• “To test our hypothesis of the importance in getting the data right to drive meaningful improvement dialogue with our clients, EY worked with an independent analytics firm, New Constructs." Page 3
• "New Constructs uses machine learning technology to read through annual and quarterly reports and make numerous adjustments to close accounting loopholes and provide apples-to-apples comparability across thousands of companies.” Page 3

• "We conducted an ROIC analysis of four firms and analyzed the side-by-side results of New Constructs data in comparison to two of the largest financial data houses.” Page 4

[Note: the four firms are Dow, Comcast, UPS, & Caterpillar.]

• "Our analysis demonstrates that the ROIC calculations for the selected companies varies significantly across the three data sources over a six year period." Page 4

[Note: the three data sources are New Constructs, Bloomberg and Capital IQ.]

•  “The majority of the difference, though, comes from New Constructs’ machine learning approach, which leverages technology to calculate ROIC by applying accounting adjustments that may be buried deeply in the footnotes across thousands of companies.” Page 4

•  “The other advantage of New Constructs is the transparency in disclosing calculations and all the data behind them. It is hard to determine the calculations at a granular level from the reports of many data providers.” Page 4

• “Below are examples of ROIC calculation adjustments that New Constructs made for Peer 2, Peer 3 and Peer 4.” Page 4
• The entire Appendix for a side-by-side comparison of our analysis of DOW’s 2015 results to Capital IQ and Bloomberg.