Denodo Senior Vice President Ravi Shankar says organisations must take a longer-term view on where their Business Intelligence strategy is headed as we enter an era of the digitalisation of everything.
By Ravi Shankar
More than 59 zettabytes (ZB) of data have been created, captured, copied, and consumed in the world this year, fuelled in part by the COVID-19 pandemic, according to IDC. With the increase in digital activities and employees working from home, richer mixes of data sets have been created in tandem.
While there is no shortage of data out there, the ability to access and act on it whenever, wherever, and however you need is something else altogether. Here’s where business intelligence (BI) modernisation comes into the picture.
Against this backdrop, BI projects have risen in priority for many businesses at present, as they seek to empower better data-driven decisions and take informed actions. The adoption of BI and analytics tools, however, does not guarantee success for businesses seeking to become more data-driven. In fact, multiple ingredients are necessary to define success for the enterprise, particularly as businesses embark on the next phase of BI modernisation – one that embraces cloud technologies and artificial intelligence (AI).
Faced with this situation, one must ask what the most common mistakes in the use and modernisation of Business Intelligence are.
Lack of collaboration between business and IT
A holistic, strategic view of BI outcome is necessary for mature and successful BI implementations. This view should be benchmarked against data collected across all business units. Clear roadmaps should be drawn up in the boardroom, collectively defining milestones and performance measurement thresholds for both business and IT teams.
Without this, business and IT teams tend to approach BI modernisation in silos, as standalone projects. Such is understandable, as at first glance, both teams appear to approach BI with opposing processes and outcomes in mind. From an IT perspective, BI might be synonymous with presenting a view on data. A more elaborate way on accurate reporting. To a sales manager, BI may be associated with information and intelligence that can help drive sales and marketing tactics in a timely manner.
However, it is important to think longer term. While IT teams have traditionally been tasked to prepare data for analysis, the expectations for faster analysis and interaction with data should ideally shift processes to self-service BI.
Enabling business users with self-service tools to access data will increase the efficiency of modern BI and ultimately frees IT to work on improving availability, interoperability, security, and overall corporate agility of data.
Not providing for data governance agility
Data governance requires a formal programme to be in place. A programme that guides processes so that businesses comply with governing standards for the collection and use of data, to protect sensitive data from landing in the wrong hands. However, most BI solutions present serious regulatory risks as they lack mechanisms to identify or classify sensitive data, making it easy to download and export data.
Increasingly, users demand agility in data governance to quicken the process of accessing data and its relevant information, from where they are located to who has access to them. Swiftness is key in setting up rules about who can access what data, even down to the level of the individual cells. This is critical particularly in modern deployments that leverage complex distributed data sets with large numbers of distributed content authors.
Ideally, successful BI modernisation should incorporate data virtualisation and machine learning (ML) technologies in its data catalogues to enhance and automate search features. Data virtualisation provides built-in capabilities for easy control by not replicating or storing source data. The metadata is used to create virtual views of all data in the catalogue, from its location and availability to full data lineage and auditing information. It also provides a central access control point to set up data governance rules. Meanwhile, ML capabilities automate repetitive tasks and make recommendations on next steps base analysis of usage patterns.
Not adapting to a cloud-based future
Businesses have been shifting data storage and workload to the cloud. Leading cloud service providers such as Alibaba, Amazon, Google and Microsoft are competing to add cloud zones across Asia Pacific, and data centre investments continue to surge in the region.
Successful BI modernisations will require agile, real-time data integration that can span multiple cloud data sources and on-premise systems. Once again, data virtualisation presents itself as a trusted methodology that can help BI systems meet this pressing need. By delivering data transparency, data virtualisation establishes a hybrid data fabric to provide seamless, real-time access across cloud and on-premises system.
Simply put, modern BI applications need to be able to automate the balancing of workload between systems and the cloud, and free up data workers from the need to sort data manually.
Modern BI is more than a single application; It is an entire infrastructure of technology, people and processes working together.
Businesses need to take a longer-term view on where their BI strategy is headed, as we head into a an era of digitalisation of everything. We are currently in the next phase of BI modernisation, one that sees businesses embracing artificial intelligence, cloud technologies and progressing in tandem with a digitalising economy.
BI will need to provide businesses with faster and more accurate analytics and insights. As part of this progress, data management will be further streamlined through incorporating advantages of AI and ML, to augment much of what are now manual processes.
(Ed. Featured image courtesy of Fauxels.)