Objectives
The main objective of the Practice Module is for participants to assimilate the knowledge gained through the four component courses and apply them in a holistic manner to solve real-world platform analytics problems.
Components
The practice module utilises a fully project-based assessment model that balances group collaboration with individual accountability. The total grade is divided evenly between group and individual evaluations.
Group Assessment (50%)
This section evaluates the team's collective ability to solve a practical platform analytics problem and communicate its value.
- Group Project Report (30%): This report must document the entirety of the team project, covering aspects like the problem statement, methodology, solution design, and evaluation results. It is designed to demonstrate the team's critical thinking, technical depth, and analytical reasoning.
- Project Pitch and Demonstration (20%): Teams are required to deliver a strategic pitch and demonstrate their deployed capability or solution. The presentation must communicate the project's business value and justify key analytical decisions. Additionally, teams must answer questions from assessors to prove their ability to translate analytical outcomes into practical business recommendations.
Individual Assessment (50%)
This section ensures personal accountability by evaluating each participant's independent mastery of concepts and their specific contributions to the group.
- Individual Contribution Portfolio and Evidence Audit (20%): Participants must document their specific roles, responsibilities, and contributions. This portfolio requires concrete evidence, such as code commits, testing results or decision logs, to verify the authenticity and significance of their work.
- Individual Technical Analysis and Justification Report (20%): Each participant must submit a report focusing on a specific technical, design, or implementation decision they personally contributed to. The report must explain the rationale behind the decision, detail trade-offs and alternatives considered and use project evidence to justify the choices made.
- Individual Reflection Journal (10%): Throughout the module, participants must maintain a journal documenting their learning journey, challenges faced and lessons learned. The reflections must be evidence-linked and demonstrate the ability to synthesize concepts learned across the four component courses.
Typical examples of projects to be undertaken
1. End-to-End Platform Build and Launch
In this project, teams are required to build and deploy their own real-world, multi-sided platform application from end to end. This involves identifying a viable platform opportunity, establishing value propositions for each side, and implementing core analytics capabilities. Teams must articulate the business model and pricing strategy, design the data model, and build an intelligent discovery or recommendation system. Additionally, they must integrate operational analytics, such as platform health dashboards and real-time machine learning inference components, while designing an experiment to measure the impact of a specific platform feature. The platform must be accessible to real users for a defined evaluation period, with team performance gauged by actual public engagement and business outcome metrics. The final deliverables for this track include the deployed platform, technical documentation, a business performance report, and a live demonstration.
2. Analytics Integration in a Simulated Platform Environment
In this project, teams will integrate their analytics capabilities via API into a simulated, AI agent-based multi-sided marketplace. Required developmental features include a recommendation engine, a real-time inference component for operational decisions and an experimentation or personalisation module. The strategic framing and business rationale for each of these capabilities must be grounded in fundamental platform economics principles. The simulation will run the integrated analytics stack through market scenarios that feature varying supply-demand conditions, competitive dynamics and agent behavioural shifts. Team performance is then evaluated using business outcome metrics generated directly from the simulation, such as gross merchandise value and match quality scores. Deliverables for this track consist of the deployed API integration package, a technical architecture document, an experiment results report and a presentation connecting the analytics decisions to the simulated business outcomes.
3. Industry Partner Analytics Audit and Capability Build
In this project, teams collaborate with a real organisation operating a platform business to audit their existing analytics stack, identify capability gaps, and develop a targeted analytics capability that addresses a genuine business need. The project begins with a strategic assessment of the partner's platform model, competitive positioning, and current analytics maturity. Based on these audit findings, teams will scope and build a specific solution, which could range from a recommendation engine or operational monitoring system to a fairness audit or causal experimentation framework. The primary deliverable is a working prototype utilising real or real-adjacent data, accompanied by a strategic recommendation deck that contextualises the built capability within the organisation's broader platform strategy. This strategic deck must comprehensively address the business model, data architecture, operational readiness, and personalisation opportunities. Final assessment evaluates the prototype's technical quality, the rigor of the audit, the strategic framing and the partner organisation's feedback regarding the relevance and feasibility of the deliverables.
Deliverables and success criteria for the problem statements:
- Apply platform economics concepts and strategic frameworks to scope, justify and evaluate platform analytics initiatives.
- Design and implement data models, event schemas, and feature engineering pipelines appropriate for multi-sided platform data.
- Build and deploy intelligent discovery, recommendation or matching systems with appropriate evaluation metrics.
- Implement at least one real-time ML inference or operational analytics capability in a production-grade or simulation-integrated setting.
- Design and execute causal experiments or quasi-experimental analyses adapted for platform environments with network effects.
- Develop data-driven personalisation, dynamic pricing, or incentive strategies grounded in treatment effect estimation and experimentation.
- Apply governance, fairness, and trust-and-safety principles to platform analytics, including bias auditing, anomaly detection or synthetic data generation where applicable.
- Demonstrate critical thinking, technical communication and the ability to link analytics design decisions to business outcomes.