Analytics Need to Provide Business Value

As a part of my ongoing conversation with senior executive leaders, Conversations on Big Data, I recently sat down with Dean Silverman, Senior Advisor to the Commissioner, Office of Compliance Analytics, at the Internal Revenue Service (IRS). These Big Data conversations tend to focus on best practices in using big data and analytics in creative and interesting ways to support real-world business goals.

Dean joined the IRS in 2011 to build an advanced analytics program. According to Dean, for analytics to be relevant, analytics programs have to provide practical outcomes and measurable business value. “There’s an endless amount of data…but what good does it do for me [if it does not lead to] any meaningful, actionable strategy?”  Furthermore, he believes that “analysis is important … but if it’s not overlaid with problem-solving, you’re going to spin around and answer the wrong question”. Since the capabilities needed for analytics often require considerable investment, a strategic goal associated with specific outcomes provides a framework to measure that future return on the required investment.

In developing the IRS’s Office of Compliance Analytics (OCA), he describes the team as:

1.       Empowered to ask the hard, sometimes uncomfortable questions

2.       Helps structure and participate in major change programs

3.       Support the core business in framing the problem

4.       Uses data that exists to find a solution

Dean feels strongly that an analytics program depends on executive sponsorship, to provide the above “secret sauce”. Those champions need to engage key stakeholders to identify the problems that are currently keeping the organization from accomplishing its goals now, and promote the inclusion of “analytics as a way of doing business”. Specifically, this approach helped to frame Dean’s primary objectives for analytics:

1.       Reduce fraud and improper payments, primarily through Identity Theft and the Earned Income Tax Credit (EITC)

2.       Reduce the tax gap

3.       Identify core compliance enhancements

Dean and his team started by reaching out to the operating and IT divisions to “partner hand and glove … in agreeing on what the problem is, solving it, [and] …then measuring the results.” He considers this the “embedded partner approach”. To be successful, an outsider who has been “embedded” in the organization by a senior leader needs to create a sense of mutual trust. It is critical to convince the internal partners that you are there to help focus on their results by leveraging their expertise and understanding of current process and available data.

In our interview, Dean provides two examples on how the embedded partnerships within an organization work successfully. To learn more about Dean’s successes, listen to my complete interview with Dean at http://www.ourpublicservice.org/bigdata/.

Leave a Comment

One Comment

Leave a Reply

Catherine Andrews

Thanks, Brian! I love these real-world examples and interviews you share. It really grounds the big data topic for me.