Business Intelligence models are frameworks that determine how data and analytics capabilities are managed and utilised within an organisation to support decision-making processes. The choice of model can significantly influence the agility, accuracy, and timeliness of insights derived from data.
The Centralised Business Intelligence Model
In a centralised Business Intelligence (BI) operating model, a designated individual or team within an organisation is responsible for providing insights, dashboards, and data reports. Typically, the demand for these reports far exceeds the capacity of the BI team. A single BI professional might be trying to meet the needs of hundreds of colleagues, leading to significant backlogs.
This bottleneck forces the BI professional to spend a considerable portion of their time – often up to 60% – on maintaining and running reports. This leaves only 40% of their time for developing new reports and insights, making it challenging to keep up with the constant demand. End users, therefore, lack the ability to generate their own reports or perform data analysis, relying entirely on the central team.
Limitations of Centralised Models
Centralised BI models often result in inefficiencies:
- Skill Limitations: BI professionals may excel in specific areas, such as report generation, data engineering, or analysis, but finding someone proficient in all areas is rare. Each individual’s strengths and weaknesses impact their ability to meet the diverse demands of the organisation.
- User Dependency: Business users are dependent on the central team for their data needs, which can stifle their ability to make timely decisions and force them to resort to less sophisticated methods, like using spreadsheets, which are often substandard.
- Scalability Issues: This problem is not limited to small organisations; even large enterprises can experience these bottlenecks. For instance, a single data professional might be responsible for an entire department, causing delays and inefficiencies.
- Key Person Risk: Relying heavily on one or two team members for data or reporting creates operational risk if they leave the organisation.
- Business Acumen: Centralised teams tend to be further removed from the business and therefore lack business knowledge.
Technological Advancements: A Shift Towards Distributed Models
With advancements in cloud technology and BI tools, it is now possible for non-technical users to generate their own reports and insights. Tools like ThoughtSpot and Power BI are integrating natural language processing capabilities, allowing users to create reports conversationally. This shift enables a distributed BI model where technology acts as the central platform, reducing the dependency on a centralised BI team.
The Role of the Central Team in a Distributed Model
In a distributed BI model, the role of the central team evolves:
- Data Curation: The central team ensures that the data is clean, curated, and properly labeled. This is crucial for accurate report generation and decision-making.
- Governance and Standardisation: Even in a distributed model, governance is essential. Defining key metrics, standardising definitions, and controlling data access are critical to maintaining data integrity and security.
- Support and Training: The central team provides on-the-job training and real-time support to help end users become self-sufficient in using BI tools.
The Balanced BI Model: Finding the Middle Ground
A balanced BI model strikes a compromise between centralised and fully distributed models. In this approach, a select group of end users have full access to BI tools to create reports and perform analysis. This model still maintains governance, standards, definitions, oversight, and compliance centrally, but it allows for quicker development of insights by enabling some business users to develop their data analysis skills.
Preparing for the Future: Embracing Gen AI and Upskilling
As Generative AI (Gen AI) becomes mainstream, organisations relying on centralised BI models may struggle to keep up. To stay competitive, companies need to shift towards a balanced model and eventually a more distributed model. This requires investing in modern BI technologies and upskilling employees to create their own insights.
Planning the Transition
Organisations must assess their current BI operating model and identify barriers to moving towards a more balanced or distributed approach. Key elements for this transition include robust technology, good data governance, and comprehensive training programs. Embracing these changes will enable continuous innovation and improvement, empowering employees to make data-driven decisions and fostering a culture of agility and responsiveness.
Are you ready to transition to a balanced or distributed BI model? Do you have a roadmap and the necessary resources to support this shift? With Gen AI revolutionising the landscape, now is the time to act and ensure your organisation remains at the forefront of business intelligence innovation. If you like to explore this topic further, feel free to reach out to us on sanan.thamo@ingrity.com
