A Framework for Interpreting Realized Economic Value From Data Science Projects

Written with Daniel Fleck and David Lubert

The field of Data Science has grown immensely over the last decade and permeated through virtually every industry. In LinkedIn’s 2020 US Emerging Jobs Report, ‘Data Scientist’ is now third in the top fifteen emerging professions in the US with projected growth of 37% in 2020. In addition, Artificial Intelligence Specialist is the top job in this report with anticipated growth of 74% in 2020. The allure of using data thoughtfully and productively has led companies to dive headfirst into this emerging field in order to stay competitive within their industry.

Business leaders are staffing new data science teams internally and connecting with third-party consultants to implement data science techniques within projects such as forecasting, customer segmentation, and optimization -but how are they assessing these potential projects? Are they measuring these results against traditional economic evaluation metrics to determine the actual value being created? In this series, we will examine different ways that data science projects can be evaluated for economic efficacy.

The challenge for senior management is continuing to find novel ways to create value for the organization by increasing revenue, identifying and implementing cost efficiencies, and enhancing customer service quality. It is this desire that fuels their willingness to commit to costly data science resources, but how are they quantifying the short-term and long-term economic value these specialists provide? Translating data science results using traditional economic evaluation approaches is the key to maximizing the effectiveness of these efforts -while ensuring organizational strategic alignment. Below is a table of key evaluation metrics that are commonly used in the business world, each of which could meaningfully enhance data science initiatives when employed appropriately:

For example, if a Data Scientist develops a model that is 10% more accurate in detecting credit card fraud, what is the ‘true’ business impact after deployment? Taking this a step further, how will this ultimately affect free cash flow? As the use of data science within the business environment continues to evolve, it is important to ensure that these projects adhere to the traditional ways by which companies assess and choose projects in which to invest. This will allow for seamless integration of data science projects into the traditional business flow.

Without connecting these evaluation metrics to data science projects, the real value created by data science initiatives could be significantly over or underestimated. It is the goal of this blog series to connect these dots by providing a framework through which data science projects can be assessed. We hope to ensure that projects with the greatest potential financial and economic impact are prioritized. In addition, we will also provide specific guidance regarding the calculation of ongoing value a business realizes once a data science project is deployed.

Links to More Information:
Net Present Value (NPV)
Internal Rate of Return (IRR)
Profitability Index (PI)
Break-Even Point (BEP)
Payback Period (PP)
Discounted Payback Period (DPP)
Economic Value Add (EVA)
Return on Investment (ROI)

Links to the Rest of the Series:
Part I: NPV, IRR, PI
Part II: BEP, PP, DPP
Part III: EVA, ROI

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store