At the recent Business Intelligence and Analytics Summit in London, Gartner analyst Dan Sommer employed a clever analogy to illustrate the divide he sees in today’s analytics landscape. He called it “Suits vs. Hoodies”.
The “Suits” represent traditional, centralized BI. This model is typically associated with legacy BI platforms that support centrally-governed enterprise deployments, but have been losing market share in recent years to more agile data discovery tools – the “Hoodies” – adopted by decentralized teams looking for greater speed, ease of use, and autonomy.
The analogy effectively depicts a very real challenge facing organizations today: how to satisfy business users’ growing demand for discovery capabilities and less restricted access to information without sacrificing the governance that allows people to trust their data.
While it’s tempting to portray the “Suits” as stiff and old-fashioned, and equally tempting to portray the “Hoodies” as the fresh, new way of doing analytics, the reality is that the proliferation of data discovery tools inside an organization brings with it a greater risk of analytical silos and reporting chaos.
Gartner estimates that, by 2016, less than 10% of self-service BI initiatives will be governed sufficiently to prevent inconsistencies that adversely affect the business1. In other words, we’ll spend more time arguing about whose numbers are right instead of having meaningful business conversations.
But is this apparent conflict between “Suits” and “Hoodies” inevitable”? Must we choose between governance and agility? I posed this question on Twitter and was presented with a brilliant alternative: a “Hooded Suit” (tip of the cap to@NickJewell).
The “Hooded Suit,” in my opinion, is a better analogy because it symbolizes the great opportunity in the BI and analytics market today: the ability to bridge the gap between centralized BI teams supporting enterprise requirements and user-led decentralized teams looking for greater autonomy.
In a recent research paper, Create a Centralized and Decentralized Organizational Model for Business Intelligence2, Gartner argues that “IT leaders should create a 2-tier organizational model where the business intelligence competency center collaborates with decentralized teams” in order to ensure success with BI and analytics.
Enabling a 2-tier BI analytics begins with the right organizational structure, but having the right technology foundation underneath is just as necessary. Companies evaluating BI and analytics solutions should look for capabilities that support this model:
- Adaptive user interfaces that seamlessly transition between reports, dashboards and visual discovery
- Self-service data preparation capabilities that ingest, cleanse and organize data for analysis without IT involvement
- Native mobile consumption modes that supports offline interaction and responsive design
- Automation technology that can accelerate the traditionally manual process of extracting, transforming and modeling data
- Common and reusable semantic layers that ensure a single version of the truth and allow users to make decisions confidently
- Flexible delivery options that support cloud, on-premises, or hybrid deployments
By creating a 2-tier model, there’s no need to choose between governance and agility. In this model, “Suits” can be flexible and “Hoodies” can be accountable. The “Hooded Suit” is a smart and stylish representation of a world where central BI teams work in collaboration with decentralized business units. Now, if we can only do something about mock turtlenecks…
2 Create a Centralized and Decentralized Organizational Model for Business Intelligence, Gartner