Data and Analytics
Invigorating the Analytics Value Chain
Analytics has become a mainstream idea, as there is ample evidence that data-driven decisions tend to be better business decisions. However, simply integrating and governing data assets does not guarantee that business value is generated. Governance has to cover the analytics value chain and be integrated with an overall business strategy; using machine-learning to augment human intelligence and contextual awareness across this value chain will also help. Finally, we are witnessing a rapid increase in data privacy regulations across the globe, which will drive further governance and data management activity.
Demonstrating measurable business value through analytics to organization leaders will be inescapable
Analytics has now become pervasive in organizations, both in operational and in more general decision-making scenarios. However, the latter kind of solution, built on top of a data lake or a data warehouse gathering data from all around the organization, is a necessary but not sufficient condition to generate enterprise value. Not long ago, the main concern was lack of trust in the data. A first generation of data governance tools and processes introduced control policies, processes and accountabilities to transform data into a trusted corporate asset. However, this is not enough when data assets are too easy to copy around or to ingest in data platforms which are often poorly managed. Recent data governance tools address this data sprawl and disorganization problem by grounding themselves on a curated inventory of available, distributed assets (a Data Catalog).
A second, more fundamental problem is that simply having well-governed data assets does not guarantee that business value is being generated. Generating value from data requires a more intimate knowledge of the value-generating chain.
For this value to materialize, it is highly advisable to take into account:
- An expanded form of governance, called data value governance, covering the entire analytics value chain, from data and insights to people and business processes.
- An integration of this data value governance with an overall business strategy that aligns to a data-driven business model.
We are not yet at the point where data value models, metrics, and a supporting data value platform are universally understood, accepted and adapted to an organization’s needs. Nevertheless, the trend has started with an attempt to comprehend, catalog and publish the interactions between data assets, people, and business models and their supporting business process workflows, to better foster value creation from data. We foresee data value governance to mature based on stakeholder engagement with published data artifacts, derived insights and their use in action-based process workflows.
Machine Learning (ML) and Natural Language Processing (NLP) will accelerate the data to data-driven decision-making process
Central to the previous development is the use of machine learning automation to augment human intelligence and contextual awareness across the entire data and analytics workflow, from data to insight, and from insight to action. Two examples of this trend are
- Self-service data preparation platforms, where two different and complementary aspects are being developed to shorten the time in getting the data ready to train ML models. First, data preparation is being specialized by including new operations such as sampling, scaling, enhanced validation, and robust methods to deal with missing values. And second, ML technology is used to learn formats and schemas of data sets, and how users interact with data, to then recommend steps to expedite data preparation.
- Conversational analytics, backed by NLP technology, making the value generation process out of data accessible to all employees. Specific example technologies include smart searching with natural language (NL) queries, and NL generation to provide advice and next steps to the end-user in prescriptive solutions.
Privacy laws will soon be inescapable
The European Union’s General Data Protection Regulation (GDPR) has been attracting attention because of its far-reaching impact on data privacy and security. There are several similar regulations coming up now:
- California’s Consumer Privacy Act (CCPA) of 2018 mirrors many of the GDPR data protections for California residents and extends those protections beyond the ‘consumer’ role to employees, patients, tenants, students, parents, and children.
- Vermont’s data broker laws regulate businesses that buy and sell personal information to third parties, requiring them to register with the state, and mirror GDPR with respect to informed consent, opt-out, and security standards.
These type of regulations will soon spread globally. Indeed,
- Consumers have taken notice: In the wake of major data breaches and the Cambridge Analytica news in 2018, more than half the respondents in a recent poll expressed concerns about trusting companies to use their data properly.
- Policymakers continue to react: The EU Parliament has adopted a resolution for the suspension of the EU-US Privacy Shield, a data transferring agreement, seeking to bridge the respective data protection regimes, because it fails to provide enough data protection for EU citizens.
- Corporations are paying attention: Data privacy has become a board-level issue, as regulations are making businesses think about what data they collect and how they use it, not just how they protect it.
|Technology enabling insights for non-experts||Data governance based on a data catalog||Machine learning workloads in analytics tools||Data management automation for regulatory compliance||Management of analytics data assets with an analytics platform|
- Make sure your data governance processes allow your data to be trusted for analysis.
- Align your data with your business strategy. Understand how data and analytics impact your business goals and how applicable regulations impact your current data management function.
- Expand the reach and relevance of analysis by using technology to enable casual users to access insights more easily.
You May Be Interested In
Sameer Dixit, GM, Data, Analytics, AI/ML at Persistent Systems talks about the significant impact of AI & ML on Industrial manufacturing
Abhiram Modak discusses the benefits of data priming for a smoother RPA implementation
The potential for Automation & AI to raise throughput, increase productivity – explained