Q: For readers unfamiliar with Peloton and its services, how would you best describe them?
A: Peloton is a nationally recognized leader in Business Analytics, Enterprise Performance Management, and Cloud Solutions. We partner with clients in the planning and execution of business and technology initiatives focused on driving insight and analytics.

In addition to our EPM practice, we have a variety of BI and Analytics services and offerings for our customers.  We provide expertise with data governance and master data management programs as well as strong development and architecture capabilities with data warehousing and reporting platforms.

Q: What is the chief business challenge your customers are trying to solve?
A: It’s really varied. There’s tremendous interest from our clients in how to move their analytic capabilities up the analytics continuum from descriptive reporting on one end to fully prescriptive analytic capabilities on the other. When we get started on an analytics initiative, we often begin by baselining where their current analytic capabilities are and where they need to be. One size doesn’t necessarily fit all here. We look along multiple different dimensions of people, process and technology to see if there’s a gap they need to address. Some clients have opportunities to improve their governance processes; some need to go look deeper into their technologies and tools.

One common finding is that although there is a lot of interest in big data and predictive analytics, there are still significant challenges with data governance and data quality at many organizations. Big initiatives in building enterprise data management capabilities are starting to pay off but there is a continuous stream of new data to integrate and the desire to analyze the data in new ways.

Q: What are some specific examples of how you’ve seen companies take advantage of Big Data?
A: Many companies are still trying to understand how best to get started with Big Data. We’ve seen some good use cases for using a Data Lake or Data Reservoir for staging both structured and unstructured content. This pattern can provide a lot of potential value for organizations looking for a cost effective way to store large volumes of varied data for discovery and analysis. It is a good complementary approach that can let organizations leverage their existing investments in their data warehouses and reporting tools with the capabilities the Big Data technologies enable.

Peloton is an Oracle Platinum Partner. We are excited about Oracle’s Big Data Discovery product. This product lets users explore data on a Hadoop platform visually and transform the data so that it can be made available for further analysis or reporting. The visual approach will make the data on the Hadoop platform accessible to more users than before. We see a lot of potential use cases for this. It’s a great complement to a data lake and should speed up the process to get insights into decision makers’ hands.

Q: Where are we in regards to attaining true “Self-Serve BI”? Are your customers having success?  Why or why not?
A: There has definitely been a lot of energy around this topic for clients. There is a huge appetite to enable self-service, but not everyone always agrees on what it is. Some think that it’s as simple as just giving users a dashboard and letting them select their own parameters or giving users access to an ad-hoc reporting tool that is pointing to an enterprise data warehouse. Of course true Self-Serve BI is much more than this. We see the opportunity here really present itself when we see just how many business users are using their dashboards or ad-hoc reporting tools as ETL tools that they use to dump data into Excel, Access or other desktop tools where they can really do what they want. It’s there where they blend in new data, create their own metrics and hierarchies and present the data the way that they need to. This is how Self-Serve BI has been getting done in places where it isn’t being addressed directly.

Now organizations have more options in this space than ever before. Self-Service BI tools let users interact with and present data much more intuitively and let them access all sorts of different data sources outside of the enterprise warehouse. There are also new user focused tools that give users the ability to perform ETL without going through IT. The adoption of these tools is growing and users are expecting these types of capabilities from all their BI tools. However, there is a real legitimate risk to this path. Organizations still require data quality, data governance and security. The challenge for customers is finding a way to balance the need for analytic agility with that of quality, governance and security. We see this as being a real focus area for our customers. We think that to really address self-service effectively, customers will need to look at optimizing the people, processes and technology in an organization to better support self-service BI.

On the people side, organizations will need to understand the roles and personas of their users. Not everyone is a data scientist who needs the ability to mash up data from a variety of sources and perform complex analysis on it. On the other hand, not everyone is going to be able to answer all their questions from a set of static dashboards. Organizations need to understand and define who needs to do what and what their information needs are.

Organizations also need to develop processes to support their analytic needs and provide a path to promote analytic content from sandbox and discovery areas to secured and governed areas like standard reports and dashboards.

For technology, it’s important not to overextend a tool. Organizations should utilize complementary tools appropriately, according to the role and function of the user.

Q: What are you seeing on the data visualization front?  What is driving increased focus in this area? What products are users adopting?
A: The newer user focused visualization tools have really impacted the way that users look at BI tools in general. Desktop tools like Tableau and QlikView have enabled people outside of IT to create incredibly insightful visualizations very quickly and easily. I think there are a few things driving the increased focus. The first is simply ease of use. These tools are designed to be very user friendly. All the way from the installation process to building out your first visualizations. Users can get going in a few hours and do it with little to no support. Another huge driver is around the power of effective visualizations. The types of visualizations these tools create are very powerful and let users easily tell compelling stories with their data. The visualizations let users quickly answer questions with large volumes of data that would be difficult and time-consuming otherwise. The interactivity of the tools is really powerful and, again, very intuitive. Finally, the variety of data sources these tools support is impressive. Users can connect to flat files, spreadsheets, nearly any relational source and many of these tools also support big data sources like Hadoop and Hive, as well as Cloud data sources. All these make it very easy for users to tell powerful stories, quickly, using almost any data they can get their hands on.

The challenge most organizations have is on how to best integrate this capability into their broader BI and analytics strategy. Understanding where tools and capabilities overlap and working through the trade-offs between enabling end users and having a proliferation of different (potentially inconsistent) views of information is something organizations often struggle with.

About Dave Harter: Dave Harter is a Senior Director at Peloton and leads their Business Analytics Practice.  Dave has over 15 years of experience leading Business Intelligence, Analytics and Data Management initiatives. Before joining Peloton, Dave worked at Fidelity Investments where he oversaw multiple Business Intelligence and Analytics initiatives.