What is Data as a Service? The 3 Key Dimensions

In order to remain competitive in today’s digital economy, businesses must overcome challenges emerging from the steadily increasing volume and variety of accessible data. Strategic business decisions rely on rapid consumption and analysis of structurally diverse internal and external data that together provide multiple perspectives on business challenges.

Disconnected internal data residing in data silos must be combined and provisioned in order to contribute meaningfully to decision making. Enterprises are motivated to expose purpose-built analytics in support of speedy decision making. As a result, organizations would benefit from adopting an innovative approach known as data as a service to overcome these and other business challenges. Let’s begin with the basics.

What is Data as a Service?

Data as a service (DaaS) is a business-centric service that transforms raw data into meaningful and reusable data assets, and delivers these data assets on-demand via a standard connectivity protocol in a pre-determined, configurable format and frequency for internal and external consumption.

DaaS entails one or more functions, in any combination, such as collection, integration, enrichment, curation, contextualization, aggregation or analysis of internal and external multi-structured data. Further, DaaS makes the entire process of transforming raw data to end-product data (including pre-built KPIs and metrics) transparent to the end user.

Put simply, DaaS can be characterized as the following:

  • Business-centric, tailored to satisfy business needs for data assets.
  • Not a service-oriented technology, although such technology might be leveraged in the technical implementation used to enable DaaS.
  • Unsuited for one-off use cases but rather intended for broad consumption of reusable data assets.
  • Incorporates integrated governance.

DaaS Enabler, Consumer or Both?

Technically, an enterprise can be a DaaS enabler as well as a DaaS consumer. External data brokers (enablers) are in the business of making curated data available to enterprises (consumers). Marketing, weather, and research data are all examples of data made available through DaaS.

Enterprises can employ DaaS to make their data assets available in support of business initiatives for consumption by their internal business units or by their end B2B and/or B2C customers, thereby transforming them into DaaS enablers.

Business Challenges Facing Today’s Enterprises

Today’s evolving data landscape has spawned new business challenges that require innovative solutions. Sample situations include:

  • Strategic decision making, which rely on multiple perspectives such as social and economic factors which require combining internal and external data.
  • Focusing on business outcomes rather than on developing and managing an intricate network of data.
  • Accounting for the increased volume and structural complexity of today’s data, and increased frequency required in delivering data assets.
  • Coping with data silos which house data that must be combined and provisioned to support decision making.
  • Exposing purpose-built analytics such as supply chain for consumption in order to expedite decision making.

DaaS can overcome these business conditions from three dimensions: information-rich external data assets, connecting data silos, and enabling pre-built and packaged analytics. Looking at each element in more detail, DaaS helps organizations manage the following:

1. External data assets

External data assets provide an enterprise with additional perspectives in the decision making process. Combined with internal data assets, the enterprise benefits from greater context and more complete information overall.

External data assets are available to enterprises through subscriptions with DaaS enablers such as Data collectors and aggregators who are in the business of integrating, cleansing, enriching, and aggregating external data from web, mobile, open data sources, and data providers, and then making the resulting data assets available for consumption by organizations on a scheduled or on-demand basis.

2. Coping with data silos

Enterprises need solutions to cope with data silos which hinder decision making. Siloed data needs to be combined and then made available to both internal and external business stake holders to support analysis and decision making.

With DaaS, enterprises make data readily available to internal business users as well as business customers and consumers all while masking the complexity of connecting data silos.

3. Prebuilt and packaged analytics

Prebuilt packaged analytics solutions such as KPIs and metrics are needed to expedite delivery of domain-oriented analytics and data products which represent the end product of analyses.

Enterprises provision packaged analytics using the DaaS approach so they can build the analytics themselves for consumption by internal business users or external customers, or else leverage pre-built analytics offerings from vendors who enable DaaS. 

Data Virtualization Fuels Data as a Service

Ongoing innovations in data virtualization make DaaS possible along each of the three dimensions above. Data virtualization facilitates the delivery of external data assets, combines internal and external data assets, and provisions data for consumption through APIs. Data virtualization connects disparate data sources and provides unified access to siloed data assets through logical data models. Through logical data warehouse (LDW), data virtualization supports building measures and metrics such as prebuild analytics which can be accessed by downstream analytical tools.

Data virtualization is a virtual data layer that combines disparate data from a variety of source systems into complete, connected information and then provisions the information assets to business users through applications, services and reporting solutions. Data virtualization has grown to become a critical part of modern data architectures. Data virtualization can be exploited as a powerful approach to enabling DaaS, allowing organizations to reap the benefits of timely, relevant data to support decision making.

Plan Ahead to Alleviate Problems Later

As data as a service continues to gain in popularity, it is important to be mindful of a few concerns accompanying DaaS all of which can be alleviated through careful upfront planning. These include:

  • Data licensing and ownership issues: can be averted by clearly defining the data ownership and monetization model.
  • Non-resilient technical architectures: can be avoided by validating that your technical architecture supporting DaaS is sufficiently robust to accommodate changes in requirements.
  • Potential impact of business model changes: can be averted by ensuring that your business strategy is sufficiently flexible to accommodate changes in both the DaaS enablers and consumer business models.
  • Risk of encouraging retention and expansion of information silos: can be circumvented by restricting unnecessary data movement which can lead to additional data silos.

The rising adoption and increasing comfort with cloud-based data are fueling a new treatment of data across enterprises. DaaS’ benefits go deep, particularly in terms of increased business speed or agility. But let’s face it, enterprises will be using a combination of traditional infrastructure, and cloud deployments for some time.

Data virtualization plays a key role in this evolution as it unifies both structured and unstructured data in real-time to power analytical and operational use cases and to make DaaS more actionable. It delivers a simplified, unified, and integrated view of trusted business data – be it internal or external – in real time or near real time as needed by the consuming applications, processes, analytics, or business users. The result is faster access to all data, less replication and cost, and more agility to change.


Subscribe to Big Data Quarterly E-Edition