Five minutes with rated_at in movie-ratings

Someone asked what is in rated_at on movie-ratings, and the honest answer took one query.

9,000 rows, no nulls, 8,999 distinct values. When the rating was submitted.

It runs from 2023-01-01 00:22:37 to 2025-06-19 23:13:27, covering 901 distinct days.

It is a timestamp, not a date, which means every comparison against a bare date is a comparison against midnight. rated_at <= date '2023-06-30' silently excludes almost all of 30 June. Half-open ranges — >= start and < end — avoid the whole class of off-by-one-day bugs and read no worse.

select
  count(*)                   as rows,
  count(rated_at)            as present,
  count(*) - count(rated_at) as nulls,
  count(distinct rated_at)   as distinct_values
from movie_ratings;

Check the range before filtering on it. Half the "the dashboard is empty" reports we have seen are a date filter outside the data's actual range, and the query is not wrong so nothing errors.

Where this bites: 901 populated days is what any window function over this column has to work with. A seven-day lag counts rows, not days — so if a day is missing, lag(7) quietly compares against eight days ago and the week-over-week number is wrong in a way that looks plausible.

The grain is one row per user-film rating, which is the context every one of those numbers depends on. None of them survive a change of grain, which is why "profile the column" and "profile the table" are the same job. Full schema, and the CSV, on the dataset page.