The need for organisations to have access to trusted and reliable information for sound decision making has not changed since Data Warehousing became topical in the 1980s. In fact, almost 30 years later the need is far greater.
Organisations are dealing with a plethora of data from a variety of data sources, including streaming data, IOT, or even the ‘Internet of Everything’ related data and can you believe it, Flat File data as well. Oh yeah, Excel is still in the picture, fortunately or unfortunately. Historically, data management teams spent many a late-night drinking flat coke and eating stale pizza making every effort to ensure that the relevant data sources have been identified, that the data source connectors are performing and ensuring that the data is extracted, transformed, and loaded into a data management system optimised for user access through specialised Business Intelligence, Reporting and Analytical tools.
Now, the need for near real-time access or even real-time access to that sourced and managed data puts the entire process at risk. Users want, no demand access to up to date data to enable them to make informed business decisions. The idea of ‘Data Batch loading windows’ need to be reconsidered considering modern data management techniques and user requirements.
It is key to start with a foundation based on sound data management principles aligned to data governance practices to enable business innovation and growth, one that promises valuable market and customer insights. Yet many organizations struggle to determine where to begin with a unified data and analytics initiative.
Organisations need guidance and pragmatic steps to unlock the value of enterprise data management and AI driven analytics. A framework should be considered to identify and target investment areas based on desired business outcomes and identifiable analytics patterns.
Where is the best place to start?
Modern data governed organisations should build a data and analytics strategy taking into consideration the enterprises data literacy maturity level of the organisation, supported by a well-structured data governance framework. One important starting point is to identify the appropriate investments needed to build on your data and analytics experience with the goal to provide self-service capabilities that can be employed with little to no IT support.
Modern Data Management must be based on the ability to integrating data across the IT landscape with the aim to provide data consumers with intelligent, relevant, and contextual insights for informed and insightful decision-making augmented by Artificial Intelligence.
The main idea behind modern data management is to Process distributed data. Data Architects and Data Integrators need to evolve data integration to data orchestration by using an enterprise data fabric.
This enterprise data fabric for enterprise-wide data management should enable the ability to easily discover, classify, profile, understand, and prepare all your data through an enterprise data catalogue. The outcome of this is to help organisations manage, govern, integrate, and optimize enterprise data.
Let the past, be a foundation for your next step. A lot of emphasis was placed on visualising data. The emphasis of traditional Business Intelligence Tools was on visualising the data. But focusing only on data visualisation and trying best to interpret the data can lead to un-insightful decision making. Making decisions without AI driven data insights can be like putting lipstick on a pig. No real outcome other than beautiful dashboards with spinning charts and fancy colours. Only focusing on the presentation of data and not the quality data has its own repercussions. Garbage in, Garbage out. A joint effort between business and IT must be established to ensure data accountability. Accountability must be in line with your enterprise data governance framework and supported by a data literate user community. With this approach joint approach, supported by investments in unified data and analytics platforms, the organisation can develop required data and analytics capabilities for managing data and driving the right organisational growth, efficiency, and financial impact.
To use data-based insights augmented by AI to make fast, well-informed decisions, organisations should evaluate a unified, business-centric platform with an open architecture that turns data into business value. This unified, business centric open platform should offer compute resource flexibility, the unified data management and analytics platform must support accelerated organisational outcomes through data orchestration, data management and data-based insights augmented by AI. A platform likes this will lead to data-to-value outcomes across the organisation.
Modern data-to-value platforms should be cloud based to capitalise on all the benefits of cloud. Allowing for interconnectivity to Off-Cloud (i.e., On-premises) systems in a hybrid and modular data architecture. Gone are the days of hardware provision, with a 5-year lifecycle and recuring procurements processes. When you need more compute power to support your data-to-value strategy, it should be available as and when it is needed.
ABOUT THE AUTHOR
Shaid FA Greeff
SAP Business Technology Platform Solution Advisor (SAP Data Warehouse Cloud, Data Intelligence, Analytics Cloud)
SAP