In the Panel Session on “Driving Analytics Journey”, the panelists Paul Ablack, Sundeep Kapur, Anne Legg, and Andre Iervolino discuss the data analytics journey and various approaches towards it which can be adopted through the aid of influential case studies and practical examples.

Case Study - The Analytics Journey: Ideal Credit Union by Paul Ablack | Founder and CEO, OnApproach

This is a case study about member loyalty and about keeping members. The idea of the case study is that there are so many choices available to members, but amidst all these choices,  how do we still keep them engaged and loyal. Also at the same time, how do we keep making money as a credit union?

According to Paul,  holding on to your most profitable members is the ROI of analytics. In a credit union, there’s a model that 20% of your members make 80% of your profit. In this case study, he talks about predictive data analytics, descriptive data analytics, integrating the data and working with it.

The Challenge

Ideal Credit union is 692$ million in assets, and they have 52000 members. They are very committed towards delivering quality service to these members and helping them achieve financial and life goals. They used to have MCIF ( Marketing Customer Information File) system for their member rewards program and it was based on products. Eventually, The CEO Questioned his investment returns and wanted proof that he was making money out of these members. So he found out that 8000 members were contributing to 80% of the net income in 2012 and called these members VIP members. He wanted to grow the base of these VIP members and reward them for their behavior. He also needed to determine how these members were using the products and in order to do this, he had to look into the transaction data, which was tied up in data silos.

The Solution

They created three levels of VIP Patronage. So the income which they generated, a part of that would go to pay the dividend to the credit union on different levels. There were 66 criteria for allocation of these members into the three levels. The standards were - VIP Patronage dividend, VIP Loan Rebate Dividend and VIP Deposit Bonus Dividend. These three levels of profit required integration of data. To do the data integration they - a) purchased a CU Enterprise Dataware solution b) Integrated disparate data c) applied 64 distinct business rules at the API level. The algorithm so constructed would enforce these 64 rules to determine how much the member contributed as net income and interest income and so forth.

According to Shari Riley, Manager of Financial Reporting and Risk Management of Ideal -

“From the program support perspective, we see members starting to think about the program...they are thinking about it throughout the year and proactively trying to be sure they will meet the criteria for the upcoming year-end. They became goal oriented.”

As a result of these solutions :

• The number of members since 2012 almost doubled.
• As of January 2018, more than $3.1 million has been paid to these VIP members as a result of this program.
• Although the members were profitable, they were earlier not using the 4 C’s program - Checking, Credit Card, Car, and Casa. So they started directing these products towards the profitable members.
• There has been a 20% increase since 2014 in the members with at least three products.

According to the CEO, Dennis Bauer, “VIP is the most popular member programs we have ever launched. Now with the additional ability to drive other marketing programs with the VIP information, we are truly leveraging the value of our data.”

Selling Starting And Effectively Executing A Big Data Project by |Sundeep Kapur| Educator, Digital Credence

Sundeep started his session with three examples that were related to a Big Data Project of Anonymous Organizations.                             

Example - 1: Anytime, Anywhere

Anytime Anywhere Communications, they power mobile devices and a few years ago they bought one of the largest television entertainment companies in the industry. They said - “Normally in the old days, everybody would get every offer and it would be difficult to understand why their rates were so high.” But they took a very pragmatic systematic approach towards Big Data.

Approach  :

• Sorted customers based on products and services.
• Further segmented by channel adoption.
• Created a multi-channel messaging plan.

Use cases :

• If a consumer has all products, thank them and offer an extended and discounted renewal.
• If a consumer is missing a product, make graded offers to him based on convenience, incentive and more.
• Promote offers systematically across the channel.

Analytics Centre of Expertise

• This team carried a financial quota and this drove the budget.
• It was held accountable for promises made.
• It chartered with continuous improvement.

Results - From the number perspective, the results were spectacular. In 2014, they drove 16  million dollars in return, in 2017 they had 310 million dollars because of what was originally a small data transformation team that has taken things further.

