Use the splitting and API habits from Chapter 3 and the serving and monitoring habits from Chapter 6. Real pricing experiments, approvals and waiting for returns take extra calendar time.
Read the causal primer first. The schedule assumes prepared opportunity data, your earlier environment and an existing API boundary. Rebuilding data from scratch or a new serving integration needs extra setup time. Finish the approved fixed-price baseline, supported demand comparison and constrained grid before the optional 2026 research extensions.
| Time | Focus | Build | Done when |
|---|---|---|---|
| 3 hours | Decision and data | Price unit, horizon, cost definition, exposure and inventory checks | A data contract distinguishes zero demand, no exposure and stockout |
| 5 hours | Baselines | Current-price policy, seasonal demand baseline, log-price GLM | Rolling validation and elasticity assumptions are documented |
| 4 hours | Challenger and uncertainty | One boosted demand model; demand calibration and intervals | Accuracy and calibration compared by SKU, price and period |
| 4 hours | Recommendation | Enumerate an approved price grid; enforce constraints and fallback | Unsupported prices and unsafe inputs cannot become recommendations |
| 4 hours | Serve and evaluate | Recommendation API, decision log, simulated experiment | Request can be replayed and matched to a matured outcome |
| 5 hours | Buffer and write-up | Failure exercise, experiment memo, model and policy cards | Claims clearly distinguish simulation, prediction and causal evidence |