EA — Growth Analytics — Product Sense & Metrics

Role context: Data Scientist, Growth Analytics · Est. study time: 35 min · 4 practice questions

How to prepare for this role

EA Growth Analytics is a games marketing-science role: measure the causal, incremental impact of marketing spend and allocate budget across channels and a games portfolio. It's judged on incremental player LTV (not attribution), MMMs calibrated by experiments, retargeting over-attribution, forecasting game performance, and marginal-ROI budgeting — communicated to media buyers.

The day-to-day is building Bayesian hierarchical MMMs, geo/retargeting incrementality experiments, budget-optimization tools, and game-performance forecasts. The instinct that separates strong candidates: acquiring on predicted LTV vs CAC (not install volume), and knowing that post-IDFA attribution over-credits — so incrementality and MMM, not attributed installs, are the truth.

Where to spend your prep time:

  • Games-marketing product sense (this article) — the LTV-driven UA funnel, why incremental beats attributed, and retargeting's over-attribution.
  • Incrementality, MMM, and forecasting — geo experiments, Bayesian hierarchical MMM, ghost-ad holdouts for retargeting, Prophet forecasting. See the A/B Test & Causal Inference section.
  • Python/R, SQL, and communication — be fluent (incl. Bayesian modeling), and turn an incrementality result into a budget recommendation for media buyers.

The through-line: acquire incremental player LTV at a healthy LTV:CAC, measure it causally, and allocate on marginal ROI.

What EA Growth Analytics actually is

EA is a large games publisher — EA Sports FC (the FIFA successor), Apex Legends, Madden, The Sims, Battlefield — earning boxed/premium sales plus live-service revenue (in-game purchases, Ultimate Team, battle passes, cosmetics). The identity to carry into every answer: a games portfolio whose marketing acquires and re-engages players, judged on the incremental player LTV it drives at an efficient cost — where user-level attribution is unreliable post-IDFA, so incrementality and MMM are the trustworthy measurement.

Two things make this different from a normal marketing-analytics job:

  • UA is LTV-driven, not install-driven. The games funnel is install → retention → monetization → LTV, so you acquire on predicted LTV vs CAC, not install volume — and you predict a player's LTV early (from first-days behavior) to bid on the right players. A cheap install that never monetizes is a loss.
  • Measurement is causal, because attribution is broken. Apple's IDFA deprecation broke user-level ad attribution, so attributed installs over-credit and mislead. The truth is geo incrementality experiments and MMM (calibrated by those experiments) — and retargeting is especially over-attributed (you target players about to return anyway).