With the students mapped, this is what we built: a repeatable way to gamify any activity, the frameworks that decide which reward fits which moment, and the information architecture and wireframes that took it from theory to a clean build handoff.
The team needed something they could run again without us. So the core deliverable was a repeatable five-step flow for gamifying any activity on the platform:

Every activity gets sorted before it's rewarded — because the two goals want completely different rewards.

Two taxonomies let the team weigh a reward's purpose against the student it's for — the conversation that actually decides build cost.
The matrix works both ways: pick an appropriate reward from an activity's effort and frequency — or pick an activity that suits an existing reward.

Social rewards should be given plenty, and spread across the whole spectrum.
For regular-to-long-term users, offering a choice between long-term rewards can be very valuable.
Rewards
Activities
Rewards
Activities
Rewards
Activities
The whole system points one way: use reward value to move students from extrinsic motivation toward intrinsic over time — so eventually the learning is its own reward.

An information architecture tied the theory to real moments — entry points, state checks, reward outcomes — and low-fi wireframes gave design and dev a shared map before graphics and animation.

A system is only as good as the research under it.