With Analytics, It All Starts with Data

In a recent conversation with a large global firm planning their analytics journey, we quickly fell into a conversation on significant hurdles to overcome. While I expected to hear about machine learning, data science, augmented analytics, and the inevitable cultural barriers any change brings about, the conversation quickly focused on one issue: DATA.

The common denominator is Data

Back in my industry analyst days at #Gartner and AMR Research, I used a mental framework – a bit like a "hierarchy of needs"—when it came to analytics.

1. Data professionals can (and do) talk about data on its own. They talk about managing it, storing it, manipulating it, governing it, etc.  But quite often, that’s where their concerns end. 
2. Analytics professionals must be conversant with both analytics and the data that’s used to support the high value use cases.  They are focused on the sources of data, but more importantly, on the uses of data to drive business insight and action.
3. Analytic Applications managers– such as Sales, Marketing, Finance, HCM, Operations, etc.—must consider the data, the analytics, and the applications use cases and processes necessary to mesh data and analytics to solve very specific domain challenges. 

While I have categorized the hierarchy of needs by role, it really is meant to be a call to action that enterprises of any size must take a holistic view of both the data it captures, but also how that data is consumed and turned from raw material to finished product – facts, insights, recommendations, and action. 

Why Data is often a contentious topic

Back to the customer conversation—the data architecture they’ve established has been in place for nearly a decade.  There are myriad sources of data – from applications in the cloud, on premises, in other clouds, from a data lake—and they’ve built a strong governance process to ensure a “single source of truth”.  But it takes them a long time to create this source; so long, in fact, that it’s become very difficult for business consumers to access the data in their normal work day. 

Reports, dashboards and analyses response times are not acceptable, and in their own analysis, they’ve determined that the challenge stems from that data model. Yet, the data professionals, who manage the data architecture, want to continue to use what they’ve built, as they have confidence in the content, but recognize its limitations. 

Since “Analytics” is the manifestation of the data to business users, to them it’s the Analytics that aren’t working. They advocate a fresh approach – bringing in new data processes, new analytics, new modes of interacting with data.  

Sound familiar? This is how self-service analytics took root in many organizations over the last decade—with the need to let business analysts and users drive data and analytics innovation, and not be constrained by architectures that are rich but rigid. 

Modern Analytics and BI is the right framework for the future

Analytics freedom needs to be balanced with business governance and policy concerns.  One size - highly governed and managed data – does not fit all. The modern business needs both freedom and a framework.  

As we talked about how to think about the future, we discussed:

• Opening up new sources of data to analytics to enable a whole new community – the data science analysts – to access a massive data lake and the governance corporate data. 
• How new data sources can be created to enhance the corporate model and provide more flexibility to managers and analysts to access what they need when they need it.  
• Using Augmented Analytics – natural language query and generation as well as machine-learning driven analysis to accelerate understanding and lead to insightful action.        

By the conclusion of our meeting, there was a renewed understanding that constraints of the existing data framework can be resolved with fresh thinking, taking into account the applications, the analytics and the data together, not as isolated pieces.

Three Takeaways:

1. Take a holistic view of both the data, but also how that data is consumed and turned from raw material to finished product.
2. Don’t be constrained by what you’ve done. Plan for what you need to do to support business needs.  Otherwise, teams will go off and do their own thing, which can lead to more challenges in the long-term.
3. Embrace Modern Analytics and BI to bring a fresh perspective on enabling freedom of analysis, with a framework of governance to ensure data and analytics consistency.