← Disney Interview Insights

Disney·Data Scientist·Technical Phone Screen·Senior

Senior
Aug 2025Remote

Summary

Disney/Hulu data science case interview, basically a full causal inference firefight compressed into one session. No holdout was configured on a 100% rollout, which set the tone for everything that followed. Brutal but fair.

Questions Asked (6)

Q1

Define the primary decision metrics and guardrails for evaluating a homepage ranking model rollout, and set concrete thresholds for each.

Product Analytics & MetricsA/B Testing & Experimentation
Author's notes

I led with watch-time per session as the north star, which felt right, but I fumbled the guardrails a bit.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the evaluation around a clear primary success metric tied to the homepage's core business goal (e.g., engagement or conversion), then define guardrails that protect user experience and long-term health. Propose concrete thresholds based on historical baselines, statistical power, and business impact, and explain how you would monitor them during the A/B test.

Pro tip: Anchor thresholds to the minimum detectable effect (MDE) you can reliably measure given traffic, and explicitly state that guardrails are non-negotiable—if breached, you pause the rollout regardless of primary metric gains.

1. Identify the primary decision metric

Choose one metric that directly reflects the homepage's purpose, such as click-through rate to key content or conversion rate, and justify why it's the north star for this rollout.

2. Define guardrail metrics

Select 2-3 metrics that ensure the change doesn't harm user experience or long-term value, like bounce rate, page load time, or downstream retention.

3. Set concrete thresholds

For each metric, specify a target (e.g., +2% relative lift for primary) and a guardrail threshold (e.g., no more than 1% relative increase in bounce rate), grounded in historical data and business constraints.

4. Plan statistical validation

Outline how you'll determine significance (e.g., power analysis, p-value < 0.05) and handle multiple comparisons, ensuring thresholds are detectable given traffic.

5. Define rollout decision rules

State the go/no-go criteria: e.g., primary metric must meet target with statistical significance and no guardrail breaches; otherwise, iterate or halt.

Key Points to Mention

  • Alignment with business goals (e.g., Disney's streaming engagement or merchandise conversion)
  • Use of historical baselines and A/B test power analysis to set realistic thresholds
  • Guardrails for user experience (e.g., bounce rate, time to interactive) and long-term metrics (e.g., retention)
  • Statistical significance and confidence intervals to avoid false positives
  • Segment analysis to check for heterogeneous effects (e.g., new vs. returning users)
  • Pre-registration of metrics and thresholds to prevent p-hacking

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

Without a configured holdout group, how would you use Android users as a difference-in-differences control for the iOS rollout? What assumptions does this require and how do you test them?

A/B Testing & ExperimentationRoot Cause Analysis
Author's notes

This is where I actually felt decent.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Explain that you would treat Android users as a synthetic control group for the iOS rollout, using a difference-in-differences (DiD) design to estimate the causal effect of the iOS change. Then outline the key assumptions—parallel trends, no spillover, stable composition—and describe how you would test each assumption using pre-period data and robustness checks.

Pro tip: Emphasize that while DiD with a non-equivalent control is weaker than a randomized holdout, it can still provide credible estimates if you validate the parallel trends assumption and conduct sensitivity analyses (e.g., placebo tests, synthetic control). This shows you understand the trade-offs and can make the most of imperfect data.

1. Define the treatment and control groups

Identify the iOS users who received the change as the treatment group and Android users as the control group. Ensure both groups are observed over the same pre- and post-rollout periods.

2. Specify the DiD model

Use a regression framework: Y = β0 + β1*Post + β2*iOS + β3*(Post*iOS) + ε, where β3 is the DiD estimate. Include covariates to control for differences between platforms if needed.

3. Test the parallel trends assumption

Plot the outcome trends for iOS and Android over the pre-period and test for parallel trends using event-study regressions or placebo tests. If trends diverge, consider alternative methods like synthetic control.

4. Check for spillovers and compositional changes

Assess whether the iOS change could affect Android users (e.g., via network effects) and whether user composition changed differentially across platforms during the rollout.

5. Conduct robustness checks and sensitivity analysis

Perform placebo tests (e.g., fake treatment dates), vary the control group (e.g., use only Android users with similar characteristics), and test for heterogeneous effects to strengthen causal inference.

