Understanding Attribution
Conceptually, attribution is easy to understand. Metrics are calculated based on allocation rules. At its simplest, attribution is based on even allocation.
But taking a look ‘under the hood’ shows that even this basic attribution model can be quite complicated. To do this, let’s isolate one order:
One product sold to Google Affiliate Network:
Two clicks in the path:
Exclusions (‘All But First’, meaning we exclude giving credit to that source unless it’s the first click in a path, so the first click (Direct) will get ½ the credit):
Correct Calculation: ($224.00 * .50) = $112.00 (Revenue)
Then, back out 50% of the total cost of goods sold and 100 % of Ad Spend to get to the total net profit, which in this case is negative.
It may seem confusing that the calculation shows ½ the revenue, then backs out ½ the cost of goods sold, yet backs out 100% of Ad Spend. However, this is the only way to accurately value each advertising source. Traditional web analytics tend to under-credit some sources and over-credit the last click sources. Search engine reports can over-credit ads, especially when people cross search engines in their research.
In the example below, both Yahoo and Google would take credit for the sale:
Taking complete Ad Spend into consideration when calculating profit is crucial to accurately value each paid media source, but there is one other consideration. What about the clicks that don’t cost anything? That is where Exclusions come in to play.
If you exclude clicks that don’t cost anything, the story changes. If the exclusions had been set to exclude all non-cost clicks, the earlier example would tell a different story.
Correct Calculation: ($224.00 * .100) = $224.00 (Revenue)
Then, back out 100% of the total cost of goods sold and 100 % of Ad Spend to get to the total net profit. In this case, the conversion would have been profitable.
Excluding giving credit to the clicks that do not cost you anything gives you the ability to analyze and optimize your total marketing budget. It answers the questions, ‘Where is money being spent making the most money?’, and ‘Where is it losing money?’
The good news is that you can still see how the impact of the non-cost clicks affects the value of each ad source by switching the display option.
Tip of the Month: Exclude your non-cost clicks to optimize your budget to your Ad Spend. Then use the complete Purchase Path to see the impact of the non-cost clicks on the value of the different ad sources. It is the best of both worlds.