The Advanced Analytics Solution Building Process
This is the second in a series of posts on business transformation through analytics and how organizations can run successful analytics pilots and build a thriving data science practice. The post is by Ilan Reiter, Principal Data Science Manager at Microsoft.
In our first post in this series we talked about the ingredients needed to successfully transform businesses through the power of analytics. In our next set of posts, starting with this one, we discuss the process behind successful implementations of advanced analytics (AA) projects.
We introduce here the Advanced Analytics Solution Building Process (AA SBP), a framework for building end to end data science solutions. This process covers the entire spectrum of activities from the discovery of business opportunities to problem formulation and successful production deployment. The process captures our leanings from working closely with many customers across a wide range of verticals and use cases.
The goal for the Advanced Analytics Solution Building Process is to provide a robust means for organizations to discover opportunities to derive business value from their data assets and to smoothly and rapidly realize that value from a set of qualified use cases.
Think of the AA SBP as a framework to accomplish the following goals:
- Systematically discover opportunities to create value from data.
- Qualify new opportunities and assess their fit and potential.
- Smoothly implement end to end AA pilots and projects.
- Maximize the number of successful AA deployments.
- Produce sustainable ongoing business value from data.
The AA SBP also helps with the creation of the appropriate metrics to track AA projects and objectively measure their success.
To meet the above goals of the AA SBP process, we adhere to some basic principles, namely:
- All AA projects must be driven by clear business needs.
- There is strong collaboration between business and technical teams to discover and qualify the opportunities that can produce business value from data.
- Business considerations and priorities determine the scope of a given AA project.
- For AA projects to deliver value, the insights they deliver they must actionable. Therefore, such projects must be fully integrated into the company’s business flow, with appropriate actions taken as a result of insights delivered.
- The business value delivered by an AA project is measurable and used to derive the data science and the performance metrics.
- Data science and data engineering must meet the required business goals and metrics.
- The business stakeholder has the responsibility to assess the impact and value produced by an AA project, using the designated metrics.
A high level illustration of the AA SBP is provided below. As seen, the process consists of four strongly coupled phases, each of which must be performed through a collaboration between the business stakeholders and the data science and engineering teams:
Business Problem Formulation. The goal of this phase is the discovery, qualification, prioritization and formulation of a business problem that can be solved using AA.
Data Acquisition. This step ensures that data from critical business processes is being collected and that relevant data sets are available for solving the business problem at hand.
Data Science. The goal of this step is to perform the underlying data science and analytics work needed to deliver the business performance or business value that is expected from the process.
Business Integration. This is a critical phase during which the business tracks and validates the solution and integrates it into their existing workflow, so that the value of implementing it can be fully realized.
Each phase consists of multiple steps that can be executed either sequentially or concurrently by multiple teams of individuals working in close collaboration with each other.
The process described here, along with a modern platform such as Cortana Analytics, provides a great foundation upon which organizations can transform their businesses through the power of data and analytics. In our next post in this series we will take a closer look at the first phase above, i.e. the discovery and formulation of business problems or opportunities that can benefit from the power of advanced analytics.
Subscribe to this blog.