My statistics degree taught me to compute a p-value; a student-council election taught me that nobody in the room cared. As data secretary, I built a genuinely careful turnout model for our campus elections and presented it with confidence intervals and caveats — and watched the committee's eyes glaze until I said one sentence: 'We'll lose the evening voters if we don't move the booth.' They acted on the sentence, not the model. That gap between correct analysis and used analysis is the entire reason I am applying for a master's in business analytics rather than pure statistics.
I built strong technical foundations — probability, regression, and a working command of R and SQL that I sharpened by analyzing three years of our college fest's ticketing data to find why certain events consistently under-sold. But I deliberately paired every technical project with the translation problem: how to turn a coefficient into a decision a non-technical person will actually make. I ran a workshop teaching junior students to read a dashboard, and discovered I was better at that than at the modeling, which told me something about where I belong.
Business analytics is the precise field for someone who finds the last mile — analysis to decision — more interesting than the modeling itself. I want the coursework that statistics programs skip: how organizations actually decide, how to frame an analysis around a business question, how to build the dashboard that changes behavior rather than the model that impresses reviewers.
My goal is a decision-analytics or consulting role where the deliverable is a changed decision, not a notebook. The evening-voters sentence worked because it translated a model into an action; I want a career spent building that bridge, for stakes higher than a booth location.
I have spent two years proving that our marketing works, and I have started to suspect I have been proving it wrong. As an analyst at a D2C brand, I report the ROAS everyone celebrates, but the last-click attribution underneath it is a convenient fiction I understand well enough to distrust and not well enough to fix. When our CMO asked whether we should cut our upper-funnel spend, my honest answer was that my numbers couldn't say — and a master's in business analytics is my response to having given that answer.
My work has real analytical weight. I own our marketing-mix reporting, I built the cohort-LTV model that reset our customer-acquisition-cost ceiling and stopped us overspending on a channel that looked good on last-click, and my A/B testing framework is now standard across the growth team. But every one of these lives at the edge of my formal training — I run experiments whose statistics I half-understand and build attribution models whose assumptions I can't defend to a skeptic.
I want the graduate coursework that would turn my instincts into methods: causal inference and experimental design done properly, the marketing-science literature on attribution and media mix, and the optimization to move from measuring spend to allocating it. My employer has framed this as an investment, with an analytics-lead role waiting on my return.
Long term, I want to lead a marketing-analytics function that allocates budget on defensible causal evidence rather than attribution theater. I have been the analyst who reports the number everyone likes. I am applying to become the one who can tell them when the number is lying.