Guide to video analytics for OTT Platform: Key metrics

OTT platforms process millions of video streams every day, but how do they determine which content resonates with viewers? Whether it's the latest season of Stranger Things or a new release like Hitman, understanding audience preferences is needed. Video analytics drives these insights, helping platforms like Netflix and Prime Video refine content delivery and enhance viewer experience.

Video analytics involves collecting, processing, and analyzing data from video streams to generate actionable insights. These insights provide a deeper understanding of viewer behavior, content performance, and technical details like buffering times and device usage, allowing platforms to optimize every aspect of content delivery.

‍

‍

Understanding video analytics in OTT

‍

Core metrics for video analytics

Even if you're not deeply technical, video analytics can provide you with valuable insights through intuitive, user-friendly metrics that help monitor platform performance.

• Viewership (Total views, Unique viewers): These metrics give you a sense of how many times content has been watched. Total views count all plays, while unique viewers tell you how many new individual users have engaged. These insights reveal which content is most popular, helping to guide future production and marketing decisions.
• ‍Watch time (Session duration, Completion rate): Session duration measures how long viewers stay on a video, and completion rate reveals how many watch it to the end. This helps content creators understand how engaging their material is and whether it resonates with viewers.
• ‍Engagement: Monitor viewer engagement metrics, including watch time, drop-off points, and interaction rates. This data provides insights into how viewers interact with your content and where improvements can be made. To add-on, likes, comments, and shares on the front-end also offer a direct reflection of how audiences are interacting with content. Higher engagement signals that content resonates and has the potential to reach a wider audience.

‍

Advanced metrics for OTT  

For engineers managing OTT platforms, technical metrics dig deeper into stream performance and the underlying infrastructure.

• Bitrate, buffering events, rebuffering ratio: Bitrate measures the quality of the video stream, while buffering events track interruptions. Monitoring the rebuffering ratio can highlight issues with delivery networks, helping to ensure seamless playback.

‍

‍

For example, this graph depicts the number of buffering incidents over the months of January, February, and March. It shows a clear downward trend, with incidents decreasing from 150 in January to 100 in March. This suggests that there has been a consistent improvement in buffering performance over this period.
‍

• Device and network analysis (Wi-Fi vs mobile data): Understanding whether viewers are streaming via Wi-Fi or mobile data is essential. Different networks affect video quality and playback times. Adaptive bitrate streaming helps address these issues, automatically adjusting video quality based on available bandwidth.

‍

How video analytics transform OTT platforms

‍

Optimizing user experience and personalization

Video analytics allows OTT platforms to deliver a smoother, more personalized experience by understanding user behavior and preferences.

• Improved recommendations: Platforms like Netflix and Prime Video rely on video analytics to deliver personalized content recommendations. By tracking what users engage with, platforms can curate custom suggestions, keeping viewers on-site longer.
• ‍Enhanced UI/UX: Analytics highlight problem areas in the user interface. For example, high drop-off rates at specific sections may indicate issues with navigation or usability, guiding design improvements to create a more intuitive experience.
• ‍Predicting content trends: Platforms can use analytics to predict which content will drive engagement based on trends, seasonality, and time of day. This ensures that popular content is highlighted at the right time to maximize views.

‍

Monetization strategies with video analytics

Analytics are also key to refining monetization efforts through smarter ads, subscription strategies, and targeted marketing.

• Impact on ads: Platforms use video analytics to deliver highly targeted ads. These ads are based on viewer behavior, increasing engagement and, consequently, revenue. Real-time data allows for adjusting ad frequency and placement to ensure ads are neither intrusive nor ineffective.
• ‍Subscription models: Analytics help track which subscription tiers are performing better and where users are most likely to upgrade or churn. Predictive analytics further enhance retention efforts by offering personalized subscription offers or discounts.
• ‍Targeted marketing: With rich data on user preferences and habits, OTT platforms can create laser-focused marketing campaigns. Personalized content suggestions, in-app notifications, and promotional offers boost both subscription rates and ad revenue.

‍

Tools and technologies for video analytics

OTT platforms utilize various tools to track user data and video performance. Here's a look at some key tools:

• Google analytics: This tool provides basic metrics like session duration and viewer retention, but lacks deeper video-specific insights, such as buffering events or playback quality.
• ‍Mixpanel: With event-based tracking, Mixpanel offers a more granular view of user interactions, including video play, pause, and completion events. Its funnel analysis reveals drop-off points for a more detailed understanding of user behavior.
• ‍FastPix Video Data: A dedicated tool for advanced video analytics, FastPix provides real-time data on viewer engagement, buffering, and playback success. It’s useful for diagnosing technical issues in video like slow start and optimizing content delivery through predictive analytics.

‍

‍

How to Access Video Stats on OTT Platforms

‍

Many OTT platforms provide ways to access video stats:

Netflix: On Mac, press Ctrl + Option + Shift + D; on Windows, press Ctrl + Shift + Alt + D to view detailed stats, including bitrate and buffer state.

‍

‍

YouTube: Right-click the video and select "Stats for Nerds" to display resolution, buffering details, and network activity.

‍

‍

Prime Video: Use browser developer tools to access network performance and video stats like bitrate and buffering.

‍

‍

Technical tips for developers implementing video analytics

‍

1. Optimizing adaptive bitrate (ABR) streaming for regional viewers:

Challenge: Viewer regions often have varying internet speeds and device capabilities, which can affect the video quality delivered.

