CUES Member Community

Expand all | Collapse all

Data Warehousing Software / Enterprise Data House

  • 1.  Data Warehousing Software / Enterprise Data House

    Posted Aug 14, 2020 11:34 AM
    ​Hello,

    I'm curious if any of your CUs utilize data warehousing software to better consolidate data from all the programs used by various departments.  I'd like to be able to better collect, store, integrate, and analyze all our data in one (hopefully) easy-to-use tool that could help get analytical reporting into the hands of anyone who needs it and therefore help improve the results of our overall CU strategies, marketing campaigns, member relations, etc.

    Thank you for any input!

    ------------------------------
    David Dayrit
    Sr. Financial Analyst
    IDB Global FCU
    Washington DC
    202.623.3444
    ------------------------------


  • 2.  RE: Data Warehousing Software / Enterprise Data House

    Posted Aug 15, 2020 10:03 AM
    We are currently going through this same analysis/challenge regarding how to manage our data more efficiently so I would appreciate this information as well. 
    Thank you!
    Kim Westphal
    Home Town Credit Union, MN






  • 3.  RE: Data Warehousing Software / Enterprise Data House

    Posted Aug 17, 2020 10:28 AM

    Thanks for this thread.  I would like to be involved in the discussion as we are constantly trying to have systems support and give us what we need from them to help plan and change for the needs of our employees and members.

     

     

    Leona Warner  CPLP PID

    Manager of People Solutions

    G&F Hub | G&F Financial Group


    778 837 9571

    lwarner@gffg.com

    gffg.com | Facebook | Instagram | Twitter | LinkedIn

     

     

     

     

     

     






  • 4.  RE: Data Warehousing Software / Enterprise Data House

    Posted Aug 15, 2020 11:20 AM
    ​David,
    We have just completed/gone live on the Trellance/OnApproach platform. It uses an industry-standard data architecture, serves as a data warehouse, and is flexible to incorporate data from multiple sources, in some cases (e.g. card processor) via a direct data import interface and in other cases (e.g. mortgage or other data for which a custom interface does not currently exist) via supplemental flat tables. There are available predictive analytics modules (additional cost) and it can use Tableau or other user interfaces for actual analytics and presentation/delivery.

    For about 8 years we have heavily utilized ARCU on our Episys in-house system, with many dozens of reports for various departments and management levels, but we will begin migrating off of ARCU to eventually produce all such reports via OnApproach. One caution about "data warehouses" is that they generally only support what I've come to call "dumb data dump reports," because the ability to do actual analytics is absent or very limited/deficient. That type of system also creates "black box" reporting where it is difficult to verify the query logic used for any particular report, and tends to make the organization dependent (and often bottle-necked) on an I.T. programmer every time you need a new or modified report. If you install only a warehouse, you may not be happy with the results. Be sure to consider whether the warehouse supports true analytics (i.e. statistical analysis of various kinds) or at least offers flexible interface for real analytics and "data democracy" where appropriate employees can have self-service access to reasonable portions of data and access to standard reporting and reasonable ability to create their own reports as needed. Do not buy or let someone convince you to focus just on a data warehouse; actual analytics and reporting capabilities and flexibility are must-haves. Also, to get the best utilization out of an analytics/warehouse platform, you need an employee who understands data structures, basic statistics, report creation/visualization, etc. to lead the system (they don't need to be a high-priced data scientist).

    There are several other emerging CU-industry warehouse/analytics platforms (I'm not familiar with their details), but I hope my 2¢ can be helpful in evaluating any system you may consider.

    ------------------------------
    Paul Meissner CCUE, MBA
    Chief Financial Officer
    Credit Union of America
    Wichita KS
    316.265.3272
    ------------------------------



  • 5.  RE: Data Warehousing Software / Enterprise Data House

    Posted Aug 16, 2020 09:16 PM
    Edited by Richard Borden Aug 16, 2020 09:19 PM
    We are in the process of re-deploying ARCU (Symitar Episys shop) with consulting, programming, and analytics (Spotfire) layered on via CFG Consulting Group. To date we had not realized ARCU's potential and we are excited now to be on the brink of bringing analytics to our strategy and decision making process.

    ------------------------------
    rk Borden, CISM
    Information Security Officer
    Blackhawk Community CU
    Janesville WI
    ------------------------------



  • 6.  RE: Data Warehousing Software / Enterprise Data House

    Posted Aug 17, 2020 02:55 PM
    Thanks for the details Paul! In my basic research so far, ​Trellance/OnApproach was a platform that came up. Great to hear that you are happy with your choice.  How long did it take to go completely live would you estimate?

