ANAHEIM, Calif.–You can’t drive the credit union by looking in the rear-view mirror.
That observation and a number of other insights were shared during a panel discussion at NACUSO’s annual meeting here on “Setting the Definition for a Data-Driven Future.” Panelists, including one credit union, shared where analytics—at a low cost in many cases—can be better used by credit unions and what lies ahead.
The panel members included Rojan Nair of Celero, Nick Curcuru of Mastercard, and Andrew Bertrand of Our Community Credit Union in Washington State.
Here’s a look at the Q&A:
Q: At a high level, what are you seeing with analytics, and where are CUs doing well and where is there room for growth?
Nair: Most of banks in Canada (where Celero is based) have announced 60-minute preapproval of mortgages, and now the credit unions want to do that, too. It’s pretty interesting. We have reached a stage where the buying, borrowing and paying experiences can’t be separated; they must be the same. I talk to credit unions saying I am losing all my auto loans because Honda and Toyota are providing those loans fast. People don’t have time to go back to a credit union when they are in a showroom; they want the experience to be real time.
Curcuru: One of the things we see credit unions doing is trying to move that personal experience out to the digital channels. How can they give the kind of experience you get in a branch through a mobile device? They want to know how to do that with the analytics.
The five big things we are working on and where we see interest are churn analysis, acquisition of members, cross sell, segmentation and action. Those are the things we see using the data analytics for in the credit union space.
Q: From a credit union perspective, how does Our Community CU compare and where are you in this journey?
Bertram: Loan payments was one of the things I worked on early on to see where members making payments to, and then finding out why a member left, who was the last interaction with. What channel did they last interact with? What’s the loss-ratio in that channel?
We’ve been doing ROI per channel and segmenting the members. Who are they and what do they look like? We have 35,000 members, but approximately 10,000 look exactly the same, so you have to find other ways to find out what makes them different. Why do they come to the branch rather than calling, for instance?
We started about four years ago in analytics and started with prescriptive analytics. It helped us to react to the market faster. It used to be you waited until the end of the month and got the reports. With indirect lending, we were able to see a trend and we reacted quickly and changed loan rates and hit our goal and actually lowered some of that volume.
Q: What other issues need to be addressed?
Nair: Who is your biggest competition? How many believe it’s banks (about 5% of the audience). The reality is of the 100 institutions we have (as clients) we are able to look at that data you would be surprised that even in rural communities fintechs are even breaking into those markets. We started looking at these external behaviors. We figured out around the sixth year of a relationship, members start looking around. But you can only determine this if you have a pool of data. A lot of social circles and economic behaviors are helping to shape this.
Q: What are the challenges that credit unions in the beginning of their journey in analytics are facing?
Curcuru: There are always the same three challenges, the first being the cost of analytics technology. The good news is that cost has come down significantly; it’s about one-third of what it used to be. The other two are talent—you are always fighting for talent in this space. And that’s where you have to be a little creative. The other part is what data to go after. Most people say I’ve got my data, but there is also social media data, Google data, Facebook data, mobile apps data. That’s where things to start to fall off, they have a thousand data elements. But you don’t need a thousand, usually you need just eight to 10. At they used to say when I was with the Disney Company, ‘Think big, start small, act fast.’
What is the right use case that will make the business grow to drive results? I would venture to say it’s probably churn.
Nair: Your quality of data drives ROI. This is a big difference between credit unions and banks. They are not going to collaborate. You can create an ecosystem and it doesn’t mean violating privacy. You can leverage data and patterns across each other’s ecosystems rather than trying to do it on your own.
Q: What hurdles have you overcome in four years?
Bertram: The main thing is you just have to start. Some credit unions say they want to get a data warehouse, but what are you doing right now? Do you have buy-in? Who is looking at your reports now?
I like the idea of starting small, because it’s a cultural transformation and my CEO was great, but the data itself and where it lives and who owns it are the big issues. Now, we have a policy that every time we sign with a new vendor, (we address) where the data is going to reside. Is it compatible with our system and can we get it into our data warehouse?
I talk to some credit unions where marketing still does there thing or accounting doesn’t trust something else. So, we automated our data so it balances and matches up monthly to get the trust of accounting, and then the trust of lending officers. There are also the challenges of marrying all that data. I had to talk to every department and sat down with them to get their feedback and find out what things mean to them. Everything is trackable.
Nair: You don’t drive using your rear-view mirror. We took on the challenge of bringing in this hybrid data set and asked, ‘Who are members likely to leave in the next 30 to 45 days and what are the reasons they would do that?’ The next question we asked was what do we offer a member when they next walk in to be their next best product. That took about six months to work out. Then the discussion got into a member profitability index and using analytics there. We shouldn’t be asking a member who’s been there five years the same questions as a member who is just joining.
The member is right at the core of experience, and then you look at their personality and their experience expectations. Don’t ever see someone coming for a mortgage as someone shopping for the best rate. How do you plan to be part of their home buying experience? If you can answer that question you have a use-case for analytics.
Curcuru: One of the biggest things we see is the need to tie the analytics to a business goal. If you can help someone make their bonus you will create a champion for you in that organization. We had an organization where no one wanted analytics. We sat down and said how can we sit down and help you out, and we worked with a mortgage officer who saw mortgages leaving. We did something really simple. We took the mortgages on their books and got the MLS listings and matched them to the listings. Did you know this member’s house is for sale? In our particular area, people were only moving five to eight miles. This was one way to go to the member and say we have some really good rates. (The mortgage officer) got a 47% uplift and that got him his bonus, and people took notice. We didn’t do any data mining. No regression. We just matched this to this. If someone is selling a house, they are probably looking for a mortgage.