Skip to content

Instantly share code, notes, and snippets.

@markrittman

markrittman/compare_segment_rudderstack_ga4_page_views_by_page_title.sql

Last active May 21, 2023 18:10
Download ZIP

Revisions

  1. markrittman renamed this gist May 21, 2023. 1 changed file with 0 additions and 0 deletions.
    0 gistfile1.txt → ...derstack_ga4_page_views_by_page_title.sql
    View file
    File renamed without changes.
  2. markrittman created this gist May 21, 2023.
    74 changes: 74 additions & 0 deletions gistfile1.txt
    View file
    Original file line number Diff line number Diff line change
    @@ -0,0 +1,74 @@
    with segment_pageviews as (
    select
    timestamp_trunc(p.received_at,day) as day_day,
    context_page_title as page_title,
    count(distinct id) as page_views
    from
    `ra-development.company_website.pages_view` p
    where
    date(timestamp_trunc(p.received_at,DAY)) > '2023-05-01'
    and context_page_url not like '%info.rittmananalytics.com%'
    group by
    1,2
    ),
    rudderstack_pageviews as (
    select
    timestamp_trunc(p.received_at,day) as day_day,
    context_page_title as page_title,
    count(distinct id) as page_views
    from
    `ra-development.rudderstack_ra_website.pages_view` p
    where
    date(timestamp_trunc(p.received_at,DAY)) > '2023-05-01'
    group by
    1,2



    ),
    ga4_pageviews as (
    select
    timestamp_trunc(timestamp_micros(g.event_timestamp),day) as day_day,
    (
    SELECT
    value.string_value
    FROM
    UNNEST(event_params)
    WHERE
    event_name = 'page_view'
    AND KEY = 'page_title') AS page_title,
    sum(1) as page_views
    from `ra-development.analytics_277223877.events_*` g
    where event_name = 'page_view'
    and
    timestamp_trunc(timestamp_micros(g.event_timestamp),day) > '2023-05-01'
    group by 1,2
    ),
    comparison as (
    select
    date(coalesce(s.day_day, r.day_day,g.day_day)) as day, coalesce(s.page_title,r.page_title,g.page_title) as page_title, s.page_views as segment_page_views, r.page_views as rudderstack_page_views, g.page_views
    from
    segment_pageviews s
    full outer join
    rudderstack_pageviews r
    on
    s.day_day = r.day_day
    and
    s.page_title = r.page_title
    full outer join
    ga4_pageviews g
    on
    s.day_day = g.day_day
    and
    s.page_title = g.page_title
    )
    select
    page_title, sum(segment_page_views) as total_segment_page_views, sum(rudderstack_page_views) as total_rudderstack_page_views,
    1-((sum(segment_page_views) - sum(rudderstack_page_views))/sum(segment_page_views)) as rudderstack_variance,
    sum(page_views) as total_ga_page_views,
    1-((sum(segment_page_views) - sum(page_views)) /sum(segment_page_views)) as ga4_variance
    from
    comparison
    group by
    1
    order by 1