A:360 Podcast #19 - Crawl, Walk, Run Progression of Analytics

Today’s A:360 discusses the importance of the “crawl, walk, run” progression when getting started with analytics. Feel free to read the summary of the podcast below or scroll towards the bottom of the page to watch or listen!

Like most major projects and strategies, realistic goals and timelines need to be adhered to for the greatest chance of success. Many organizations may want to immediately start using “big data” and “data science” yet they haven’t even tackled the basics of traditional business intelligence. This is where the crawl, walk, run mentality comes into the picture.

The “crawl” phase is what I would consider traditional business intelligence. This is the phase where the organization stops living and dying by their Excel VLOOKUPs and starts to use relational databases and common reporting tools. Visualization tools (like WebFOCUS, PowerBI or Tableau) are implemented allowing the organization to consume information in a fashion other than a spreadsheet.

The “walk” phase of analytics is where “real” analytics can begin. Data governance has been set in place so key definitions and terms are understood by all within the organization. Data is no longer stored and reported on in silos. Data transparency and data integration allow the organization to see a 360-degree view of the member. At this phase, your staff does not need to go to ten different places to get data for a report. Near real-time or real-time analytics can be developed. Questions that are asked are not “what did we do last month?” but “what will we do next month?”.

The “run” phase of analytics is where data science and statistical models become fully realized and leveraged. This is the point where your organization may employ or work with outsourced data scientists. Statistical models are developed based on your underlying data structures to do any number of things including:

Advanced Member Attrition Analysis – who is likely to leave the credit union (or no longer use us as their PFI) and when?
Refined Risk Modeling – is using credit score really the best way to manage risk and maximize NIM? Can we layer in other attributes in the origination process to underwrite traditionally “riskier” loans without impacting our risk profile?
Advanced analysis of card transaction data to identify opportunities for improved interchange income and greater utilization (i.e. top of wallet) by the member.

There are about a hundred other examples of “run” phase analytics that could be leveraged. The idea, however, is that in the “run” phase the data is working for us in every way imaginable. At this phase, the organization has the structure, the culture, and the skills to make full use of the data’s potential.