The Attribution Model Analysis feature in Cafe24 Analytics is a powerful tool that helps you analyze customers' complex purchase journeys from multiple perspectives, uncover the hidden contribution of each marketing channel, and develop optimal advertising strategies.
Use detailed data to see how customers visit and purchase from your store through various paths, understand the role each channel plays in driving purchase decisions, and find the performance measurement model that best fits your store.
2 key insights from Attribution Model Analysis
- Determine which attribution model is right for you (Comparing attribution models)
- Dive deep into channel performance using the attribution model best suited for your store (Analyzing a specific attribution model)
to develop detailed strategies.
Info - Data is available from April 11, 2025 onward.
- All data in Attribution Model Analysis is available from April 11, 2025 onward.
- If the comparison period is before April 11, 2025, comparison data cannot be displayed.
- However, the Data-Driven Model under Ad Channel Analysis>Attribution Model Analysis provides data from January 7, 2026 onward.
Comparing attribution models
Compare and analyze channel performance across multiple evaluation criteria (models) such as Last Click and First Click to find the attribution model that best fits your store.
All attribution models comprehensively analyze media auto-tracking parameters, UTM parameters, and domain (referral) information to evaluate the actual impact each channel has on conversions.
The five attribution models are as follows:
Example: A customer sees a YouTube ad, searches on Naver a few days later, and finally searches on Google to make a purchase of 100,000 KRW.
Last Click Model
- This model assigns all credit to the last channel visited before the purchase.
- Google, the last channel, receives 100% of the 100,000 KRW in revenue.
First Click Model
- This model assigns all credit to the first channel that brought the customer to your store.
- YouTube, the first channel, receives 100% of the 100,000 KRW in revenue.
U-Shaped Model
- This balanced model assigns higher weight to the first and last channels in the journey, with lower weight given to the channels in between.
- The revenue is split among YouTube (approx. 40,000 KRW), Google (approx. 40,000 KRW), and Naver in the middle (approx. 20,000 KRW).
Linear Model
- This is the most equitable model, distributing credit equally across all channels in the journey.
- All three channels in the example each receive an equal share of approximately 33,333 KRW.
Data-Driven Model
- This model uses data analysis to automatically calculate the contribution of each channel.
- The actual contribution of each of the three channels in the example is calculated based on data analysis results.
Info - About the Data-Driven Model
- The Data-Driven Model provides data from January 7, 2026, up to the previous day. Today's analysis results can be viewed the next day.
- Up to the top 10 channels with the highest daily conversions are analyzed.
Insight summary
| Metric | Description |
|---|---|
| Stable channels | These are channels whose contribution remains relatively consistent regardless of which attribution model is applied. They can be interpreted as channels that steadily contribute to conversions. |
| Sensitive channels | These are channels whose contribution varies significantly depending on the attribution model selected. Careful interpretation is needed when analyzing performance to determine which model to use as the baseline. |
| Contribution difference by attribution model | Shows how differently each channel's performance can be evaluated depending on the attribution model. The contribution difference is displayed as 0-100%, and the higher this value, the more important the choice of model becomes.
Example:
|
Top channel by attribution model
Shows the contribution of 8 major channels across the 5 attribution models.
Contribution refers to the share of revenue generated through a given channel relative to the store's total revenue.
(Analyzed channels: Instagram, Naver, Facebook, YouTube, Kakao, Google, TikTok, Twitter/X)
Channel influence by attribution model
This chart compares how major channels are evaluated under each attribution model. The shape of the chart allows you to intuitively understand the characteristics of each channel.
Contribution refers to the share of revenue generated through a given channel relative to the store's total revenue.
Contribution by channel
You can intuitively compare the contribution of each channel across all models. This lets you see under which attribution model each channel has the greatest influence.
Contribution refers to the share of revenue generated through a given channel relative to the store's total revenue.
Analyzing a specific attribution model
Select one evaluation criterion (model) from the 5 attribution models and dive deep into channel-level performance to develop detailed strategies.
Insight summary
| Metric | Description |
|---|---|
| Top contributing channel | The channel that contributed the most to revenue under the selected model. |
| Contribution of top channel | The contribution of the channel that contributed the most to revenue under the selected model. Contribution refers to the share of revenue generated through a given channel relative to the store's total revenue. |
| Attributed revenue of top channel | The attributed revenue of the channel that contributed the most to revenue under the selected model. Attributed revenue refers to the revenue generated through a given channel when the selected attribution model is applied. |
| Top previous visit channel | The channel most frequently visited just before the conversion, helping you identify the key assisting channel for the final conversion.
|
| Top next visit channel | Shows the key bridging channel that maintained customer interest after the first visit.
|
| Top mid-funnel channel | The most influential channel in maintaining customer interest during the middle stages of the purchase journey.
|
| Average number of channels before purchase | Shows the average number of channels a customer passes through before making a purchase.
|
Contribution analysis by channel
A performance scorecard showing which channels delivered the greatest results under the selected model.
The screen displays analysis for 8 major channels only. Click the View all button to see data for all channels.
(Analyzed channels: Instagram, Naver, Facebook, YouTube, Kakao, Google, TikTok, Twitter/X)
| Metric | Description |
|---|---|
| Contribution | Shows the contribution of each channel when the selected attribution model is applied. Contribution refers to the share of revenue generated through a given channel relative to the store's total revenue. Since only the attributed revenue share of the 8 major channels is shown in the chart, the total may not add up to 100%. You can view the attributed revenue share of all remaining channels by clicking View all. |
| Attributed revenue | The revenue generated through a given channel when the selected attribution model is applied. |
Attributed revenue trend by channel
Shows how the attributed revenue of each channel has changed over time.
You can determine whether a channel's own performance has improved by comparing it against the total revenue.
| Metric | Description |
|---|---|
| Attributed revenue | Shows the trend of attributed revenue for the selected channel.
Available by daily, weekly, or monthly views.
|
Contribution trend by channel
Identify which channels have become more or less important compared to the previous period.
Contribution refers to the share of revenue generated through a given channel relative to the store's total revenue.
| Metric | Description |
|---|---|
| Selected period | Represents the contribution of each channel for the period configured at the top of the screen.
|
| Comparison period | Represents the contribution of each channel for the equivalent period immediately preceding the selected period.
|