Decentralized business intelligence definition

An analytics leader at a Healthcare company recently lamented that because inefficient queries are such a drag on their data warehouse, a group of his DBAвЂ™s are forced to spend a better part of their day killing bad queries. This is a vast improvement from the вЂњcollaborationвЂќ which often occurs in top-down analytics organizations. Instead, engineers are embedded alongside analysts within autonomous business teams organized around specific business objectives. ItвЂ™s telling that Amazon, one of the most successful analytics companies in the world, has no central analytics team. In this new world, successful analytics organizations focus on outcomes, then organize & empower cross-functional teams around those outcomes with as few dependencies outside their team as possible. The future of analytics is decentralized, empoweredвЂ¦ and collaborative.ДЄn organizationвЂ™s analytics program must keep up so business teams can respond quickly to new data challenges in order to drive outcomes.Д¬ollaborative analyticsВ recognizes that the experts on the business are those sitting on the front lines, and as such, the overall structure of an organizationвЂ™s analytics must be organic to take advantage of this knowledge and evolving business needs.Д¬ollaborative Analytics brings engineers and analysts together to deliver outcomes and increase competitive advantage. The analytics function must also evolve to meet the needs of the business and realize the benefits of this new data mesh architecture. The capabilities and economics of capacity-on-demand have changed the game. Highly elastic data platforms can be scaled up or down within minutes or seconds. The old requirements for pre-defined server capacity have given way to dynamic cloud-based scaling.

The old world of expensive, fixed server architecture demanded it. When analysts work in isolation, you end up with disconnected, inefficient, inconsistent data silos representing hundreds of versions of the truth.ДЇor all the drawbacks of centralized analytics, this approach was formerly the only option. ItвЂ™s also the reason companies are plagued with orphan analytics. This is the reason Excel is the most popular business tool on the planet. In a top-down engineering environment, analysts are often forced to go it alone. To the extent an analytic model is disconnected from the evolving business context, it becomes little more than an intellectual exercise.Д®ven so, the work must get done. The opportunity will evaporate before the necessary data or analytics make it back to them. ThereвЂ™s simply no way a centralized data engineering team can understand and build all the analytic capabilities every team in the business needsвЂ¦nor can it be fast enough.Д«usinesses canвЂ™t afford to wait weeks or months for the centralized engineering team to approve, scope, build, test, and deploy requested changes. Centralized analytics moves too slowly to respond to an evolving set of business needs.

Decentralized business intelligence definition