In today’s digital world, analytics plays an important role as an emerging competency with varied capabilities and has come a long way from traditional reporting and data presentations, thus transforming the technologies today into a more robust models, algorithms and data pieces. What’s more, with the evolvement of new Analytics, BI, Data Science, Data Management platforms the battleground for having a common process to deploy, re-purpose and secure an effective solution has never been so complicated and competitive. An enhanced analytics governance framework would help track the common issues around various practices and policies revolving around our data and analytics and come across a set of guiding principles.
Governance today is a critical element around your data and analytics capabilities and can be said easily to be a set of guiding principles for streamlining the work of managers, analytics and data management practitioners to work towards a streamlined approach, thereby create a function for face major challenges surrounding our day-to-day work structure, from human capital to structure around broader enterprise to legal and regulatory practices.
“The Pareto Principal of Data Science” states, that data scientists spend around 80% of their time finding, cleaning & recognizing data & rest only 20% time on data analysis.
Now the question that comes to the mind is why we need to align with Analytics Data Governance and the answer lies within.
I simply believe that addressing the most restless V’s is the solution to what we look forward to. I picked here the top five V’s that resonate with my finding’s here.
Breakaway companies scale analytics by outperforming the data driven decisions across the critical layers, but mostly they fail to comply with some of the extraordinary increases in the types and amount of data that are being generated and the useful perspective around the ideal governance framework is somehow missed.
- Same report with different results
- Data shared cubicle to cubicle
- Inconsistent and inappropriate data
- Revalidation of data
- Multiple data marts house the same “golden result”
- Tracking data origin is a forensic exercise
Companies might plan and embed the analytics code across each layer of the organization to meet their global needs and capture a portion of the $9.5 trillion to $15.4 trillion of value which McKinsey Global Institute estimates advanced analytics can enable across industries globally. But with huge investments as part of this strategy, are they able to benefit or rather scale the analytics to real value? Though it looks promising, it still needs attention.
Dave Wells through his article “Technologies for Data Governance” presented the emerging issue around traditional data governance quite handsomely bringing self-service as an emerging area to offer solutions.
A distinct analytics governance framework is something which can address all the complications and issues surrounding the processes around the data and analytics function. These may include, setting and enforcing policy, data maintenance, architecture and technology adoption and integration, legal and regulatory environment, etc.
Now that we have better understanding around “Analytics Governance” and how it functions, it is also important to understand that the framework for establishing the policy, strategy and objective is grouped under three basic models:
- Algorithm governance
- Model governance
- Reports governance
When we better plan to align with the above models, we must ensure that the centralized and de-centralized components are also taken care of.
One thing we need to understand that data and analytics will continue to serve the businesses with various customer experiences and results. And to have the analytics function accurately, the governance framework must be designed in such a way that it covers the below key processes.
Process 1: The “Know” Quotient
So important, when your entire decisions and forecasting is based on analytics, especially knowing what business outcomes you are expecting and how your KPIs are calculated. The know layers can exist around the data, the sources, the expectation and the ultimate results.
Having an answer to the above “KNOW” areas, can help address the foundation of an organizational model and act and play accordingly. In short, it provides a definite path to the scope we are building.
Process 2: Structuring the Framework
The second most important process around self-service governance will be around structuring and building the framework. Starting point here can be assigning role-based persona from what we have learn in Process 1 above.
Process 3: Persona based activity
Now that we have created the framework around four different types of personas, it is important to assign activities based on each persona. And here roles and accountability can be assigned for each persona to function as desired.
Now, here the persona has been based on four major critical level in the organization, also based on what big organization and practitioners recommend. However, this can be framed or created differently how each organization or client foresee the expectations from data analytics governance. The idea here is to at least create a structure based on the objective to plan and achieve the required outcome.
Process 4: Role modelling and Interactions
Once the activity part is decided, next is to define the models based on each persona and plan about interactions between different personas.
One of the major interactions that we see normally is drafted below that can typically respond to how you plan to structure your self-service governance. With the interactions, what it provides is the layer of responsibilities and decisions rights lies with each holder.
2019 has shaped up to be the year where organizations are back with intense pressure and ask to get the data and analytics run with best of measures and results and one of the models the organization today are seamlessly planning to have is self-service governance framework.
So stop worrying and start planning around aligning with the new structure, build the infrastructure to cut through the internal challenges and win the global situation by being top in the Analytics and Data provisions.