Statistics query

[Ryan Marks]

I'm trying to capture data from Revive by way of a MySQL query that will be consumed by a client-facing statistics dashboard.  The data I'm interested in is Client name, Client ID (stored in Client Comments), Date, Total Impressions, Total Clicks, Affiliate name.

I've written the following query, but it takes a while to respond:

SELECT
    client.clientname AS `Client`
    , client.comments AS `Client ID`
    , DATE_FORMAT( CONVERT_TZ( STR_TO_DATE(stats.date_time, '%Y-%m-%d'), 'GMT', 'America/Chicago'), '%Y-%m-%d') AS `Date`
    , SUM( stats.`impressions` ) AS `Impressions`
    , SUM( stats.`clicks` ) AS `Clicks`
    , affiliate.name as 'Website'
FROM
    `ox_clients` client
    LEFT JOIN `ox_campaigns` campaign USING (clientid)
    LEFT JOIN `ox_banners` banner USING (campaignid)
    LEFT JOIN `ox_data_summary_ad_hourly` stats ON banner.bannerid = stats.ad_id
    LEFT JOIN `ox_zones` zone ON stats.zone_id = zone.zoneid
    LEFT JOIN `ox_affiliates` affiliate USING (affiliateid)
WHERE
    client.comments <> '' -- we only want clients with the comments field populated
    AND DATE_FORMAT( CONVERT_TZ( STR_TO_DATE(stats.date_time, '%Y-%m-%d'), 'GMT', 'America/Chicago'), '%Y-%m-%d') = DATE(NOW()) - INTERVAL 1 DAY
GROUP BY
    client.clientid
    , affiliate.name
    , DATE_FORMAT( CONVERT_TZ( STR_TO_DATE(stats.date_time, '%Y-%m-%d %H:%i:%s'), 'GMT', 'America/Chicago'), '%Y-%m-%d')
ORDER BY
    DATE_FORMAT( CONVERT_TZ( STR_TO_DATE(stats.date_time, '%Y-%m-%d %H:%i:%s'), 'GMT', 'America/Chicago'), '%Y-%m-%d')

The above query took 48 seconds to run.  I ran an explain which is below.

Is there a query I can use that will help me get to the dimensions and metrics I've mentioned above?

[andrewatfornax]

Hi Ryan,

I haven't actually run your query, so, just to clarify - is the issue that your query isn't returning the data you want (and you need help getting the right info), or is the issue that it takes a while to return (and you need help making it faster)?

Thanks!