AI-Powered Personalization Boosts Podcast Engagement 7×

Podcast Platform Boosts Engagement 7× Using AI Recommendations

A leading podcast platform partnered with Dataforest to replace manual recommendations with an AI-powered personalization engine. The new system analyzes user behavior and context in real time to deliver tailored suggestions in under 0.5 seconds, handling up to 20 recommendations per second. This resulted in 7× higher user engagement, enhancing listener experience and significantly increasing the client’s revenue.

Media

Data Science

Data Insights & Forecasting

higher user engagement

0.5

secs

average recommendation delivery speed

recommendations/sec throughput

The client is a prominent web and mobile podcast platform based in Saudi Arabia, widely recognized across the MENA region for delivering high-quality, authentic journalism across text, visual, and audio formats. It operates as part of a major media holding that manages multiple outlets spanning news, digital platforms, and social media channels.

Databricks

TensorFlow

Spark

PostgreSQL

Databricks vector search

THE CHALLENGE

The podcast platform faced very limited personalization due to its manual recommendation process

The podcast platform relied on static, manually curated recommendations that couldn’t adapt to user behavior. This restricted engagement, slowed revenue growth, and left the platform behind competitors who leveraged dynamic personalization.

No Adaptability to User Behavior

Recommendations weren’t responsive to user preferences, resulting in low engagement and poor discovery of new content.

Missing Real-Time and Unified Data Infrastructure

The system lacked pipelines to process and unify data across sources, preventing real-time insights and consistent content delivery. This also led to data duplication — with content coming from multiple sources, users often received recommendations for podcasts they had already seen on other platforms.

Cold Start Problem for New Users

With no recommendation system for users without history, new listeners had a poor first experience.

Scalability Gaps

The legacy approach couldn’t support rapid growth in users and content, limiting future expansion.

Boost Engagement 7× with AI-Powered Personalization

Validate your strategy with a 1-Week PoC

THE SOLUTION

AI-Powered Recommendation Engine with 7x Engagement Boost

We built a flexible recommendation model that processes diverse user signals in real time.

The solution architecture included four interconnected recommendation models based on two-tower architectures, each designed to process different user signals and content types.

It delivers highly relevant podcast suggestions, improving user engagement by 7x and enabling scalable growth.

Real-Time and Unified Data Infrastructure

We developed automated ETL pipelines that continuously collect, clean, and synchronize data from all media channels — including the website, social platforms, and radio. This eliminated data duplication and enabled instant processing for consistent, high-quality recommendations across all user touchpoints.

Contextual Cold Start Solution

For new listeners without behavioral history, we introduced a context-aware recommendation module that leverages signals like time of day, device type, location, and language. This ensured that even first-time users received meaningful, personalized recommendations — improving retention from the very first session.

Adaptive Personalization Engine

To address low adaptability to user behavior, we implemented four recommendation models built on two-tower architectures. These models analyze multiple user signals such as listening history, engagement patterns, and metadata, dynamically adjusting to user preferences and delivering relevant podcast suggestions in real time.

Modular and Scalable Architecture

We designed the system as a modular, multi-model framework capable of expanding with the platform’s growing user base and content library. Each recommendation model functions independently but integrates seamlessly through a centralized ranking layer, ensuring consistent performance and scalability as data volume increases.

THE RESULT

AI-Powered Personalization: 7x More Podcast Engagement and 30% Higher User Interaction

A leading podcast platform in Saudi Arabia and the MENA region needed to replace its static, manually curated recommendations to drive growth and user satisfaction. DATAFOREST delivered a scalable AI-powered recommendation engine that personalizes podcast suggestions in real time based on user behavior, preferences, and context.

We developed a modular system with a learning-based ranking model and a real-time data pipeline to process user activity efficiently. Key challenges included scaling the architecture, integrating diverse interaction signals, handling data duplication across multiple content sources, and solving the cold start problem for new users. These were addressed by unifying data streams through real-time ETL pipelines, applying deduplication logic to eliminate repeated content, and combining behavioral data with contextual metadata (e.g., time, language, location). Recommendations now auto-update every 48 hours, ensuring ongoing relevance and eliminating manual work.

This transformation enabled the podcast to personalize content at scale, increase user satisfaction, boost revenue, and future-proof its platform with a flexible, data-driven solution.

<0.5 sec

average recommendation delivery speed

~20

personalized recommendations processed per second

7×

higher user engagement compared to the manual system (A/B tested)

Podcast Platform Boosts Engagement 7× Using AI Recommendations

Steps of providing data scraping services

Step 1 of 5

Free consultation

It's a good time to get info about each other, share values and discuss your project in detail. We will advise you on a solution and try to help to understand if we are a perfect match for you.

Step 2 of 5

Discovering and feasibility analysis

One of our core values is flexibility, hence we work with either one page high level requirements or with a full pack of tech docs. At this stage, we need to ensure that we understand the full scope of the project. Receive from you or perform a set of interviews and prepare the following documents: list of features with detailed description and acceptance criteria; list of fields that need to be scraped, solution architecture. Ultimately we make a project plan which we strictly follow. We are a result-oriented company, and that is one of our core values as well.

Step 3 of 5

Solution development

At this stage, we develop the scraping engine core logic. We run multiple tests to ensure that the solution is working properly. We map the fields and run the scraping. While scraping, we keep the full raw data so the final model can be enlarged easily. Ultimately we store data in any database and run quality assurance tests.

Step 4 of 5

Data delivery

After quality assurance tests are completed, we deliver data and solutions to the client. Though we have over 15 years of expertise in data engineering, we expect client’s participation in the project. While developing and crawling data, we provide midterm results so you can always see where we are and provide us with feedback. By the way, a high-level of communication is also our core value.

Step 5 of 5

Support and continuous improvement

We understand how crucial the solutions that we code for our clients are! Our goal is to build long-term relations, so we provide guarantees and support agreements. What is more, we are always happy to assist with further developments and statistics show that for us, 97% of our clients return to us with new projects.