Design a Causal Measurement Plan for EcoWear's Ad Campaign
Overview
What this challenge is about.
Design a difference-in-differences plan to measure a fashion ad campaign’s effect, test parallel trends, and run a placebo check to earn your verifiable certificate.
The scenario
EcoWear is a small direct-to-consumer sustainable fashion brand that sells organic-cotton and recycled-fabric apparel through its own website, competing in a price-sensitive category where seasonal demand swings sharply around end-of-year holidays.
The Brief
What you'll do, and what you'll demonstrate.
Design a credible difference-in-differences measurement plan that isolates how much of EcoWear's 2023 online sales change was caused by its digital ad campaign rather than by seasonality or market-wide trends.
Earning criteria — what you'll demonstrate
- Translate a vague business question into a falsifiable causal measurement design
- Test and honestly report the parallel-trends assumption that difference-in-differences depends on
- Specify a difference-in-differences model that controls for seasonality and produces valid standard errors
- Use a placebo test to argue for or against the credibility of a causal estimate
- Communicate a causal result and its uncertainty so a non-technical stakeholder can make a spending decision
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Aligned coursework coming soon.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
- Difference In Differences
Apply difference in differences to solve real industry problems and demonstrate production-level capability.
- Causal Inference
Apply causal inference to solve real industry problems and demonstrate production-level capability.
- Time Series
Apply time series to solve real industry problems and demonstrate production-level capability.
- Panel Data
Apply panel data to solve real industry problems and demonstrate production-level capability.
Careers
Career paths this challenge builds toward
Completing this challenge demonstrates skills that transfer directly to these roles:
Marketing Data Scientist
Marketing teams constantly ask whether a campaign actually worked. This challenge builds the exact reflex employers want: turning messy weekly sales into a defensible causal number, stress-testing it, and recommending budget moves with quantified uncertainty.
This challenge sharpens
- difference-in-differences
- causal-inference
- time-series
Applied Econometrician
Consultancies and policy teams hire econometricians to estimate program and intervention effects. Designing a parallel-trends-validated difference-in-differences study on real panel data mirrors the core deliverable of that role almost exactly.
This challenge sharpens
- difference-in-differences
- causal-inference
- panel-data
Product Analytics Analyst
Product analysts evaluate launches and experiments where clean randomization is impossible. This challenge teaches the quasi-experimental toolkit and the discipline of communicating effect sizes and caveats to product and growth stakeholders.
This challenge sharpens
- causal-inference
- time-series
- panel-data