    ------------------------------
    David Dayrit
    Sr. Financial Analyst
    IDB Global FCU
    Washington DC
    202.623.3444
    ------------------------------



  • 7.  RE: Data Warehousing Software / Enterprise Data House

    Posted Aug 17, 2020 10:10 AM
    We are in the middle of implementing a brand new data warehouse utilizing Snowflake as our foundational platform.  Our overarching strategy is to collect all data from production systems in a centralized repository and then restructure the data based on traditional design patterns (STAR Schema) for enterprise reporting.  This allows us to provide our analyst with both the agility to data mine and also provide an consistent enterprise perspective towards data.

    We choose Snowflake given that it's a cloud data platform designed for analytics, performance tuning and application life cycle management requires much less effort, we can focus on designing data solutions.  This also enables us to have a foundational platform towards data science and machine learning.

    ------------------------------
    Jimmy Lin B.S. Computer Science
    Manager, IT Data Services
    SchoolsFirst FCU
    Tustin CA
    714.466.8597
    ------------------------------



  • 8.  RE: Data Warehousing Software / Enterprise Data House

    Posted Aug 18, 2020 06:37 AM
    David,

    At Hudson Valley we are building our own internal data warehouse and moving away from ARCU.  The ARCU data warehouse is a bit limited.  This internal data warehouse has 34 dimension tables, 7 fact tables and 2 aggregate tables.  The data governance council meets internally monthly and discusses/reviews additions and changes to the internal data warehouse

    The dimension tables are joined by one column keys to aggregate and fact tables.  This is done for deployment of tabular models using Microsoft visual studios.  For machine learning models we are currently using R in house and using RODBC package to pull data from the data warehouse and to push model output back to tables.  At some point we'll get around to deploying R to our Microsoft SQL server stack to execute the R on the server using stored procedures.  Each dimensional table is slowly changing dimension to cut down on data bloat with indexes and partitions for data recovery as well as performance improvement.

    If you have never built your own data warehouse or implemented a data visualization I would definitely recommend contacting an outside consulting firm to help guide you through the process.  Just remember with outside consultants you own the inhouse knowledge when it comes to your data.  So you need to make sure you pair that with the architecture to implement new reporting.

    I have worked with CFS consulting group and Convergence Consulting Group.  I recommend both as I have had very positive experiences with both.  CFS is really good with Spotfire and has some great third party data connectors to gain an early data advantage you may not have otherwise.  If you are a Microsoft shop like Hudson Valley you will save more money leveraging that relationship with regards to Power BI licensing.  For Power BI I would recommend Convergence Consulting.

    It really depends on what the appetite is for Buy versus Grow as a strategy with regards to a data warehouse implementation at your credit union.

    ------------------------------
    Troy Del Valle
    VP, Business Intelligence
    Hudson Valley Credit Union
    Poughkeepsie NY
    845.463.3011
    ------------------------------



  • 9.  RE: Data Warehousing Software / Enterprise Data House

    Posted Aug 18, 2020 08:10 AM

    Hello all,

    Here at TDECU, we are using Microsoft SQL Server for our Enterprise Data Warehouse. We are also using the Microsoft Business Intelligence Stack (SQL Server Analysis Services (SSAS), SQL Server Reporting Services (SSRS), SQL Server Integration Services (SSIS), and Power BI for analysis, reporting, ETL and data visualization. Our long-term plan is to eventually migrate to a cloud-based solution such as Snowflake or Microsoft Azure. We are a heavy Microsoft shop and our Microsoft Enterprise Agreement (EA) gives us the opportunity to leverage Microsoft tools without dramatically increasing costs. Microsoft is also the leader on Gartner's Magic Quadrant in 2020 for Analytics and Business Intelligence tools.

     

    I hope this helps.



    ------------------------------
    Marcus Manning BS
    Director - Information Management & Data Analytics
    TDECU
    Lake Jackson TX
    979.297.1154
    ------------------------------



  • 10.  RE: Data Warehousing Software / Enterprise Data House

    Posted Aug 18, 2020 10:10 AM
    Ideally a cloud platform like Snowflake still have some dependencies on a native cloud platform.  We are a large Microsoft shop as well so we are taking advantage of that to built out cloud infrastructure for future analytical needs on Microsoft Azure.

    ------------------------------
    Jimmy Lin B.S. Computer Science
    Manager, IT Data Services
    SchoolsFirst FCU
    Tustin CA
    714.466.8597
    ------------------------------