Vendor

Is Your Culture Ready for Data Analytics?

 

 

 

 

When most of us think about data, we think about numbers, charts, graphs, and reports. However, we need to talk about CULTURE.

Collecting, aggregating, normalizing, manipulating, and presenting data seems to consume the conversation. However, here is the question we need to ask ourselves, "Is our culture ready to become truly data-driven?" Too often organizations spend lots of money on software, tools, and expertise, only to learn the culture was not ready to be data-driven.

How do I get the culture ready?

  1. Remove the data silos. Too often I see data owned by departments. Lending owns lending and collection data, marketing owns marketing data, finance owns finance data, and call center owns call center data. The problem data silos bring to the organization is the loss of data integrity. If the Chief Lending Officer is responsible for making lending and collection goals and they own the data in a silo, there is a risk that the data can be manipulated to tell the story lending wants to show. The only way to wean the credit union from silos is to have the data centrally collected, managed, cleansed, and normalized, to have the definitions of commonly used data terms standardized, and to have the data report queries standardized. Centralizing data is a painful process because some of these silo owners may not like this neutral party controlling the story the data is telling the organization.
  2. Make all data accessible to the organization. Data is an organizational asset and needs to be available for all to see. This accessibility and availability of data will allow anyone in the organization to see what is happening and to even ask questions of the silo owner or the data for clarification or understanding.
  3. Create a Business Intelligence (BI) unit in your organization. Their job is to collect, aggregate, store, cleanse, and manage all data. The BI staff does not write reports but are the data experts. With accessibility, tools and training are available to the subject matter experts, whether they are marketers, lenders, operational staff, or finance staff, to run their reports. However, the queries, definitions, and timing of reports are understood so variances in reports can be easily explained. This method of data management is called centrally managed, widely distributed or the "hub and spoke model."

The organization needs to see Business Intelligence the same way they see Human Resources, IT, or Marketing. It is an independent business unit that makes data available to the entire organization in a way they can access and query the data. The organization structure needs to include a strategic owner, usually an SVP or EVP, and a tactical owner. The strategic owner keeps data projects, software, and tools visible in the planning and budgeting process the tactical owner owns the work of collecting, aggregating, normalizing, cleansing, and monitoring of the data, definitions, and queries.

Data is the credit unions most valuable asset; it needs to be right and has a strategic focus.

E-mail me when people leave their comments –

You need to be a member of Credit Union Big Data Analytics Community and Summit - CULytics to add comments!

Join Credit Union Big Data Analytics Community and Summit - CULytics

Comments

  • CU Employee Community Chair

    Thanks for sharing these thoughts, Richard. The concept of situating analytics in the widest possible enterprise context in early stages is a provocative one, sure to excite as quickly as it challenges. If you don’t mind, I’ll offer a contrasting view, and urge the CULytics community to weigh in with their own thoughts on which view has greater merits for their credit unions.

    My thought is that while credit unions will absolutely want to encourage analytics on top of a data model that gives a minimally holistic view of our members, I suspect that MUCH of the earliest wins with analytics for most will be firmly seated within individual business silos, typically where there’s initial passion for scientifically solving challenges with data, and that we should embrace that reality and use a lighter touch when asking analytics too early to solve bigger issues at the enterprise level.

    Many of us who’ve seen the genesis of advanced analytics programs in credit unions recognize that it’s VERY often a particular one or two very specific business problems that ultimately justify the significant initial investment in infrastructure and resources to stand up data warehouses and get analytics practices launched. These initiatives are born more often from fraud departments or finance departments or card service departments or marketing departments when they outgrow their nascent (spreadsheet) data management strategies.

    But even after significant credit union investment is made to take analytics to the next higher level — with the merging of a half-dozen data sources and some enterprise query/reporting tools in place, data passion is apt to be uneven across the credit union. Some departments are early adopters. Others are later to the analytics game. Is there wisdom in overtly enforcing analytics as a priority in a top-down executive way? I’m not so sure. Isn’t it a path of lesser resistance to let the early departments prove the value of analytics with their initiatives, and let their success cultivate the interest in the laggard departments? It won’t take too many analytics wins in an area or two to have everyone in the organization clamoring, “I want MY piece of that action” — and that bottom-up demand for analytics would seem to have significant advantages in growing a program. Can we trust that dynamic?

    Sure, analytics maturity is likely to proceed unevenly within the organization for some time. The fact is, limits on resources almost always mean that all areas of a credit union are unable to share in the benefits of analytics all at once, anyways. Everyone takes a number and waits his turn for his or her area’s second-tier data sources to get wired into an ETL. Letting pockets of enthusiasm drive where investments get early priority is as rational an approach as a lot of other schemes… one can hope that, at a minimum, data investments will get leveraged promptly, and not risk sitting on a shelf.

    This is a far more “laissez faire” way of thinking about growing analytics than what you're proposing, I daresay, and many may be uncomfortable about the lack of central control driving direction. And yes, it trusts that the broader data culture will emerge over time, organically, based on an ethic of merit/success rather than executive persuasion. But perhaps it’ll have its value for some, who see wisdom in this alternative approach.

    Dale Davaz
    STCU Research & Development

This reply was deleted.

Related Post