Specialised Models for Analytics Engineers
Specialised models are modelling patterns for common business domains. Event and product analytics models turn clickstream data into sessions and funnels, while finance and operations models handle ledgers, revenue recognition and inventory, where accuracy and auditability matter more than speed.
Stage: Data Modelling · Part of the Analytics Engineer roadmap · Last reviewed 19 November 2026
What to learn in Specialised Models
- Build sessions from raw events using a time gap rule
- Model a funnel with ordered steps and conversion rates
- Create a revenue table that reconciles to the ledger
- Handle late-arriving and corrected records
- Design snapshots to answer what was true on a date
Specialised Models interview questions
How do you sessionise raw event data?
Order each user's events by time and start a new session when the gap since the previous event exceeds a threshold, often 30 minutes. A window function with lag and a running sum of session starts assigns session ids.
Why must finance models reconcile to source systems?
Finance numbers feed reporting and audits, so totals must match the accounting system. Reconciliation tests compare model totals to source totals, which exposes missing rows, duplicates and timing differences early.
Where Specialised Models fits in the Analytics Engineer roadmap
Specialised Models belongs to the Data Modelling stage (intermediate level), which takes about 5 hours. Structure data for people and machines.
Other units in this stage
Back to the full Analytics Engineer roadmap