Those of us steeped in the world of analytics have a well-articulated view of the modern analytics and BI workflow. You access data, optionally ingest it into an in-memory engine, prepare and enrich that data, analyze it anyway in countless ways, then share results with your team (or your enterprise) so they have insight into what drives your business. Address those functions, and you're done, right?
Not quite. From the first bit of data captured to the last action taken to achieve the outcomes you want, the data and analytics supply chain traverses a much more diverse landscape. As you architect your approach, you need to plan for an integrated and open approach to addressing the broader data and analytics supply chain.
Data is an ever-renewing resource. It comes in all shapes and sizes; it's stored in many forms. It can exist in the cloud, on-premises, in desktop files, from third parties, in many different clouds. Any data and analytics workflow starts at the data, and that data needs to be captured, managed, and integrated for myriad reasons…analytics being only one use case. By definition, open integration is key, as the data landscape is diverse and often messy. Any data architecture you choose must be open to data from any source, whether refined or raw.
There are many constituents in your analytics workflow:
There's a plethora of tools and technology these participants use to get their jobs done. Any analytics architecture must be able to use a variety of cloud services as needed to get their job done. Open integration is part of the fabric of the new analytics workflow.
To get to the ultimate business user, data and analytics can't be "separate but equal" to business applications. Analytics must be embedded and surfaced in the context of role and responsibility, used to plan activities both tactically and strategically, and finally, act on results and recommendations.
"Insight to Action" has long been a mantra for BI and analytics vendors. But action was disconnected from the insight process, often relying on (and reminding) others to do something based on what was uncovered. If your analysis senses that demand is outstripping supply for product in key geographies, you want to close the loop and shift inventory to those territories with higher demand while not shorting regions where inventory is in oversupply. Any end-to-end data and analytics architecture must be able to integrate to the business applications that drive the business, both embedding analysis in the workflow of the application and triggering action based on recommendations.
Workflow implies a beginning and an end. In reality, data generated as a result of taking action needs to be managed and integrated, analyzed, etc. It's a loop that repeats over and over. Data gets richer, insights get refined, recommendations get more accurate, action gets more targeted. When you plan for a comprehensive data and analytics workflow and the architecture to support it, it must be integrated and open, responsive, and dynamic to reflect the demands of an ever-changing digital business.
To learn how you can benefit from Oracle Analytics, visit