Gangarapu, Sandeep Kumar2023-02-162023-02-162022-11https://hdl.handle.net/11299/252546University of Minnesota Ph.D. dissertation. November 2022. Major: Business Administration. Advisor: Ravi Bapna. 1 computer file (PDF); x, 110 pages.Digital transformation is defined as the use of digital technologies to transform every area of an existing business. A 2019 McKinsey report \citep{bughin2020} found that the top 10\% of the digitized incumbents earned 80\% of the revenue in their industries. One of the key strategic themes of digital transformation is to turn data the company collects into assets \citep{rogers2016digital}. In this thesis, we look at two different contexts, randomized control trials and referral marketing. In the first essay, we define a prescriptive analytics framework that addresses the needs of a budget constrained decision-maker facing, ex-ante, unknown costs and benefits of multiple policy levers to maximize overall utility from the randomized control trial data that the company already stores. we find a targeting strategy that produces an order of magnitude improvement in expected total utility compared to existing methods. In the second essay, we solve some of the challenges associated with referral marketing by turning the referral data that the company has into an asset. We investigate how referral targeting compares to algorithmic targeting in its effectiveness. We understand the mechanisms behind a referral and why they are valuable. We also unpack the effects of `information' and `influence' that play a role in the purchase decision of a referred customer.enDigital TransformationHeterogeneous treatment effectsrandomized control trialsreferral marketingThesis or Dissertation