LAG in BigQuery

This page is a quick reference checkpoint for LAG in BigQuery: behavior, syntax rules, edge cases, and a minimal example; plus the official vendor documentation.


Function Details

LAG lets you look back at a previous row's value within the same partition.

Returns the value of value_expression on a preceding row. - offset must be non-negative; default_expression is used when no row exists at the specified offset. (type and behavior details as documented)

Warning: An error occurs if the offset is NULL or a negative value.

If this behavior feels unintuitive, the tutorial below explains the underlying pattern step-by-step.

LAG(value_expression[, offset [, default_expression ]]) OVER (window_clause)

SELECT division, runner_name, finish_time, LAG(runner_name) OVER (PARTITION BY division ORDER BY finish_time) AS previous_runner FROM race_results;

What should you do next?

If you came here to confirm syntax, you’re done. If you came here to get better at window functions, choose your next step.

Understand the pattern

LAG is part of a bigger window-function pattern. If you want the “why”, start here: Lead Lag

Prove it with a real query

Reading docs is useful. Writing the query correctly under pressure is the skill.

Two Orders Before, Two Orders After: The Price Prophecy

Support Status

  • Supported: yes
  • Minimum Version: BigQuery is continuously updated and has no versioned SQL engine, so minimum version information does not apply.

Official Documentation

For the authoritative spec, use the vendor docs. This page is the fast “sanity check”.

View BigQuery Documentation →

Looking for more functions across all SQL dialects? Visit the full SQL Dialects & Window Functions Documentation.