--- title: "Attribution: Multiple Assignment Models" slug: "attribution-multiple-assignment-models" updated: 2026-01-14T14:15:38Z published: 2026-01-14T14:15:38Z canonical: "docs.mapp.com/attribution-multiple-assignment-models" --- > ## Documentation Index > Fetch the complete documentation index at: https://docs.mapp.com/llms.txt > Use this file to discover all available pages before exploring further. # Attribution: Multiple Assignment Models Multiple assignment models allow you to distribute attribution credit across several campaigns or channels, providing a comprehensive view of the customer journey. This approach helps you evaluate the combined impact of multiple marketing touchpoints instead of assigning the full conversion value to a single interaction. **Why Use Multiple Assignment Models?** Multiple assignment models are useful when: - Customer journeys involve several campaigns or channels before conversion. - You want to avoid overvaluing a single touchpoint (for example, last click). - Credit should be shared more fairly across contributing marketing activities. --- ## 1 Overview of Multiple Assignment Models We differentiate between **manual models**, where the user configures weightings or distribution rules, and the **automatic model**, which dynamically calculates channel contributions using data-driven insights. The table below summarizes the key characteristics of the multiple assignment models, helping you decide which one best fits your needs: | Model | Type | Description | Best For | | --- | --- | --- | --- | | **Data-Driven Attribution (DDA)** | Automatic | Attribution credit is calculated dynamically using Shapley Values and historical customer journey data. | Complex journeys with many touchpoints | | **Distribution Across All Ad Media** | Manual | Credit is distributed across all campaigns with manually configured weightings. | Simple use cases requiring precise control | | **Distribution Across the Last Five Ad Media** | Manual | Credit is distributed among the last five campaigns using user-defined percentages. | Scenarios where recent interactions matter most. | --- ## 2 Data-Driven Attribution (DDA) Data-Driven Attribution (DDA) distributes conversion value across marketing channels based on their observed contribution in real customer journeys. The model is based on **Simplified Shapley Values**, a transparent, data-driven approach derived from game theory. DDA operates at **channel level** (not individual campaign level) and is designed to support strategic budget allocation rather than granular media optimization. ### 2.1 Why Use DDA? DDA helps marketers by: - Distributing credit fairly across all contributing channels. - Reflecting actual customer behavior instead of fixed attribution rules. - Adapting automatically as customer journeys change over time. ### 2.2 How DDA works DDA follows two main steps: 1. **Shapley Values Calculation** Historical customer journeys are analyzed to determine each channel’s contribution to conversions. 2. **Attribution to New Journeys** The calculated Shapley Values are applied to new journeys to distribute conversion value. --- #### Step 1: Shapley Value Calculation (Simplified Example) Shapley Values measure the **marginal contribution** of each channel across all observed journeys. In simplified terms: - Each channel participating in a journey contributes to the total conversion value. - Contributions are aggregated across all journeys. - Each channel’s share of the total conversion value becomes its Shapley Value. ****Example (simplified, without positioning)**** | Customer Journeys | Conversion Value (€) | Contribution SEA (€) | Contribution Email (€) | Contribution Display (€) | | --- | --- | --- | --- | --- | | SEA > Display > Email | 300 | 100 | 100 | 100 | | Email > Display | 100 | 0 | 50 | 50 | | Email > SEA | 600 | 300 | 300 | 0 | | **Totals** | **1000** | **400** | **450** | **150** | The Shapley Values are then calculated by dividing each channel's total contribution by the overall conversion value (1000 €): - **SEA:** 400 / 1000 = 40% - **Email:** 450 / 1000 = 45% - **Display:** 150 / 1000 = 15% | **Channel** | **Simplified Shapley Value (%)** | | --- | --- | | SEA | 40% | | Email | 45% | | Display | 15% | > [!NOTE] > Note > > This example illustrates the Shapley principle in a simplified form. Position-specific effects are omitted for clarity. In the actual DDA model, channel contributions are calculated based on their ordered position within the customer journey. **Position in the Customer Journey** In Mapp Intelligence, DDA explicitly considers the **position of each channel within the customer journey**. - Customer journeys are evaluated as **ordered sequences** of channels. - Contributions are calculated **per channel occurrence and position** (for example: first, middle, penultimate, last). - A channel can appear multiple times within a journey and may receive different attribution values depending on its position. Position is **not handled through fixed weightings** (such as first-touch or last-touch rules). Instead, position-sensitive effects are reflected through the Shapley calculation based on observed journey patterns. --- #### Step 2: Attribution to New Customer Journeys Once Shapley Values are available, they are applied to new customer journeys: - Only the channels present in the journey are considered. - Shapley Values are **normalized** so that the total attribution equals 100%. - Conversion value is distributed according to the normalized contributions. **Example:** | **Customer Journey** | **Conversion Value (€)** | **SEA (%)** | **Email (%)** | **Display (%)** | | --- | --- | --- | --- | --- | | Display > Email | 200 | 0% | 75% (normalized: 45/(45+15)) | 25% (normalized: 15/(45+15)) | | SEA > Display > Email | 300 | 40% | 