How many duplicated, dirty data pipelines are running throughout your organization? Probably more than you can count. Data-as-a-Service (DaaS) streamlines the chaos of ungoverned data pipelines and reporting silos created by users who are eager to use data, whether they are simple data inquiries from business analysts to more complex data questions from data science teams.

With Birst, you can create DaaS today so that data is used efficiently and consistently to support different layers of reporting and analytic needs – from a small centralized IT team to tens of thousands of decentralized users, with varying levels of data maturity and refinement. To put it simply, by setting up DaaS, Birst offers a better way to democratize data without sacrificing security, governance, and control.

Register for November 12 webinar, where a CIO explains how flexible and secure DaaS is delivered today with Birst.

DaaS is no longer something that analysts only talk about in the data management hype cycle; it is real and can be delivered today with Birst’s modern analytics platform. I spoke with Kevin Hodgkins, Vice President and Chief Architect at Birst, an Infor company, to explain what DaaS is, what challenges it solves, and what it takes to deliver.

 1. What is your definition of DaaS?

DaaS is a core component of modern data architecture. It provides a governed standard for accessing existing data objects and pipelines for sharing new data objects within an organization. Because it hides the underlying complexities of connecting to and preparing data sources, DaaS helps expand usage of available data. As data complexities increase over time, the demand for technical skills to manage that data increases. Without DaaS, as the complexity of data increases, it becomes increasingly difficult for non-expert users to generate value from the available data.

Because DaaS allows the reuse of data objects, new value can be pulled from data more easily. I really like to state that this is a faster time to repeatable value. This concept is opposed to what desktop discovery tools provide, which is value for the individual user only, and “quick time to first insight.” When we look at tools like Microsoft’s Power BI and Tableau, you must recreate complex data objects repeatedly across different teams and use cases. This is not conducive to ongoing and repeatable insights and value generation out of your data assets.

2.  You have been designing and implementing modern data and analytics architectures for organizations for most of your analytics career. How does DaaS fit into this architecture?

I believe very strongly in self-service. It’s common for people to say they can do self-service analytics by creating and customizing their own dashboards, reports, and visualizations. But that is only part of self-service. If you can’t get accurate data in the form that you need quickly and repeatedly, then true self-service is not really possible. Additionally, data needs are diverse across departments and roles; therefore, DaaS in the modern architecture must accommodate this variety of skill levels and data requirements.

3.  What issues does DaaS solve?

DaaS solves the tension between governance and agility so that there is repeatable value. A balance is needed between IT having control over most of the semantics while offering departments or individual user freedom of self-service. With the right balance, data object reuse and user productivity are maximized.

4.  With Birst, DaaS supports a wide range of data, reporting, and analytics use cases. What are they?

DaaS is all about allowing access to data that has already been dealt with, governed, and persisted. Data engineers and data modelers, probably within a central IT team, have often already taken the time to build and operationalize the organization’s most important data assets. This includes ETL processes and subsequent augmented and extended data sets. Those data sets or objects could be in a data lake and then used by an enterprise data warehouse, a polyglot enterprise data warehouse, analytic data marts, AI applications, or sandboxes. DaaS helps serve up solutions across all of the different use cases more quickly and repeatably.

5.  Can you give us an example of setting up data as a service for the business analyst? What value does this provide over other methods for acquiring data?

Customer segmentation is a great use case for a business analyst. The underlying CRM or ERP application may already have many customer attributes, but a business analyst may come up with different segmentation types based on a unique attribute they track outside of the core system. Traditionally, customer data would need to be exported from the CRM or ERP and then linked to new segmentation types in Excel. Now you have data sitting within a silo that will get out of sync with master data from the source system. Also, this new segmentation information may be useful to share with other team members, which will end up being emailed, creating duplicate out-of-date copies, and more silos.

With DaaS, we can enable the analyst to quickly use the pre-existing data, merge in their unique non-core data, and then share that new consolidated data object for reuse. Now anyone can analyze everything related to the customer, including the new segmentation types. This is important for repeatable value and maximizing speed of insight.

This functionality is available in Birst today.

Mona Patel works in Birst’s Product Strategy team. With more than 20 years of experience building analytic solutions at The Department of Water and Power, Air Touch Communications, Oracle, MicroStrategy, EMC and IBM, Mona is now growing her career at Birst. Mona received her Bachelor of Science degree in Electrical Engineering from UCLA.