Using continous aggregates for total (all-time) aggregates?

TimescaleDB and PostgreSQL
1

I’m experimenting with Timescale for tracking views on a website, and while continuous aggregates work great for aggregating views per day/week/month/etc., I was wondering how to use them for “all-time views”, something like this:

CREATE MATERIALIZED VIEW metrics_episode_views_total WITH (timescaledb.continuous)
AS
SELECT
    episode_id,
    COUNT(1) AS views
FROM metrics_episode_views
GROUP BY
    episode_id
WITH NO DATA
;


SELECT add_continuous_aggregate_policy('metrics_episode_views_total',
    start_offset => INTERVAL '2 hours',
    end_offset => INTERVAL '1 hour',
    schedule_interval => INTERVAL '1 hour'
);

If I try this, Timescale complains that continuous aggregate view must include a valid time bucket function.

I could try to cheat that by giving a time bucket of 1000 years or so, but it made me wonder if I was approaching this wrong? Is continuous aggregates not the right tool to use for this kind of thing?