45% | 15% | **Attributed Values:** | **Customer Journey** | **SEA (€)** | **Email (€)** | **Display (€)** | | --- | --- | --- | --- | | Display > Email | 0 | 150 | 50 | | SEA > Display > Email | 120 | 135 | 45 | The percentages in the table above are normalized for the relevant channels in each customer journey to ensure the total equals 100%. For example, in the *Display > Email* journey, Email's contribution of 45% and Display's 15% are recalculated as 75% and 25%, respectively. This step-by-step approach ensures that credit is distributed fairly to each channel based on its role and position in the customer journey. --- ### 2.3 Additional Information for Marketers - **Clear Visibility:** Use the attribution metrics for analysis and the [Smart notification](/v1/docs/smart-notifications) to understand the calculated shared each channel has. - **Dynamic Updates:** Shapley Values are recalculated daily using the most recent data. - **Early Phase Behavior:** If insufficient data is available, attribution is distributed equally across channels. - **Transparency:** The model provides consistent and explainable attribution results based on real customer journeys. --- ## 3 Distribution across all Ad Media This manual model assigns attribution credit to **all campaigns** recorded in the customer journey. - Weightings are configured manually. - Position-based percentages can be defined. - If fewer campaigns are present than configured positions, weightings are normalized automatically. This model is suitable when you want **full control** over how attribution credit is distributed. **Example:** Configuration including percentage-based distribution per position for metric Order Value ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/CleanShot 2024-10-09 at 17.59.35@2x.png) - Customer Journey ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/image-1728489711506.png) - Analysis ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/ss1(169).png) If less than 5 Ad media are available, the weightings will be distributed. ****Example:** Overview of normalized weightings with varying numbers of campaign contacts** ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/CleanShot 2024-10-09 at 17.59.35@2x(1).png) | | Number of campaign contacts in the Customer Journey | | --- | --- | | | 1 | 2 | 3 | 4 | 5 | 6 | | Endmost | 100% | 55% *last + next to last +* *(0.5 * others)* | 50% *last + next to last* | 42.5% *last + (0.25* other)* | 40% | 40% | | Second to last | | 45% *first + second +* *(0,5 * other)* | 10% *other* | 12.5% *next to last + (0.25*other)* | 10% | 10% | | Third to last | | | 40% *first + second* | 12.5% *second + (0.25*other)* | 10% | 5% *0.5*other* | | Fourth to last | | | | 32.5% *first + (0.25*other)* | 10% | 5% *0.5*other* | | Fifth to last | | | | | 30% | 10% | | Sixth to last | | | | | | 30% | **Traffic source Weighting** In addition to the position within the Customer Journey, the campaign type can also be used for analysis purposes with traffic source weightings. ****Example:** Overview of normalized weightings with varying numbers of campaign contacts** ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/ss1(170).png) | Position | Campaign | Position weighting | Traffic source weighting | Overall weighting | Normalization factor (based on overall weighting) | Final normalization (overall weighting * normalization factor) | | --- | --- | --- | --- | --- | --- | --- | | Endmost | SEM | 40% | 100% | 40% | 100 / 80 | 50% | | Second to last | SEM | 10% | 100% | 10% | 100 / 80 | 12.5% | | Third to last | Newsletter | 10% | 30% | 3% | 100 / 80 | 3.75% | | Fourth to last | Newsletter | 10% | 30% | 3% | 100 / 80 | 3.75% | | Fifth to last | Banner | 30% | 80% | 24% | 100 / 80 | 30% | | | | 100% | | 80% | | 100% | --- ## 4 Distribution across the last five Ad Media This model focuses on the **five most recent campaigns** before conversion. - Attribution is based on user-defined position weightings. - If fewer than five campaigns are present, values are normalized. - Campaigns beyond the fifth position receive no credit. This approach is useful when recent interactions are considered more influential than earlier touchpoints. **Example:** Configuration including percentage-based distribution per position for metric Order Value ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/CleanShot 2024-10-09 at 17.53.05@2x.png) - Customer Journey ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/ss1(162).png) - Analysis ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/ss1(163).png) If less than five Ad media are available, the data will be normalized. ****Example:** Normalizing three Ad Media** ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/ss1(164).png) | | Position weighting | Normalization factor (corresponds to the sum of the position weightings of the Ad media) | Normalized weighting (Position weighting * normalization factor) | | --- | --- | --- | --- | | Endmost | 50% | 100 / 85 = 1.17647 | 1.17647 * 50% = 58.82% | | Second to last | 20% | 100 / 85 = 1.17647 | 1.17647 * 20% = 23.53% | | Third to last | 15% | 100 / 85 = 1.17647 | 1.17647 * 15% = 17.65% | | Fourth to last | - | | | | Fifth to last | - | | | | Total | 85% | | 100% | ****Example:** Overview of normalized weightings with varying numbers of campaign contacts** ![](https://cdn.document360.io/554b9b98-6720-4d8b-9919-c7b203d72648/Images/Documentation/ss1(165).png) | | Number of campaign contacts in the Customer Journey | | --- | --- | | | 1 | 2 | 3 | 4 | 5 | 6 | | Endmost | 100% | 71.43% | 58.82% | 52.63% | 50% | 50% | | Second to last | | 28.57% | 23.53% | 21.05% | 20% | 20% | | Third to last | | | 17.65% | 15.79% | 15% | 15% | | Fourth to last | | | | 10.53% | 10% | 10% | | Fifth to last | | | | | 5% | 5% | | Sixth to last | | | | | | 0% |