- Enterprises must change the way they work to overcome the limitations of:
- Siloed systems
- Data integration problems
- Resources and skills to tackle them…
- … In order to look up to their partners to help:
- Industrialize their Analytics capabilities by creating an Analytics strategy.
- Build an operational framework.
- Select the right Analytics tools/technologies.
- Select the right people with appropriate skillsets.
“Being Data Native is the core for enterprises wanting to embrace digital”
Data Governance mesures are the key issue here (see Data Governance below).
- Companies need a data layer that can sit on top of all their multi-cloud and hybrid cloud architecture.
- With this data layer architecture, firms can deploy part of the data layer on premise and move the data in and out of any cloud at any time, with no downtime or re-writing of their applications.
- Instead of spending time validating and fixing data errors, companies should focus on their core mission.
- Graph databases are emerging as an important tool for finance firms.
- Two complementary ways to add value:
- Managers of SMEs, can build competitive advantage by better integrating social media channels and BDA, realizing the super-additive value from social media Analytics, and growing on steroids in the marketplace.
- Managers of Large Firms can combine their complex use of social media channels and BDA, in order to utilize social media Analytics to enhance market performance.
- They can realize the super-additive value from the complementary use of both of them, if sufficient integration efforts are made. Read more
- Agile Data makes it possible to quickly and easily:
- Add new Data Processing, Machine Learning (ML) and Analytics use cases to support the development of business models and initiatives, regardless of data volume, variety or velocity.
- Boost development data pipelines from growth
into production to potentially multiple different execution environments (on-prem, cloud, Hadoop, Spark, etc.), without re-coding the business logic.
- Handle changing upstream data source changes without manual intervention. Data pipelines and Analytics engines need to be able to automatically deal with upstream changes, like adding or deleting a column.
- Rehearse data engineering, ML and Analytics pipelines. As business logic changes, the data pipelines must be easily evolved and iterated.
- Manage the operational environment for large numbers of data processing, ML and Analytics use cases.
- Agile operationalization of data pipelines, referred to as “DataOps”, is as critical as the development process, in order to scale the total number of these.
- Companies should broaden their Data Governance program and shift their focus beyond just governing data.
- Metadata should be the center of the Data Governance effort, as understanding the context of the data content is the central concept of data stewardship. Read more
To create strategies and build a data-driven culture, senior technologial executives and senior leaders need to work together.
Roles & Responsibilities (here’s a detailed example of how an R&R structure for Data Governance in a company would look like)A) Support Levels:
Data Governance Team (DGT):
- Data Governance Chair.
- Program Administration.
- Data Governance Council facilitator.
- Advisory from other levels.
Data Governance Partners (DGP) Regulatory & Compliance:
- Project Management Office (PMO).
- Information Technology (IT).
Staff includes Application Development, Data Design, Security and other Data Resource Management.
B) Operational Level
- Operational Data Stewards:
- Managers and individual contributors, Data Definers, Producers, Users… These people are presently defining producing and using.
C) Tactical Level
- Data Domain Stewards (DDS): Directors and Managers Identified Subject Areas.
- Data Steward Coordinators (DSC): Directors and Managers Per Business or Functional Unit.
D) Strategic Level
- Data Governance Council (DGC): Senior Management-Vps and Directors of Group of similar participation
One person, plus alternate, for each Business Unit represented & IT.
E) Executive Level
- Senior Leadership Team (SLT): Senior most level knowledge of the program.
- Leverage existing business structure as possible.
- If no structure exists, use a Steering Committee or an Executive founded Board or Council
- Analytics to be useful should be Prescriptive; but it can use statistical or modeling techniques that are Descriptive or Predictive for instance.
- Providing a churn propensity*, for instance, without taking it into the context of a decision rule on a Prescriptive action is empty.
- It is not actionable insights that should be the holy grail of Analytics but the conversion of these insights into effective actions.
- Conversion: rule or system of automated learning based on past actions by customers (historic record) to generate Prescriptive solutions for firms to take action. In other words, from Predictive to Prescriptive.
- Analytics professionals should still be the flag bearers of that conversion process, rather than purely relying on it from an organizational perspective.
- Mckinsey recommends firms to:
- Start by identifying the decision-making processes they could improve to generate additional value, in the context of the company’s business strategy…
- And then work backward to determine what type of data insights are required to influence these decisions…
- And how the company can supply them.
* Note: “Conversion process of insight into effective action”: rule or system of automated learning based on past actions by customers (historic record) to generate Prescriptive solutions for firms to take action. In other words, from Predictive to Prescriptive.