Key Points to Mention

  • Difference-in-differences (DiD) design and its causal interpretation
  • Parallel trends assumption and how to test it (event study, placebo tests)
  • No spillover effects between iOS and Android users
  • Stable group composition over time (no differential attrition or selection)
  • Potential confounding factors (e.g., platform-specific trends, seasonality)
  • Alternative methods if assumptions fail (synthetic control, matching, instrumental variables)

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

How would you design a geo-based synthetic control within iOS as an alternative identification strategy, and what diagnostics would you run?

A/B Testing & ExperimentationAdaptability & Ambiguity
Author's notes

Blanked for a second here.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the problem: when randomization is infeasible or contaminated, a geo-based synthetic control can estimate causal effects by constructing a weighted combination of untreated geographies that mimics the treated market's pre-intervention trend. Then walk through the design steps—selecting donor pools, building the synthetic control, and validating with placebo tests and pre-period fit—while emphasizing diagnostics that assess robustness and uncertainty.

Pro tip: Mention that you would pre-register the donor pool and weight constraints to avoid p-hacking, and use in-time placebo tests (e.g., backdating the intervention) to demonstrate that the method would not have found spurious effects in the pre-period.

1. Define the intervention and outcome

Clarify the treatment (e.g., a new feature launch in a specific region), the unit of analysis (DMA, country, or city), and the primary metric (e.g., engagement, conversion). Specify the pre- and post-intervention periods.

2. Select donor pool and construct synthetic control

Choose untreated geographies that are similar to the treated unit on key covariates and pre-treatment outcomes. Use a weighted combination (e.g., via optimization) to create a synthetic counterfactual that closely tracks the treated unit's pre-intervention trajectory.

3. Estimate effect and quantify uncertainty

Compute the treatment effect as the difference between the treated unit and its synthetic control post-intervention. Use placebo tests (in-space and in-time) and permutation inference to derive p-values or confidence intervals.

4. Run diagnostics and robustness checks

Assess pre-treatment fit (e.g., RMSE, R-squared), check for spillovers or interference, test sensitivity to donor pool and weight constraints, and examine whether other events could confound the results.

5. Interpret and communicate caveats

Discuss the limitations of synthetic control (e.g., reliance on donor pool, no randomization) and how the results complement or triangulate with other evidence. Recommend next steps if needed.

Key Points to Mention

  • Donor pool selection: criteria for including/excluding geographies, avoiding treated or contaminated units, and ensuring pre-treatment similarity.
  • Weight optimization: methods like constrained regression or quadratic programming to minimize pre-treatment difference, and regularization to avoid overfitting.
  • Placebo tests: in-space (randomly assign treatment to donor units) and in-time (backdate intervention) to assess significance and rule out spurious effects.
  • Pre-treatment fit metrics: RMSE, R-squared, and visual inspection of trends to ensure the synthetic control is credible.
  • Spillover and interference: checking for cross-geo effects (e.g., via border discontinuities or network effects) that could bias estimates.
  • Sensitivity analysis: varying donor pool, weight constraints, and outcome definitions to test robustness of conclusions.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q4

The rollout window includes a major content premiere and a payments outage. How do you adjust your analysis to account for these events, and what fixed effects or controls do you include?

Root Cause AnalysisA/B Testing & ExperimentationProduct Analytics & Metrics
Author's notes

Day-of-week fixed effects, an indicator for the premiere date, and a separate indicator for the outage window felt like the right answer.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

First, acknowledge that the premiere and outage are exogenous shocks that can bias your analysis if not properly controlled. Then, propose a robust strategy: either exclude affected periods if they are not representative, or include fixed effects and time-varying controls to isolate the treatment effect. Finally, emphasize the importance of sensitivity analysis to validate your findings.

Pro tip: Always pre-register your analysis plan and define exclusion criteria before looking at the data to avoid p-hacking and ensure credibility.

1. Identify and Characterize Events

Determine the exact timing, duration, and scope of the content premiere and payments outage. Assess whether these events affect the entire user base or specific segments.

2. Decide on Exclusion vs. Control

If the events are unrelated to the treatment and affect all users equally, consider excluding those periods. If they interact with the treatment, include controls to account for their impact.

3. Include Fixed Effects and Controls

Add time fixed effects (e.g., day, hour) to absorb common shocks, and user or segment fixed effects to control for unobserved heterogeneity. Include time-varying covariates such as outage indicator, premiere indicator, and their interactions.

4. Conduct Sensitivity Analysis

Test the robustness of your results by varying exclusion windows, control specifications, and model types. Report how the treatment effect estimate changes.

