Case Study

How The Zebra Optimizes Funnel Conversion with Realtime Personalization

The Zebra, the nation's leading insurance comparison site, leveraged Sumatra to dramatically increase revenue per applicant by segmenting visitors in real time and delivering differentiated site experiences based on predicted intent.

Case Study

How The Zebra Optimizes Funnel Conversion with Realtime Personalization

The Zebra, the nation's leading insurance comparison site, leveraged Sumatra to dramatically increase revenue per applicant by segmenting visitors in real time and delivering differentiated site experiences based on predicted intent.

Case Study

How The Zebra Optimizes Funnel Conversion with Realtime Personalization

The Zebra, the nation's leading insurance comparison site, leveraged Sumatra to dramatically increase revenue per applicant by segmenting visitors in real time and delivering differentiated site experiences based on predicted intent.

Problem

Insurance buyers have wide-ranging coverage needs and buying intent

Balancing revenue strategies (online sales vs. phone sales vs. affiliates) is a struggle with a one-size-fits-all website experience.

Solution

Personalize the results page by segmenting customers in real time

The team delivered a solution that automatically identifies audiences and assigns the optimal buying experience based on conversion.

Result

Increased revenue per applicant and operational efficiency

By emphasizing the most appropriate revenue path, high intent buyers get a streamlined experience and phone support agents are used efficiently.

Problem

Insurance buyers have wide-ranging coverage needs and buying intent

Balancing revenue strategies (online sales vs. phone sales vs. affiliates) is a struggle with a one-size-fits-all website experience.

Solution

Personalize the results page by segmenting customers in real time

The team delivered a solution that automatically identifies audiences and assigns the optimal buying experience based on conversion.

Result

Increased revenue per applicant and operational efficiency

By emphasizing the most appropriate revenue path, high intent buyers get a streamlined experience and phone support agents are used efficiently.

Problem

Insurance buyers have wide-ranging coverage needs and buying intent

Balancing revenue strategies (online sales vs. phone sales vs. affiliates) is a struggle with a one-size-fits-all website experience.

Solution

Personalize the results page by segmenting customers in real time

The team delivered a solution that automatically identifies audiences and assigns the optimal buying experience based on conversion.

Result

Increased revenue per applicant and operational efficiency

By emphasizing the most appropriate revenue path, high intent buyers get a streamlined experience and phone support agents are used efficiently.

"Personalizing the experience and education is critical. That's where machine learning and artificial intelligence comes in. You can't build one product to serve everyone's needs."

Meetesh Karia

CTO

"The platform filled a critical technology gap. The Sumatra partnership is a core part of The Zebra's ongoing AI/ML strategy."


Claire Look

VP of Data

"Personalizing the experience and education is critical. That's where machine learning and artificial intelligence comes in. You can't build one product to serve everyone's needs."

Meetesh Karia

CTO

"The platform filled a critical technology gap. The Sumatra partnership is a core part of The Zebra's ongoing AI/ML strategy."


Claire Look

VP of Data

Problem: Insurance buyers have wide-ranging coverage needs and buying intent

Problem: Insurance buyers have wide-ranging coverage needs and buying intent

Consumer insurance buyers come to The Zebra with a wide variety of needs and intents, from seeking basic auto coverage to exploring complex home and life insurance bundles. Serving these different audiences with a one-size-fits-all survey funnel made it challenging for the company to organize a results page that catered to each visitor's specific requirements.

Further, the company's multiple revenue streams (online sales, phone sales, and affiliate partnerships) are appropriate for different customer segments. A common website experience made it difficult to effectively channel users toward the most appropriate revenue stream based on their buying intent. This resulted in missed revenue opportunities and inefficient use of resources, such as call center agents assigned to buyers more appropriate for self-service.

Finally, these audiences, characterized by intent, are not trivially identified using simple filters. Instead, The Zebra needed a solution that could extract these fuzzy customer segments automatically from the data.

