Multi-touch Attribution

Author: Kadence Leung 

Last Update：2026/04/12

 

Introduction

Standard Amazon advertising reports inherently favor a "Last Touch" bias, giving 100% of the conversion credit to the final ad a customer clicked. This completely ignores the upper-funnel awareness campaigns (like DSP or Sponsored Brands) that initially introduced the shopper to your product.

The Multi-Touch Attribution (MTA) model solves this by analyzing the entire user conversion path. By applying various data-driven and rule-based attribution logics, this model reveals the true, marginal contribution of every ad touchpoint. It goes beyond simple reporting by providing actionable budget reallocation recommendations and simulating your predicted performance improvements.

This model can help you answer questions like:

• Beyond Amazon's basic default attribution, what is the true value of my upper-funnel awareness campaigns?
• If I shift my budget away from Search and into Display, how will that impact my total conversions?
• Exactly how should I reallocate my advertising budget to maximize my overall marketing efficiency?

 

---

When to Use This Report

This model is essential for any advertiser who manages a budget across multiple ad types and wants to stop "guessing" which campaigns are working:

• Justify Awareness Spending: Use this to prove the ROI of top-of-funnel ads (DSP/SB) by showing how they "assist" your lower-funnel search ads.
• Optimize Budget Distribution: Stop relying on gut feelings. Use the budget optimization tool to get mathematical advice on exactly where to shift your next dollar.
• Forecast Future Growth: Use the simulation tool to see your predicted ROAS and total purchases before you actually make any budget changes in the ad console.
• Understand the Full Path: Identify which specific combinations of ads (like "Video ad then Search ad") create a "multiplier effect" and drive the most profit.

---

The 7 Attribution Models

MTA offers seven ways to slice your data. For the most accurate results, we recommend the Model-Based options (Markov and Shapley).

Model	How it Works	Best Use Case
Markov Chain (Recommended)	Measures how much your sales would crash if a specific ad was removed from the mix.	Finding the "Essential" ads in your strategy.
Shapley Value (Recommended)	Uses game theory to fairly distribute credit based on an ad's marginal contribution.	Complex, multi-touch media mixes.
First Touch	Gives 100% credit to the very first ad seen.	Evaluating your best "New-to-Brand" acquisition drivers.
Last Touch	Gives 100% credit to the final ad.	Matching standard Amazon Advertising Console reports.
Linear	Distributes credit equally across every ad in the journey.	Products with long consideration cycles.
Time Decay	Gives more credit to ads seen closer to the purchase time.	Short promotional cycles (like Prime Day).
Position-Based	Gives 40% to the first ad, 40% to the last, and 20% to the middle.	Valuing both the "Hook" and the "Closer."

 

How to Use It 

 

1. Compare Attribution Results Select one of the 7 models from the top menu. The dashboard will generate a side-by-side bar chart comparing Amazon's Official Attribution against your Selected MTA Attribution.

• What to look for: Look for ad types (like DSP) whose bars are significantly higher under the MTA model than the Amazon model. This proves they are driving hidden value.

💡 Pro Tip: Understanding the Markov Chain Interactive Chart When you select the Markov Chain model, navigate to the Explanation tab to unlock an additional interactive chart. This visualizes every conversion path a user might experience.

• The Nodes (Circles): Each circle represents a specific ad touchpoint. Users passing through a circle represent the audience reached by that ad.

• Interactive Removal Effect: Click on any ad touchpoint to temporarily "remove" it from the ecosystem.

• The Impact: Look at the data window on the right. It will calculate the "Removal Effect," showing you exactly how much your overall conversion rate would drop if you stopped running that specific ad, giving you a crystal-clear view of its true attribution share.

2. Review Budget Optimization Once you understand the true value of your touchpoints, navigate to the Budget Optimization module. The system will compare your historical spending proportions against the new attribution data and provide a mathematically recommended budget split (e.g., Shift 10% from SP to DSP).

3. Analyze Result Predictions Click the Forecast Budget Optimzation Result tab. Based on the recommended budget, the system will simulate the expected improvement in your ROAS and total conversions. If the prediction bars are above the axis, your new budget allocation is forecasted to increase overall performance.

 

FAQ

Q: What counts as a "touchpoint" in this model?

A: In the predefined baseline model, DSP is attributed based on impressions in the past 14 days. Sponsored Products, Brands, and Display are attributed based on clicks in the past 7 days.

Q: Can I customize the attribution lookback windows and trigger events?

A: Yes. When building a Custom Model, you can define up to 10 custom ad groupings. For each grouping, you can define the exact event type (Impression vs. Click) and the specific lookback period.

• Pro Tip: Use "Impressions" and longer lookback windows for upper-funnel tactics (DSP). Use "Clicks" and shorter windows for lower-funnel tactics (SP).

Q: Why do my MTA results look skewed or distorted?

 A: This is usually due to your lookback window settings. If your window is too short, you will miss key early touchpoints. If it is too long (e.g., an 80-day lookback for a $5 everyday consumable product), you will pull in irrelevant data and unrelated ad interactions that didn't actually influence the final sale. Always align your lookback window with the natural decision cycle of your specific product.

Q: Does this model account for external or offline factors?

A: No. This model only tracks digital interactions on Amazon. Always consider external factors (like a big holiday or a competitor being out of stock) when reviewing budget advice.

 

Glossary

Type	Term	Description
Dimension	Event	Impression or click that attributon model
	Ads Account	The account used for managing and executing advertising campaigns.
	Model	The algorithm or method used to attribute conversions or other desired outcomes to specific events or touchpoints.
	Goal	The primary objective or result that an advertising campaign aims to achieve.
	Attribution	The process of assigning credit to different marketing touchpoints for their role in achieving the campaign goals.
Metrics	Total Purchase	The total number of purchases made as a result of the advertising campaign.
	Total NTB Purchase	The total number of purchases made by new-to-brand (NTB) customers, indicating first-time buyers influenced by the campaign.
	Actual Cost	The total expenditure on the advertising campaign.
	Suggest Budget	The recommended budget allocation for future advertising efforts, based on past performance and future goals.
	Forecast Result Change	The predicted change in campaign performance metrics as a result of proposed adjustments or ongoing trends.

Video Tutorial