brage

5 min read

NAVER WEBTOON boosts CRM strategies with Braze Cloud Data Ingestion

TopicsStreaming Data
IndustryMedia & Entertainment
ProductData Ingestion
NAVER WEBTOON boosts CRM strategies with Braze Cloud Data Ingestion
Problem

NAVER WEBTOON aimed to improve user retention and LTV (lifetime value) by connecting its proprietary user modeling data to its customer engagement campaigns.

Strategy

By connecting its proprietary user modeling data with Braze, NAVER WEBTOON was able to send personalized messages to different cohorts, focusing on users likely to bounce or purchase.

Results

NAVER WEBTOON boosted retention, LTV, and workflow efficiency with CRM strategies powered by machine learning.

company
INDUSTRY
PRODUCTS USED
BY THE METRICS

5,000,000

Users auto-synced in 10 minutes

640

hours saved monthly by streamlining CRM processes

NAVER WEBTOON is part of WEBTOON Entertainment Inc., a leading global entertainment company and home to some of the world’s largest storytelling platforms. WEBTOON Entertainment and webcomics have exploded in popularity around the world, creating a new generation of comic creator superstars.

The company’s content has also become an important source of IP for the global entertainment industry, with its two studios—Wattpad WEBTOON Studios in the United States and Korea’s Studio N—producing a string of hits on global streamers, including Netflix, Disney+, Prime Video, and Crunchyroll, along with local projects from Korea and across the Asian market.

NAVER WEBTOON leverages predictive modeling data to personalize messages at key user touchpoints. The "Perfect for you! (알아서 딱!)" feature, powered by its proprietary user modeling data, boosted read counts and expanded the range of genres consumed, resulting in more engaged readers and purchases.

Building on its success, NAVER WEBTOON extended the application of its modeling data from in-app services to its customer engagement strategy. To do this, the team seamlessly integrated two key data sets, both generated by NAVER WEBTOON's in-house cohort system, with Braze:

  1. User modeling data
  2. Recommendation modeling data

CRM strategy powered by machine learning

NAVER WEBTOON's user modeling data forms the backbone of its targeted CRM strategy. This data, generated by its in-house cohort system, combines various engagement metrics with advanced predictive modeling powered by machine learning (ML). The team utilizes this rich data set to create highly-segmented user groups based on visit frequency, read counts, purchase history, and engagement rates

NAVER WEBTOON’s Predictive User Modeling Data Groups

  1. Viewer churn likelihood: Predicts which users may stop reading specific comics and targets them with personalized recommendations.
  2. Abandonment risk: Identifies users who've stopped reading comics altogether and may leave the platform entirely.
  3. Reactivation potential: Pinpoints inactive users who are likely to return if given tailored content prompts such as personalized promotional messages.
  4. In-app currency usage: Targets users with high upselling potential based on predicted spending amount within a given period.
  5. Reward Ads engagement pattern: Identifies users at risk of disengaging from Reward Ads in the next 14 days based on the Reward Ads usage for the last 4 weeks.

NAVER WEBTOON shifted from post-campaign analysis to proactive CRM campaigns using its predictive models. To strengthen this strategy, the company integrated its predictive user modeling data with Braze, enabling more precise and efficient audience targeting in CRM campaigns.

"We've seen remarkable results with our ML-powered CRM campaigns. They've significantly boosted our user retention and LTV. Our A/B tests clearly show that targeting based on our proprietary user modeling data outperforms previous methods across all metrics including CTR, views, and purchases. Seeing these successes firsthand, we're now expanding our predictive models across more features and developing more custom models.”

— Marketing & Ads Intelligence Team

Integrating user modeling data into CRM campaigns

NAVER WEBTOON built a streamlined pipeline to leverage user modeling data in its CRM campaigns. This pipeline allows NAVER WEBTOON to efficiently use its rich user data for targeted and personalized CRM campaigns through Braze.

image

NAVER WEBTOON’s cohort system and CDI Integration:

  1. NAVER WEBTOON's cohort system generates user modeling data and sends it to Amazon S3.
  2. This data is formatted for integration with Braze Cloud Data Ingestion (CDI) and stored in Amazon Redshift.

The processed data is then sent from Amazon Redshift to Braze using Cloud Data Ingestion integration.

image

“By automating CRM processes with Braze CDI, we saved 640 work hours monthly. Our new workflow syncs up to 5 million users to Braze in just 10 minutes with a single click.”

— Marketing & Ads Intelligence Team

Leveraging user modeling data as target audiences

NAVER WEBTOON gathers valuable user modeling data by compiling the various cohorts that each user belongs to. This approach enables the company to store data in a multi-dimensional array format, which facilitates flexible and dynamic user categorization.

By organizing user data in this way, NAVER WEBTOON can easily segment their audience based on diverse attributes and behaviors, allowing for more targeted and personalized engagement strategies. This level of granularity enhances their ability to deliver relevant content and improve user experiences across the platform.

image

Here’s how they do it:

1. When creating a campaign, NAVER WEBTOON selects the most appropriate cohort as the target audience based on the campaign's objective.

a screen showing additional filters including app language and indonesian

2. NAVER WEBTOON can then create targeted campaigns leveraging its predictive models:

  • Churn prevention campaign: Targets users likely to stop reading and recommends a new comic to read.
  • Reactivation Campaign: Targets users at high risk of abandoning the platform altogether, prompting them to revisit their favorite comics.
  • Upsell Campaign: Targets users most likely to increase or decrease their spending.

"We've integrated various user modeling data types with Braze for scalability. This allows us to easily incorporate new ML-powered predictive models into our CRM system. For example, we recently integrated our Reward Ads churn prediction model into Braze. By designing for scalability from the start, we can quickly adapt our CRM capabilities to evolving marketing needs."

— Marketing & Ads Intelligence Team

Key takeaways

  1. Scalable data integration: NAVER WEBTOON developed a robust system to seamlessly integrate their in-house modeling data with CDI, enabling agile and responsive CRM efforts.
  2. Streamlined workflow: By automating cohort syncing through Braze CDI, NAVER WEBTOON saved an impressive 640 work hours each month, allowing their team to focus on strategic initiatives.

NAVER WEBTOON is a customer of AB180, the official Braze reseller in South Korea.



ALLOYS PARTNERSHIPS
company

Check out more Case Studies

View Case Studies

Join the movement to journey orchestration.

The move to highly-intelligent, always-on journey orchestration is happening. And much of it is happening on our platform. Join brands of all sizes who are taking the craft of customer engagement to the next level.