capabilities, Winship says, including:  

1. Descriptive—These data analytic components monitor service delivery to answer questions such as: What’s happening in member journeys and processes? When is it taking too long? Where are members abandoning processes because there’s some type of obstacle? Descriptive data analytics should be “granular to the point of identifying which screen may be causing an issue so credit unions can respond quickly and accurately to correct any problems,” he says. 
2. Predictive—These components help credit unions anticipate when member interactions are likely to go off the rail and take proactive steps in real time, such as offering assistance through a chatbot or phone call. 
3. Prescriptive—These algorithms support service delivery simulation and optimization to guide credit unions toward solutions and improvements.

Deploying analytics is a journey of building and testing these capabilities in smaller, successful sprints instead of one big push, he recommends. And along that journey, the organization should also be developing a strong data governance foundation.

Next Level: ‘Analytics for Robots’

Deployment is beginning to shift away from a focus on “analytics for human users,”—i.e., to guide branch managers and staff, CFOs and marketers—in favor of “analytics for banking robots,” such as digital marketing software, interactive chatbots and other systems that enable real-time member interactions, Winship suggests. 

Credit unions should be working toward that next level of analytics, which is completely real-time, API-enabled, based on machine learning and artificial intelligence, and fully integrated with key systems, Winship says. The technology is available today, but credit unions may need to cross some stepping stones to adopt that full digitization.  

The data needed to power both types of analytics is fundamentally the same, he notes. “If your credit union is already consolidating data into a large, curated, high-quality datasets to support human decision making and better process efficiency, that is essentially the same dataset you’d use for that real-time engagement for the next generation digital banking solutions.”

‘Data Lake’ for Credit Unions

Trellance is currently working on developing a hosted data analytics solution for credit unions that want “to get out of the hardware and data analytics infrastructure game,” Davis says. “At the end of the day, what they want is those insights, those ‘golden nuggets,’ and a cloud-based environment is very important to deliver that ubiquitously to all our credit unions.”

A second project now underway is the creation of what Trellance is calling the OnApproach Caspian Lake, a shared data repository for client credit unions and perhaps other companies as well. The aim is to develop a system whereby credit unions can share member data anonymously, stripped of personally identifiable information. 

“The more data you can push into these models, the more insights you can get. Caspian Lake would allow credit unions to see patterns within their member base and across the industry and take what they’ve learned back to their members and enrich the value of their offerings,” Davis says. 

Access to payments data could be a “very rich carrot” to encourage credit bureaus, local and regional merchants, and other companies to add their data to the shared repository as well, he notes.  

Credit unions are ready to embrace data analytics for applications as diverse as operationalizing the new current expected credit losses methodology to building their credit card, auto loan and mortgage business, Davis says. “I wouldn’t be surprised if within the next year or two, there are several hundred credit unions using a data analytics platform on a daily basis and building a staff around that platform.” 

Data analytics can also help financial cooperatives address “big picture challenges,” he adds. “What if Amazon moves in and starts to offer new financial services? How do we keep our products and pricing fresh? Credit unions can use data analytics to stay competitive with current and future disruptors.” 

Credit unions aren’t developing and executing their data analytic strategies in a vacuum, Manus notes. They face the same challenges as companies across business sectors—to pull together data from disparate siloes into an accessible and usable form—to meet daunting expectations. 

“The GAFAs of the world—Google, Apple, Facebook, Amazon—are setting the baseline for what a consumer experience is, and all of that is happening by leveraging data in the most helpful and meaningful way,” she says.

“This is a complex problem that every legacy industry has on its plate, and it will require a laser focus, collaboration and a commitment of cultural agility to ‘fail fast,’” Manus adds. “There’s going to be tremendous learning along the way, so how do we make sure that we build that agility and resiliency and keep moving forward? We need to collaborate not only with credit unions but industry players as well, given the costs and stakes at this point.”

Attend to Strategy and Culture First

As with other technological advances, a few credit unions are already deploying sophisticated analytics while others are just getting started. Data analytics will get “democratized” over time as implementation costs level off such that that mid-size and smaller organizations can increase their efforts in this area, says Manus.

In the meantime, “there are ways smaller credit unions can get started with this foundational strategy, looking at what data they have, what they could be doing and what member friction and use cases they’re really looking to solve,” she recommends.

Formulating a data strategy will also help credit unions consider a range of possibilities: Beyond the obvious stores of structured data, will it be possible to harness social media posts, voice data from call center interactions and other emerging data sources? Will an in-house data infrastructure or a cloud-based solution be a better fit? How will the data warehouse be secured? What new expertise and staffing will be required? 

In short, “your people, your products, your processes, your problems and capabilities—you need to look at all of this holistically to drive what solution you really want to partner for,” Manus says.

The foremost challenge facing credit unions in launching and optimizing data analytics is not technology but “mindset, culture and strategy,” says Naveen Jain, president and founder of CULytics, San Jose, California.

Across the industry, credit unions are very early in the maturity model of deploying and reaping the benefits from data analytics, according to surveys conducted by CULytics. More than 80% say they have taken at least initial steps toward implementing this tool, but on four-point scale—spanning ad hoc, basic, managed and optimized data analytic solutions—the average is about 1.4, Jain notes. 

CULytics offers an online organizational readiness assessment, a seven-page questionnaire covering five key areas of transformation (outcomes, strategy and roadmap, people, process and technology) to help credit unions plan their approach to data analytics.

Karen Bankston is a long-time contributor to Credit Union Management and writes about membership growth, operations, technology and governance. She is the proprietor of Precision Prose, Eugene, Oregon.