Why attribution shows what actually works in Digital Marketing - Performance Analysis
When we analyze attribution in digital marketing, we want to understand how the effort to track and assign credit grows as campaigns and data increase.
We ask: How does the work needed to show what actually works change when we have more channels or conversions?
Analyze the time complexity of this simplified attribution process.
for conversion in conversions:
for touchpoint in conversion.touchpoints:
assign credit to touchpoint.channel
This code goes through each conversion and then each touchpoint that led to it, assigning credit to channels involved.
Look at what repeats in the code.
- Primary operation: Looping through each conversion and then each touchpoint inside it.
- How many times: For every conversion, it processes all its touchpoints.
As the number of conversions and touchpoints grows, the work grows too.
| Input Size (n) | Approx. Operations |
|---|---|
| 10 conversions, 3 touchpoints each | 30 operations |
| 100 conversions, 3 touchpoints each | 300 operations |
| 1000 conversions, 3 touchpoints each | 3000 operations |
Pattern observation: The total work grows roughly in direct proportion to the number of conversions and their touchpoints combined.
Time Complexity: O(n * m)
This means the time to assign credit grows with both the number of conversions (n) and the number of touchpoints per conversion (m).
[X] Wrong: "Attribution time only depends on the number of conversions."
[OK] Correct: Each conversion can have many touchpoints, so the total work depends on both conversions and their touchpoints.
Understanding how attribution scales helps you explain how marketing data processing grows with campaign complexity, a useful skill in many roles.
What if we only assigned credit to the last touchpoint instead of all touchpoints? How would the time complexity change?