How to Build CoE Analytics

Analytics is a journey that more and more organizations head on, there’re road maps have been outlined, shall they also build Center of Excellence in sharing resource and govern practices? What do organizations need to consider if they are thinking about building their own internal Advanced Analytics Center of Excellence and what should be the measures of success?

1. A defined purpose for the CoE is key, measure success against those goals that support the purpose and mission of the COE. A good Cost benefit analysis, some quick hits and a motivation for folks to utilize the COE are key factors as well... 

2. Most CoEs start with and revolve around a solid methodology: from strategy and visioning, to architecture and design, and a strong dose of change management. Analytic implementations have many nuances you don't find in other more traditional app development or implementation projects, so a specific well-honed analytic approach is truly necessary. if businesses use CoE extensively, and it produces high ROI, it is successful. This assumes that data governance, data quality, security etc. is impeccable. 

3. Executive sponsorship is a key factor. Strong business involvement at all levels is critical to success, such as executive sponsorship. Quick turn around on the business's request and high utilization by business are the measure of success. The people leading an analytics center need to have a particular mix of characteristics. They must be able to speak to management but also have a very strong grasp of the modeling concepts. If they cannot convey their vision of analytics to the rest of the company, then the analytics group may eventually be perceived as interesting but not necessarily meaningful work done by back-room staff. 

4. It is crucial that domain expertise and institutional knowledge be reflected in the reports, tools, and predictive models that the center of expertise creates. Analytics is one of the most widely applicable competencies in business… and also one of the most interdisciplinary practices. A good analytics team should therefore represent a range of skills and perspectives. Within the technical realm, a good team should contain talented systems people, dedicated data experts and statistical programmers, as well as statisticians and machine learning experts who are capable of creatively fashioning technical solutions to business problems. or taking an analytics initiative to fly. 

5. Analytics Talent in CoE are “interpreters” who are capable of speaking the language of multiple disciplines. So an analytics COE should be conceived neither as a back room full of number crunchers nor as an ivory tower R&D group. Not coincidentally, the COE - and its products - will gain a lot more traction within the rest of the organization if it is organized to have strong links with the business units that it is intended to serve. In short, a range of skills and perspectives, strong intellectual and social links with the rest of the organization are key.  

6. A center of excellence would define the time frame and detailed objectives for achieving enterprise-wide excellence. Excellence can be defined and measured by applying a framework that clearly identifies the differentiators. At the highest level, businesses that have truly knowledgeable business intelligence people implement technology that works, provides clean usable data and uses proven processes to consistently maintain value. Measuring the degree of excellence is accomplished through use of the capability maturity model (CMM) against each of the pillars of success. 

Posted in: Big Data,Business Intelligence