Example - 2: Lead with Payments.

Credit Unions are a financial institution and they need to take things forward from there. This example is of a 500 million dollar credit union which supported colleges. They had two types of ACH ( Automated Clearing House) payments - Payments going out to American Express ( people interested in the brand),  and payments going to Discover card ( which gave you cash back).

Approach

• Analyze ACH transactions.
• Create multiple segments.
• Unique graded offers.

What consumers want when they use a card

• Convenience
• Security.
• Incentives

What credit unions want when customers use a card

• Top of wallet card.
• Reduced fraud.
• Different source of revenue.

The concept of the golden triangle in financial services

The usual approach of banks and credit unions is to reduce the costs of deposits, maximize interests through loans and work on payments. But what FinTech did was to lead with payments, because they were taking a little bit of each transaction at the time. The approach is to lead with payments, connect with the consumer and then bringing them to do loans and deposits.

Results

• 78% card and mobile adoption.
• 60+ transactions a month.
• Chartered with continuous improvement.

Example - 3: First impressions

You do not get a second chance to make a first impression. Think from an onboarding perspective, this credit union was sending out 14 messages over a 90 day period to connect and communicate with those they served. There were three categories of boarding - onboarding, loan boarding and deboarding.

The messages included the following

• Welcome package (information that was promised).
• Relevant testimonials based on product or service you have purchased.
• Did you know? Inform about more products and services.
• How did we treat you?

After any of these stages, they brought them in for a financial checkup to understand challenges and provide solutions, opening a lot of possibilities. Secondly, they adopted a mobile environment approach and made messaging and emails adaptable to mobile infographics.

Results

• 70% engagement, 21% conversion on three touch transactions.
• Inbound and outbound strategy for real-time communication.
• Conversations with customers.

The Strategic Impact Of Big Data

Here are the 5 key sectors in which Big Data contributes.

• Increase Revenue
• Reduce Operational Costs.
• Enhance Engagement.
• Be More Efficient.
• Be Legally Compliant.

Common objections or excuses from executives :

• “ I’ll have to pay 150k to hire a programmer.”
• “ I have outsourced my data.”
• “I do not want to disrupt anything. I am close to retiring.”

You should be able to define your purpose.

• How much money can you afford to give a prospect to become a member or for a member to stay?
• We have to make analytics a repeatable process that drives revenue.
• What is big data? Deciding on the best possible option based on an efficient scientific method that can be proven.

Ways To Ensure Step By Step Continued Learning

•     Management education and picking projects by joint business and technology focused decision delivery workshops.
•     Conducting an analytics strategy and assessment to gauge organizational readiness, and highlight maximum impact projects.
•     Implement prioritized big data journeys by applying business and technology to drive results.
•     Evaluating success with a renewed focus on continuous improvement.
•     Instilling Analytics Expertise within the organization.

Drowning in Data and Striving For Insight? How To Fuel Your Credit Union Growth by Anne Legg | Director of Client Strategy | AdvantEdge Analytics

If you walk to the member first, you are going to solve your credit union problems. To drive a successful strategy, your two questions should be

1. What is the business problem I’m going to solve?
2. What is the member problem I’m going to solve?

Common Credit Union Friction

• Harness data to improve member experience and operations.
• View member information as a single enhanced source of truth.
• Access real-time reporting dashboards with visualizations, insights, and analytics.
• Create a one-click member experience anytime, anywhere.
• Automate and streamline internal processes to be better and faster anywhere.

Approach to executing analytics opportunities: Build a roadmap

1. Source of value - Defined by business need.
2. Data Ecosystem- Appending key external Data, data integration from internal systems.
3. Modeling Insights - Predictive modeling, apply linear and non-linear modeling to drive insights.
4. Workflow integration- Integration of insights into day to day workflow, drive tech enablement.
5. Adoption- Build frontline capabilities, Proactive Change Management and tracking of adoption with performance indicators.

If you just focus on solving the member problems, you will be able to solve lot of other problems on the way.

