Smarter ideas worth writing about.

Defining a Modern Analytics Architecture

How are you leveraging the tools and technology in your company to make better decisions?

It is a necessary competency for an organization to be able to analyze their own data and use it to establish goals, measure success and plan for the future.  A typical analytical architecture involves systems and processes for collecting, manipulating, analyzing and presenting information.

One of the major challenges we've been helping businesses overcome is how to be able to leverage new technology to enhance or upgrade their existing analytics framework.  We've also been able to help enable new opportunities for data and information exploration.  In order to implement a modern analytics framework, we look to ensure that systems and processes are in place for the following operations:

    • Data collection and exploration

    • Data consolidation, manipulation and restructuring into information

    • Information exploration, analysis, presentation and delivery

While some of these elements may be established business competencies, the evolution of exploration and analysis is critical to discuss further.  In the past, these activities were almost exclusively for those with deeply technical backgrounds.  The output from their efforts would typically be incorporated into the presentation and delivery layers, but the end users are still only able to consume what is provided to them.  It would not take you long to look around the office for a person (or team of people) tasked with taking the weekly sales report and merging it with the next weeks production report to produce a forecast.  Imagine if all that information was in one place and those resources were able to focus their time to analyzing instead of simply combining.  It’s an exciting vision made possible by recent advances in the Business Intelligence space.

It is important to note that the framework is not constraint to one solution provider or technology stack.  New products and powerful tools are being developed at an astounding rate and can be very focused on solving one problem.  The reality is that a niche tool may solve a subset of problems extremely well and we’re finding ways to incorporate new tools into existing ecosystems where appropriate. 

Data collection and exploration
As a compliment to the relational database system, the modern architecture introduces a place for unstructured data.  It is necessary to repeat this: unstructured systems are a compliment and not a replacement to relational systems.  Many early adopters have learned that the unstructured data systems are great for taking new data and loading it into the inexpensive and scalable environment in order to start determining how it fits into established metrics and organizational goals.  A common scenario would include transactional and reporting systems on SQL Server and using Azure HDInsights for data not ready for more formal use, like web log data or EDI XML where you only store a portion of the message.  If you don't already have a Hadoop-based solution in your company, Azure HDInsights provides straight forward and cost effective way to incorporate Big Data technology.

Data consolidation, manipulation and restructuring into information
Once you've decided what data is ready for building goals and metrics around, it's ready for the Enterprise Data Warehouse.  The EDW has other names and can look very different between companies, but the consistent element is that it contains "one source of the truth" for reporting.  What the modern architecture enables here is taking data that is loaded in the EDW and allowing users to efficiently iterate and explore with the aforementioned unstructured data.  Tools like Power Query and Power Pivot in Excel are able to connect to disparate sources, combine them all into a single model and enable regular business users to conduct analyses in order to discover new insights.  By empowering those business users to explore data, the potential for new and innovative metrics increases exponentially. 

Information Exploration, Analysis, Presentation and Delivery
Finally, advancing the capability to present and deliver is the ability to explore and analyze information.  The modern architecture includes having analytical and visualization engines that empower end users to be able to explore without having to know code like SQL or Map Reduce.  There are no shortage of products and options here and it's up to your business to have discussions around what fits the need and the budget.  A common scenario could involve using Excel to build and share workbooks through a company portal, such as SharePoint.  To go further, open source tools for data analysis, like R, are becoming increasingly popular to perform in-depth statistical analysis and data visualization.

At Cardinal Solutions, we're enabling organizations to make better decisions.  The Modern Analytics Framework is just one of the ways we can make it happen.  If you are interested in ways to leverage new technology or simply want to discuss further, let's talk.



About The Author

Data Solutions Managing Consultant

Vinnie is a Solution Sales Specialist and Managing Consultant in Cardinal’s Columbus office. As an MCSE in Business Intelligence, he looks to enable businesses to make better decisions.