Calculating and updating session cost data in the adCost and attributedAdCost fields

Note: Session cost attribution is available only in the subscription plans Reports & Attribution and Enterprise.

OWOX BI allows assessing the effectiveness of your ad channels based on the calculated session cost. To do it, you need the following data:

To calculate the session cost, the session data tables are supplemented by the cost data from CostData_ tables automatically. And the total session cost is attributed evenly between sessions according to the UTM tags 'source', 'medium', 'campaign', 'keyword', and 'content' and written to the fields adCost and attributedAdCost.

Note:By default, Google Analytics applies the Last Non-Direct Click attribution model, in which the value of direct visits may be attributed to the last non-direct traffic source. Read more about this.

The adCost field

When all tags are present, the session's cost is attributed evenly between all sessions with the corresponding tags and is written to the AdCost field.

The attributedAdCost field

When one or more tags are unknown, the cost is attributed evenly between the sessions according to the known tags and is written to the attributedAdCost field.

Processed cases:

  1. No 'utm_content' tag — costs are attributed accordingly to 'utm_source', 'utm_medium', 'utm_campaign', 'utm_term'.
  2. No 'utm_term' and 'utm_content' tags — costs are attributed accordingly to 'utm_source', 'utm_medium', 'utm_campaign' and 'utm_term'.
  3. Only the 'utm_source' and 'utm_medium' tags are known — costs are attributed accordingly to these known tags.

Note:To distribute the attributed ad cost to a session, the presence of the 'source' and 'medium' tags is required. If even one of these tags is absent, then costs won't be attributed.

Calculation examples

We know we've spent 100$ (adCost = 100) on sessions with the following tags:

  • 'utm_source' = 'google'
  • 'utm_medium' = 'cpc'
  • 'utm_campaign' = 'spring_sale'
  • 'utm_term' = 'running+shoes'
  • 'utm_content' = 'logolink'

We also know had 5 sessions total with the same values of the tags 'utm_source', 'utm_medium', utm_campaign', 'utm_term', and 'utm_content'.

Since all the tags perfectly match in all sessions, the field AdCost will get the value calculated as 100$ / 5 sessions = 20$.

In this case, the costs are calculated for the attributedAdCost field in the same way:

'utm_source' 'utm_medium' 'utm_campaign' 'utm_term' 'utm_content' 'attributedAdCost' 'AdCost'
'google' 'cpc' 'spring_sale' 'running+shoes' 'logolink' 20 20



Another case: Google Analytics has cost data for 100$ more with this set of tags:

  • 'utm_source' = 'google'
  • 'utm_medium' = 'cpc'
  • 'utm_campaign' = 'sale'
  • 'utm_term' = 'running+shoes'
  • 'utm_content' = 'logolink'

However, Google Analytics doesn't have data on sessions with this set of tags. In this case, the sum of costs of 'google / cpc / sale' will be attributed to all 'google / cpc' sources as (100$ + 100$) / 5 sessions = 40$. But the value in the AdCost field won't change, since this time we don't gave a perfect match by tags:

 'utm_source'  'utm_medium'  'utm_campaign'  'utm_term'  'utm_content'  'attributedAdCost'  'AdCost'
'google' 'cpc' 'spring_sale' 'running+shoes' 'logolink' 40 20



Data updates

Velues in both the AdCost and attributedAdCost fields are updated every day during an actualization window preset for data imported from Google Analytics.

Changes in cost data, collected via the connected Ad Service → Google BigQuery pipelines, will be reflected in the adCost and attributed AdCost fields in 6 hours after this data is updated in the CostData_ tables.

If you use the Ad platforms → Google Analytics pipelines as a data source for the CostData_ table (see scheme №1 of the cost data import), please consider the following:
when the tags are changed in your Google Analytics account, but your ad spends remain the same, data in both the "CostData_" and "owoxbi_sessions_" tables won't be updated.
To update the tags in the OWOX BI table, contact us at bi@owox.com.

Was this article helpful?
2 out of 2 found this helpful
Have more questions? Submit a request

0 Comments

Please sign in to leave a comment.