5. Communicate Assumptions and Limitations

Clearly state the assumptions behind your chosen approach and discuss potential biases. Provide confidence intervals and effect sizes to contextualize the findings.

Key Points to Mention

  • Difference-in-differences or fixed effects models to control for time-invariant confounders
  • Time fixed effects (e.g., day-of-week, hour-of-day) to account for seasonality and common shocks
  • User-level fixed effects to control for individual heterogeneity
  • Event indicators (premiere, outage) and their interactions with treatment
  • Exclusion of contaminated periods if events are unrelated to treatment
  • Sensitivity analysis to assess robustness of results

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q5

Quantify the business impact of the observed watch-time decline with confidence intervals, and estimate the weekly revenue effect across both ad and subscription revenue paths.

Product Analytics & MetricsPricing & Monetization
Author's notes

Back-of-envelope stuff.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining the watch-time metric and quantifying the decline with confidence intervals using appropriate statistical methods. Then, map watch-time to revenue through ad and subscription paths, estimating the weekly revenue impact with sensitivity analysis. Finally, communicate the results with clear assumptions and caveats.

Pro tip: Always segment the analysis by user cohorts (e.g., subscribers vs. free users, geography, device) because the revenue impact can vary significantly, and this shows you understand the business nuances.

1. Define and Measure Watch-Time Decline

Clearly define watch-time (e.g., total hours, average per user) and calculate the decline with confidence intervals using historical data and statistical tests.

2. Link Watch-Time to Ad Revenue

Model the relationship between watch-time and ad impressions, then apply CPM rates to estimate the ad revenue impact, accounting for uncertainty.

3. Link Watch-Time to Subscription Revenue

Estimate how watch-time affects subscriber retention, acquisition, and churn, and translate that into subscription revenue impact using cohort analysis or elasticity models.

4. Aggregate and Validate Revenue Impact

Combine ad and subscription revenue effects, propagate confidence intervals, and validate assumptions through sensitivity analysis or A/B testing if possible.

5. Communicate Findings with Caveats

Present the weekly revenue effect with confidence intervals, highlighting key assumptions, limitations, and potential next steps for validation.

Key Points to Mention

  • Use of confidence intervals and statistical significance to quantify uncertainty
  • Segmentation by user type (subscribers vs. ad-supported) and demographics
  • Ad revenue calculation: watch-time to impressions to CPM
  • Subscription revenue: impact on churn, retention, and lifetime value
  • Sensitivity analysis to test robustness of assumptions
  • Clear communication of assumptions and limitations to stakeholders

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q6

Outline an immediate remediation plan for the rollout and a proper experiment design for the following month, including holdouts, pre-registration, variance reduction, and exposure logging.

A/B Testing & ExperimentationProduct Strategy
Author's notes

Felt most comfortable here.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by addressing the immediate remediation: diagnose the issue, stop the bleeding, and implement a quick fix with monitoring. Then outline a rigorous experiment design for the next month, covering holdouts, pre-registration, variance reduction, and exposure logging to ensure valid and reliable results.

Pro tip: Emphasize the importance of pre-registration to prevent p-hacking and ensure credibility, and mention using CUPED or stratification for variance reduction to increase sensitivity. Also, highlight the need for detailed exposure logging to track user interactions and debug issues.

1. Immediate Remediation

Identify the root cause of the rollout issue, pause the rollout if necessary, and implement a hotfix. Set up real-time monitoring and alerting to catch regressions.

2. Define Experiment Goals and Metrics

Clearly state the primary metric (e.g., conversion, engagement) and guardrail metrics. Determine the minimum detectable effect (MDE) and required sample size.

3. Design Experiment with Holdouts and Pre-registration

Randomly assign users to treatment and control groups, ensuring a holdout group remains untreated for long-term measurement. Pre-register the hypothesis, metrics, and analysis plan to avoid bias.

4. Implement Variance Reduction and Exposure Logging

Use techniques like CUPED or stratification to reduce variance and increase power. Log all exposures and user interactions to ensure data quality and enable debugging.

5. Analysis and Iteration

Analyze results according to the pre-registered plan, check for novelty effects, and decide on next steps (scale, iterate, or stop).

Key Points to Mention

  • Root cause analysis and hotfix for immediate remediation
  • Pre-registration to prevent p-hacking and ensure validity
  • Holdout groups for long-term impact measurement
  • Variance reduction techniques (e.g., CUPED, stratification)
  • Exposure logging for data quality and debugging
  • Sample size calculation and MDE

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.