From bcccc1d2accbf87f537f9a697b0b87fac24f3550 Mon Sep 17 00:00:00 2001 From: mirnawong1 <89008547+mirnawong1@users.noreply.github.com> Date: Wed, 29 Nov 2023 07:55:04 -0500 Subject: [PATCH] removing limitation bullet removing `metric_time` limitation as this was resolved in https://github.com/dbt-labs/metricflow/issues/825 --- website/docs/docs/build/cumulative-metrics.md | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/website/docs/docs/build/cumulative-metrics.md b/website/docs/docs/build/cumulative-metrics.md index 708045c1f3e..e4c3e1b9cbc 100644 --- a/website/docs/docs/build/cumulative-metrics.md +++ b/website/docs/docs/build/cumulative-metrics.md @@ -38,10 +38,7 @@ metrics: ## Limitations Cumulative metrics are currently under active development and have the following limitations: - -1. You can only use the [`metric_time` dimension](/docs/build/dimensions#time) to check cumulative metrics. If you don't use `metric_time` in the query, the cumulative metric will return incorrect results because it won't perform the time spine join. This means you cannot reference time dimensions other than the `metric_time` in the query. -2. If you use `metric_time` in your query filter but don't include "start_time" and "end_time," cumulative metrics will left-censor the input data. For example, if you query a cumulative metric with a 7-day window with the filter `{{ TimeDimension('metric_time') }} BETWEEN '2023-08-15' AND '2023-08-30' `, the values for `2023-08-15` to `2023-08-20` return missing or incomplete data. This is because we apply the `metric_time` filter to the aggregation input. To avoid this, you must use `start_time` and `end_time` in the query filter. - +- You can only use the [`metric_time` dimension](/docs/build/dimensions#time) to check cumulative metrics. If you don't use `metric_time` in the query, the cumulative metric will return incorrect results because it won't perform the time spine join. This means you cannot reference time dimensions other than the `metric_time` in the query. ## Cumulative metrics example