Data has become one of the world’s most valuable resources. The world’s largest companies have built their enterprises by obtaining, storing, and analyzing data.
Fintech companies that understand the importance of data are rapidly entering the market.
Credit unions have fallen behind—a result of lagging analytics acceptance—and are playing catch-up. Why? I’ve heard some interesting reasons, including:
These statements have varying levels of truth, but aren’t relevant to why credit unions are reluctant to perform analytics. The real reason? Changing how we do things is scary.
Credit unions have become more accepting to data as a means of driving operations, but still shy away, often confusing creating reports with performing analytics. But what’s the difference?
Reporting involves organizing data into informational summaries to monitor how different areas of business are performing.
Analytics is the process of exploring data and reports to extract meaningful insights, which can be used to better understand and improve business performance.
Reports compile information. Analytics use that information to tell a story.
Figure I illustrates interest rates, charge-offs, and cost of capital on a used auto loan portfolio by original credit tier. What do you see when you analyze the report?
I see higher-risk paper outperforming lower-risk paper. Further, B-rated loans are priced like A-rated loans but perform like C- to E-rated loans.
That’s a meaningful insight, but only the start of our story. For an analytics initiative to succeed, that insight must be used to increase the value provided to membership.
There are lots of directions we could go from here:
Illustrated by the large red bubble in Figure II, lower-yielding credit tiers are dominating this segment.
We might add value to our membership by taking actions to increase demand for our higher-yielding segments (dark green bubbles) such as reducing interest rates, increasing marketing dollars, or opening up qualification requirements other than the credit score to increase approval rates.
What you may notice about these visualizations is that there aren’t any numbers. Why? Because you don’t need them.
As statistician John Tukey said, it’s far better an approximate answer to the right question than an exact answer to the wrong question.
I like numbers and precise answers. I like tables and reports. You can create meaningful analytics using tables alone, but don’t miss the forest through the trees.
Reports used in analytics don’t necessarily need to be visual, but they should be:
Don’t get left behind. Use data to evaluate relevant issues with actionable outcomes, and take action where appropriate.