What are attribution models in digital marketing?

Attribution models are rules or sets of rules that guide the assignment of credit for sales and conversions to the various touchpoints in your conversion paths. Through the use of attribution modeling, marketers are able to analyze and allocate credit to the marketing touchpoints that occur at particular steps of the customer journey - all the way from searching for a product online to actually making a purchase, and every action that lies in between these points.

Why do you need an attribution model?

Attribution modeling makes it possible for marketers to figure out how much credit every marketing channel and customer touchpoint should be given for a specific conversion.

How many attribution models are there?

1. Last-click attribution: This assigns all the credit for the conversion to the last-clicked ad and the corresponding keyword.2. First-click attribution: This assigns all the credit for the conversion to the first-clicked ad and corresponding keyword.3. Linear attribution: This allocates the credit for the conversion equally across all ad interactions on the conversion path.4. Time decay: This attribution model assigns greater levels of credit to ad interactions that happened at points that are nearer to the conversion. Credit is distributed through the use of a 7-day half-life. Essentially, an ad interaction that happens 8 days before a conversion gets half as much credit as an ad interaction 1 day before a conversion.5. Position-based: This attribution model assigns 40% of credit to both the first and last ad interactions and corresponding keywords, with the remaining 20% being spread out across the other ad interactions on the path.6. Data-driven: In this attribution model, the distribution of credit for a conversion based on your past data for this conversion action. The biggest difference between the data-driven attribution model and other attribution models is that the data-driven model makes use of your account's data for the purpose of calculating the actual contribution of each interaction across the conversion path.

What are the Single-touch attribution models?

1. First touch model2. Qualified Lead Model 3. Lead creation model4. Last non-direct click attribution

What are the Multi-touch attribution models?

1. Linear attribution2. Time decay attribution3. U-shaped attribution4. W-shaped attribution5. Z-shaped attribution

Attribution models

What are attribution models?

Attribution models are the frameworks & methodologies that are used in marketing to assign credit or value to different sources, channels, campaigns, conversions, etc.

These models basically answer this questions - "Which marketing activities influenced a customer to take a desired action?"

Attribution models provide insights into the ideal customer journey that eventually helps in efficient allocation of marketing resources.

What is the importance of attribution models in marketing?

Understanding the customer journey
Improving the campaign's performance
Optimizing your customer's experience
ROI measurement
Data-driven decision making
Efficient resource allocation

By understanding the importance of attribution models, marketers can drive more effective & efficient marketing campaigns.

What are the key concepts in attribution models?

1. Touchpoint

Any interaction that a customer has with the brand or their marketing campaign.

It could be a website visit, paid ads impression, social engagements, email opens & clicks, etc.

2. Conversion

The desired action or an outcome that a marketer wants a customer to take.

It could be form submission, video views, app downloads, etc.

3. Attribution window

The time period during which the touch-points are considered for attribution.

It decides how far back in the customer journey marketers look to assign credit.

What are the challenges of marketing attribution?

1. Cross-channel management

When web assets are shared across multiple websites, channels & teams and when the data passes between multiple sub-domains & the main domain.

2. Making decisions based on very small sample sizes

For websites receiving less traffic - the sample set may be biased.

3. Lack of tracking compliance

When you rely on manual data updates along with automated data updates, the attribution would always remain inaccurate or would have missing data.

4. Data privacy

GDPR, CCPA and other compliances make it difficult to capture the user journey without any breaks in between.

What are the types of attribution models?

1. First-touch attribution

This model assigns 100% credit for a conversion to the 1st touchpoint a customer encounters in their journey.

2. Last-touch attribution

This model assigns 100% credit for a conversion to the last touchpoint a customer encounters in their journey.

3. Linear attribution

This model distributes equal credit across all the touchpoints involved in the customer journey.

4. Time-decay attribution

This model gives more credit to the touchpoints that occur closer to the conversion event rather than the first touchpoint. The credit assigned to each touchpoint decays as time passes.

5. Position-based attribution

This model assigns maximum credit to the 1st & last touchpoints encountered by the customer. The remaining credit is divided among the touchpoints in between. It is also called as "U-shaped attribution".

6. Algorithmic attribution

This model employs statistical methods & complex algorithms to assign credits to different customer touchpoints.

7. Multi-touch attribution

This model considers multiple touchpoints throughout the customer journey & assigns credits proportionally.

8. W-shaped attribution

This model gives the most credit to the 1st touchpoint, last touchpoint, and mid-funnel touchpoint before a conversion.

9. Custom attribution

This model adjusts the weights of position-based interactions & gives more credit to the higher-value touchpoints & less credits to low-value touchpoints. Multiple attribution models can be combined into this model.

What are the factors to consider in choosing an attribution model?

Business goals & objectives
Available data & resources
Customer journey complexity
Attribution model flexibility
Industry & channel characteristics
Budget & cost considerations

What are the steps to select an attribution model?

Define your objectives - optimizing marketing spend, campaign effectiveness, understanding consumer behavior, etc.
Evaluate your customer journey - Gauge the complexity & the number of touchpoints involved.
Review available data - Identify the sources, quality, campaigns, pages, country, city, etc.
Understand attribution model types - Understand how each model assigns credit & the pros and cons associated with them.
Test & experiment - Use historical data & controlled experiments to compare the outcomes of different models.
Consider industry & channel characteristics - Different industries may require different attribution approaches.
Involve stakeholders - Engage stakeholders such as marketing teams, data analysts, decision makers, etc. in the selection process.
Monitor & iterate - Keep refining your attribution approach based on new data, market changes, etc.

What are the benefits of attribution modelling?

Customer behavior insights
Optimum resource allocation
Focussed marketing efforts
Enhanced personalization & targeting
Improved campaign planning & strategy
Measurement of marketing ROI
Data-driven decision making
Improved collboration & alignment
Long-term customer value optimization
Improved customer experience
Influence customers earlier on their path to conversion

What are single-touch & multi-touch attribution models?

Single-touch attribution models give all the credit for the conversion to a single touchpoint in the customer journey. Some examples are First touch model, Qualified Lead Model, Lead creation model, etc.

Mutli-touch attribution focuses on assigning every touchpoint along the consumer journey with an appropriate amount of value.Some examples are Linear attribution, Time decay attribution, etc.