Implementation Strategies
Implementation strategy covers how a solution actually lands: phased rollout, sandbox strategy, deployment mechanics, data migration order, testing and adoption. Examiners test whether you can sequence work and pick the right environment.
Sandboxes: Developer and Developer Pro for build and unit test; Partial Copy for integrated testing with a data subset (it requires a sandbox template that defines which objects are copied); Full Copy for UAT and performance testing because it mirrors production data and configuration. Refresh intervals and storage differ, and a Full Copy refresh overwrites the sandbox entirely - a classic trap.
Deployment: change sets move metadata between related orgs; unlocked and managed packages suit distribution; a DevOps pipeline with version control is the answer whenever questions mention multiple releases, rollback or many admins. Certain changes need post-deployment steps (recalculating sharing, activating flows, assigning permission sets, reloading data, re-running report subscriptions).
Data migration order follows dependencies: users, then accounts, then contacts, then leads/opportunities and their children (products, quotes, orders, contracts), then attachments and history. Loading children before parents fails on lookups.
Phasing and adoption: pilot with one team, gather feedback, then roll out in waves; run UAT with real scenarios; train by role, not by feature; appoint champions; plan hypercare. A Center of Excellence and a release calendar prevent the "big bang then abandonment" failure mode.
Worked example: a three-phase rollout - Phase 1 accounts/contacts/activities, Phase 2 lead-to-opportunity with forecasting, Phase 3 quotes, orders and contracts - each with its own sandbox, UAT exit criteria and training wave.
Rule of thumb: migrate parent objects first, deploy metadata second, load reference data third, and never migrate into production without a Full Copy rehearsal.