Building on this CTE, we can add in DAU from earlier. Complete the query by calling the DAU query we created earlier, now aliased as daily_players:

/**/ daily_revenue as (
  select
    date(created_at) as dt,
    round(sum(price), 2) as rev
  from purchases
  where refunded_at is null
  group by 1
), 
daily_players as (
  select
    /**/ as dt,
    /**/ as players
  from gameplays
  group by 1
)
select * from daily_players order by dt;

Here’s a hint on how we created the previous DAU query.

Now that we have the revenue and DAU, join them on their dates and calculate daily ARPU. Complete the query by adding the keyword using in the join clause.

/**/ daily_revenue as (
  select
    date(created_at) as dt,
    round(sum(price), 2) as rev
  from purchases
  where refunded_at is null
  group by 1
),
daily_players as (
  select
    /**/ as dt,
    /**/ as players
  from gameplays
  group by 1
)
select
  daily_revenue.dt,
  daily_revenue.rev / daily_players.players
from daily_revenue
  join daily_players /**/ (dt);

In the final select statement, daily_revenue.dt represents the date, while daily_revenue.rev / daily_players.players is the daily revenue divided by the number of players that day. In full, it represents how much the company is making per player, per day.