Consumer insurance buyers come to The Zebra with a wide variety of needs and intents, from seeking basic auto coverage to exploring complex home and life insurance bundles. Serving these different audiences with a one-size-fits-all survey funnel made it challenging for the company to organize a results page that catered to each visitor's specific requirements.

Further, the company's multiple revenue streams (online sales, phone sales, and affiliate partnerships) are appropriate for different customer segments. A common website experience made it difficult to effectively channel users toward the most appropriate revenue stream based on their buying intent. This resulted in missed revenue opportunities and inefficient use of resources, such as call center agents assigned to buyers more appropriate for self-service.

Finally, these audiences, characterized by intent, are not trivially identified using simple filters. Instead, The Zebra needed a solution that could extract these fuzzy customer segments automatically from the data.

Solution: Personalize the results page by segmenting customers in real time

Solution: Personalize the results page by segmenting customers in real time

To address these challenges, The Zebra adopted Sumatra as the foundation of their personalization solution. Sumatra's realtime event processing and low-code data transformation capabilities filled a critical technology gap, enabling the team to serve timely customer data to audience segmentation models as well as retrain the models frequently, based on which customers saw which treatments, and whether or not they converted.

When an applicant completes the online survey funnel, a contextual bandits algorithm, assigns a variant of the results page, based on a handful of attributes relevant to predicting intent. The variants differ in their emphasis of the various revenue channels, e.g. displaying more ads more or less prominently. The bandit algorithm balances "exploration" (discovering how customer attributes influence conversion) vs. "exploitation" (optimizing revenue throughout the experiment.) The click conversion events captured by Sumatra are then used to update and improve the model on an ongoing basis.


To address these challenges, The Zebra adopted Sumatra as the foundation of their personalization solution. Sumatra's realtime event processing and low-code data transformation capabilities filled a critical technology gap, enabling the team to serve timely customer data to audience segmentation models as well as retrain the models frequently, based on which customers saw which treatments, and whether or not they converted.

When an applicant completes the online survey funnel, a contextual bandits algorithm, assigns a variant of the results page, based on a handful of attributes relevant to predicting intent. The variants differ in their emphasis of the various revenue channels, e.g. displaying more ads more or less prominently. The bandit algorithm balances "exploration" (discovering how customer attributes influence conversion) vs. "exploitation" (optimizing revenue throughout the experiment.) The click conversion events captured by Sumatra are then used to update and improve the model on an ongoing basis.


Result: Increased revenue per applicant and operational efficiency

Result: Increased revenue per applicant and operational efficiency

The Zebra was able to deliver on their Continuous Automated Personalization initiative using the Sumatra engine to:

  • Ship multiple experiences that respond in real time to customer data to optimize the buying experience

  • Quickly deploy new experiments and use cases to measure their impact on conversion rates

  • Dramatically increase revenue per applicant while reducing call center operational costs

The partnership with Sumatra has played a significant role in helping The Zebra reach their Series D milestone "to create personalized experiences including results informed by machine learning" as part of their overall growth strategy.

The Zebra was able to deliver on their Continuous Automated Personalization initiative using the Sumatra engine to:

  • Ship multiple experiences that respond in real time to customer data to optimize the buying experience

  • Quickly deploy new experiments and use cases to measure their impact on conversion rates

  • Dramatically increase revenue per applicant while reducing call center operational costs

The partnership with Sumatra has played a significant role in helping The Zebra reach their Series D milestone "to create personalized experiences including results informed by machine learning" as part of their overall growth strategy.

Start optimizing your revenue streams

Start optimizing your revenue streams

Need to target different audiences with different monetization strategies? We're here to help.

Need to target different audiences with different monetization strategies? We're here to help.

Get started

Sign up now to lock in early adopter pricing

Get started

Sign up now to lock in early adopter pricing

Get started

Sign up now to lock in early adopter pricing