Data Pipeline SLA
also called Freshness SLA, Data SLO
A published commitment about when data will be available and how fresh it will be, turning a pipeline into something consumers can design against.
Analytical pipelines are usually operated without any stated commitment, which leaves every consumer guessing whether an empty result means no activity or a late load — and guessing wrong in board meetings.
A usable commitment names three things. Freshness: the maximum lag between an event happening and it being queryable. Timeliness: the wall-clock time by which the daily load is complete, which is what report schedules and downstream jobs actually depend on. Completeness: what proportion of expected records have arrived, since a pipeline can be on time and missing a source.
Publishing these changes behaviour on both sides. Consumers stop building defensive logic against unknown latency, and the platform team gains a defensible basis for prioritisation and for refusing requests that would breach it.
The monitoring that goes with it is the part that must be automated: alert on absence — the load that did not happen is invisible to any check that only inspects data that arrived — and expose freshness as a queryable value beside the data, so a dashboard can display its own staleness rather than presenting stale numbers as current.