Data Transformation Example - Zero UI ( User Interface)

Where the member interacts with a personal assistant using zero user interface, zero UI is an interaction where an individual’s movement, voice, and even thought can all cause a system, such as financial assistance to respond to a person through their environment.

Features - Automated Loans, deposit and bill payments, predictive and prescriptive interaction.

Predictive Models

CU’s want to anticipate consumer’s next best product and make better decisions. As Credit unions, we should be able to leverage predictive analytics to customize member experience and improve operational efficiencies. Credit unions have a huge opportunity to leverage data analytics to create value across the value chain and provide members with better products and services.  If you are looking for a model, the three categories you should aim for are - commercial effectiveness, risk control, and operational efficiency.

To have a predictive HR management, you should be focusing on KPIs such as Churn Rate, to understand where the members are going and what they are up to.

Use Case - Identify triggers for account closure, dormancy to re-engage members with personalized offers and retention marketing.

Typical impact: 20-25% churn rate reduction.

Conversion of 1-2%accounts to active.

Most effective: end to end approach focused on business value.

1. Source of value- What are you doing to make the member’s life better?
2. Data ecosystem - Having the right toolset.
3. Modeling insights - Predictive modeling.
4. Workflow integration- Integrating data into the workflow.
5. Adoption- Productive change management.
6. Iterate throughout

Big Data, Small Data: What About The Right Data -  by Andre Iervolino | VP of Business Intelligence and Strategic Performance | Cabrillo Credit Union

Credit  Unions tend to look at data and use it for creating projects in silos.

But their practice was to create a BI strategic plan, incorporated into the corporate plan. Because of this, they were able to design a pyramid where they designed the vision, the mission, the strategy under the BI plan, services provided, the types of data, the tools that were needed to be used and operationalizing the overall strategic plan.

Data vision and mission :

Vision: To become a prescriptive enterprise driven by a democratized data culture - From hindsight to foresight.

Mission: As a provider of data related services, generate actionable business intelligence through Descriptive, Diagnostic, Predictive, and Prescriptive Analytics.

Data Strategy: Going through six different key result areas.

1. Revenue - Whatever was done, had to generate revenue.
2. Efficiencies - Save money for the credit union.
3. Data Governance - Develop principles and strategies for business governance.
4. Compliance - Enhance current business compliance,
5. Implementing the right BI tools.
6. Ensuring retention, acquisition, and development for a strong BI team.

Divided services into categories - Enterprise data management services and Strategic data services .

When you look at the data available to you, you can look into the following four categories, ( and it is possible to squeeze any kind of data into these categories).

1. Competitors
2. Macro and Micro economic data
3. Member data.
4. Market data.

Identify the problems you want to solve and what will help you generate the money. Then identify the data type. Having uncomplicated and easy to use tools is important.

Data Operationalization and Results

  Designed 8 different integrated programs to leverage the power of business intelligence and the right data.

1. BI Tools
2. Data Warehousing.
3. Target Marketing.
4. Data Sources, Data Governance , Data Compliance.
5. Data Science.
6. Consumer Insights, Market Insights, Research, and Data Reporting.
7. Strategic Performance
8. BI Team

Results -

1. Migrated 17% of our indirect households to a retail status by using a CRM that was built from the ground up and took three weeks to build.
2. Generated 65 million dollars in CD balances.
3. Member engagement index drove by 8 variables through BI indices.
4. Through the attrition model, increase in 12% member retention.
5. Through market analysis, placement in 7 different branches.

Three things to take away

• The size of the credit union doesn’t matter.
• Do not start with silos projects, start with vision, mission, strategy and so on.
• You do not always need to partner with external vendors to apply algorithms to the data.

From this panel, it is best to conclude that what is best for the members should be done first. While making crucial decisions about what is the best product, to ask what is member’s next is the best guide. It is only possible to answer this question if our data is relevant and right. If you can solve the member’s problem with whatever you are offering, they will trust you. And for a credit union, a member’s trust is the most crucial element.

WATCH FULL PANEL SESSION HERE