Context Stability
Time Intelligence: Why ‘Last Month’ Is Harder Than It Sounds
Time intelligence depends on clean date tables and clear definitions of time.
TL;DR
- • Time logic must be explicit and consistent.
- • AI needs to know which date field is used.
The problem (layman)
- • Multiple date fields and inconsistent logic lead to different results.
- • AI can’t infer which time definition is intended.
Why it matters
- • Time comparisons are core to business decisions.
- • Inconsistent logic causes misleading trends.
Symptoms
- • Last month results vary by report.
- • Year‑over‑year comparisons don’t align.
Root causes
- • No dedicated date table or inconsistent relationships.
- • Time intelligence implemented differently across measures.
What good looks like
- • Dedicated date table with standard time measures.
- • Explicit time logic in canonical measures.
How to fix (steps)
- • Create a single date table and use it consistently.
- • Standardize time intelligence measures.
- • Document time basis in each KPI.
Pitfalls
- • Mixing calendar and fiscal definitions without labels.
- • Using implicit date fields.
Checklist
- • Single date table used across facts.
- • Time measures standardized.
- • Time basis documented.