Best practice: Implement region-specific ABR profiles. This involves setting up multiple bitrate ladders customized for different regions. For example, in regions with slower internet connections, optimize for lower bitrates but maintain a balance in video quality to avoid excessive buffering. In regions with higher internet speeds, higher bitrate ladders can provide viewers with ultra-HD video.

Technical implementation: Use analytics tools to gather real-time network conditions and device capabilities, and leverage those insights to dynamically adjust ABR profiles. Platforms like FastPix can provide detailed network insights, helping engineers to fine-tune ABR settings for optimal performance.

‍

2. Real-time buffer monitoring and prediction:

Challenge: Buffering is one of the biggest user complaints, leading to churn if not managed effectively.

Best practice: Set up real-time monitoring of buffering events and use predictive analytics to pre-emptively adjust video delivery. By tracking past buffering events and network congestion, the system can proactively switch to a lower bitrate before the buffer occurs, ensuring smoother playback.

Technical implementation: Integrate tools like FastPix or Mux to monitor rebuffering ratios and network conditions. Use machine learning algorithms to predict buffering risks based on historical data, allowing the system to adjust streaming settings in real-time.

‍

3. Device-specific video analytics:

Challenge: Different devices (smartphones, TVs, tablets) have varied playback capabilities and viewer interaction styles.

Best practice: Customize streaming profiles and data collection based on the type of device. For example, mobile viewers might prioritize low latency over high resolution, while smart TV users prefer 4K content. Engineers should track device-specific metrics, such as screen resolution, processing power, and network type (Wi-Fi vs. mobile data).

Technical implementation: Use analytics tools that segment data by device and network type. This enables engineers to fine-tune video delivery, offering device-specific optimizations such as adjusting buffer sizes or lowering initial video bitrates for mobile users.

‍

4. Reducing latency with edge caching:

Challenge: Latency can significantly impact user experience, especially during live streaming.

Best practice: Deploy edge caching solutions that bring content closer to users. By leveraging CDNs (Content Delivery Networks) with geographically distributed edge servers, you can reduce latency and improve video start times.

Technical implementation: Engineers should monitor CDN performance using video analytics to identify which regions experience higher latency and adjust caching policies accordingly. Tools like FastPix can provide insights into CDN performance by measuring latency, rebuffering events, and regional playback delays.

‍

5. Implementing granular video segment tracking:

Challenge: Identifying exact points where viewers drop off during a video can help pinpoint content issues.

Best practice: Use segment-based tracking to analyze viewer behavior at a granular level. Track viewer engagement at specific timestamps to determine which scenes or segments lead to higher drop-off rates. ‍

Technical implementation: Set up video analytics to log viewer interaction at a segment level, collecting data on pause events, fast-forward actions, or exits. Tools like Mixpanel or FastPix allow for this level of detail, helping engineers refine content delivery or adjust ad placement.

‍

‍

Global audience insights: Using video analytics for a worldwide reach

‍

OTT platforms serve a global audience with diverse preferences, languages, and devices. It helps platforms deliver personalized content and optimize performance across different regions.

‍

1. Language preferences and localization

• Insight: Video analytics helps identify the most popular languages for audio, subtitles, and dubbing. Platforms can use this data to localize content and enhance regional engagement.
• ‍Example: Platforms like Netflix track the usage of language settings and subtitles. If Spanish subtitles are frequently selected in a particular region, more localized Spanish content can be promoted to meet demand.

‍

2. Regional content preferences

• Insight: Different regions have distinct tastes in genres and content themes. Analytics reveal what resonates in specific markets, helping platforms prioritize the right content for each region.
  
• ‍Example: A spike in Korean dramas in Western markets prompts platforms to invest more in localizing and recommending this genre to a wider audience, based on viewership trends.

‍

3. Device usage by region

• Insight: Viewers in different regions use a variety of devices to stream content, from smartphones to smart TVs. Analytics track which devices are most common, guiding optimization efforts.
• ‍Example: In regions where mobile devices are dominant, platforms can prioritize mobile-friendly streams with adaptive bitrate adjustments to ensure smooth playback on slower networks.

‍

4. Bandwidth optimization

• Insight: Internet speeds and network quality vary globally, affecting video quality. Analytics help platforms adjust bitrate and compression based on regional bandwidth capabilities, minimizing buffering and improving performance. ‍
• Example: In regions with slower mobile networks, platforms can serve lower-bitrate streams to reduce buffering. In areas with faster broadband, higher-quality streams are delivered to enhance the viewing experience.

‍

5. Predicting regional content demand

• Insight: Platforms can use video analytics to forecast demand for specific types of content based on regional trends, seasons, and events. This helps platforms curate content and optimize recommendations.
• ‍Example: During major global events like the World Cup, analytics can predict an increase in sports-related content, prompting platforms to feature relevant content and adjust marketing strategies accordingly.

‍

‍

Wrapping up…

‍

Video analytics is key for OTT platforms, helping them understand what viewers like and improve their services. By tracking metrics such as viewership patterns, buffering issues, and regional preferences, OTT streaming platforms can create a more personalized and smooth viewing experience.

‍

‍

FastPix boosts this process with its detailed video data API, offering features like audience metrics, play metrics, and quality-of-experience analytics. This helps platforms quickly fix technical problems and fine-tune content delivery. As your OTT platforms grows, using FastPix’s video infrastructure will be essential for staying ahead and meeting the needs of a diverse global audience.