Causal Inference for Real World Decision Making
Anirban Bhattacharyya
Seasoned data scientist & educator
.png&w=768&q=75)
Master Causal Inference to Make Credible Decisions from Messy Real-World Data
Most data scientists are asked to answer high-stakes “what caused what?” questions long before they have a perfect A/B test: Did this feature actually improve retention? Did the campaign drive incremental revenue? Did policy changes reduce churn, or were users already trending that way?
This workshop helps you move beyond surface-level correlations and build the judgment to make credible causal claims from messy, real-world data. You’ll learn how to frame business questions as causal questions, choose the right method when experimentation is limited, identify common sources of bias, and communicate assumptions clearly to stakeholders.
By the end, you’ll have a practical toolkit for making better product, growth, marketing, and policy decisions when randomized experiments are unavailable, delayed, or incomplete
Workshop agenda
- 12:00PM EDT
Causal Thinking for Real-World Decisions
Learn why causal inference matters when A/B tests are unavailable and how counterfactual thinking helps answer “what caused what?”
- 12:30PM EDT
Turning Business Questions into Causal Estimands
Convert vague stakeholder asks into clear treatments, outcomes, populations, time windows, and estimands like ATE, ATT, and CATE.
- 1:15PM EDT
Regression Adjustment for Product Feature Impact
Estimate feature impact by comparing naive and adjusted results while accounting for observable confounders in business data
- 2:30PM EDT
Lunch break
Pause, recharge, and return ready for hands-on causal model implementation
- 3:15PM EDT
Matching Methods for Marketing Campaign Lift
Build comparable treatment and control groups to estimate incremental impact from targeted marketing campaigns
- 4:30PM EDT
Difference-in-Differences for Product and Policy Rollouts
Measure product, policy, or regional rollout impact using pre/post trends and treatment/control comparisons
- 5:45PM EDT
Short break
Short reset before the second half of hands-on modeling and case work
- 6:00PM EDT
Propensity Scores and IPW for Churn Interventions
Use propensity scores and inverse probability weighting to estimate retention or churn intervention impact
- 7:15PM EDT
Heterogeneous Treatment Effects and Uplift Thinking
Move beyond average impact to identify which users, customers, or segments benefit most from a treatment
- 8:00PM EDT
Capstone: Method Selection and Stakeholder Communication
Choose the right causal method for business cases, then explain assumptions, uncertainty, and recommendations clearly to stakeholders
Learn directly from Anirban
.png&w=384&q=75)
Anirban Bhattacharyya
Data scientist and educator with 10+ years in experiments and causal inference
Who this workshop is for
Data scientists & analysts who need to answer “what caused what?” using messy product, growth, or business data.
Product, growth & marketing teams making decisions when clean A/B tests are unavailable, delayed, or incomplete.
Students & early-career data pros who want practical causal inference skills for interviews and real-world data work.
What's included
.png&w=1536&q=75)
Live sessions
Learn directly from Anirban Bhattacharyya in a real-time, interactive format.
Lifetime access
Go back to course content and recordings whenever you need to.
Community of peers
Stay accountable and share insights with like-minded professionals.
Certificate of completion
Share your new skills with your employer or on LinkedIn.
Maven Guarantee
Your purchase is backed by the Maven Guarantee.
Frequently asked questions
$800
USD