Closing the Monetisation Gap. Smart Clipping from Reuters Imagen

Enhancing video discovery & monetisation with Smart Clipping and metadata enrichment. 

By the end of 2025, the global volume of data is projected to rise to 181 zettabytes¹. In the broadcast and media landscape, the volume of video content being produced and consumed is exploding. Synchronised with this is the demand for ‘real-time’ that is greater than ever, from streaming platforms and news broadcasters to sports franchises and user-generated content. Whilst part of the challenge is content storage, the real challenge is making the content discoverable and monetisable for your partners.

Finding and isolating that one specific clip from the full video footage of that sporting event, or having to curate a collection of clips from that same event to fit a narrative should be easy. However, imagine having to do this across all the games or events in a season, across multiple seasons, going back years and the ‘narrative’ constantly changing based on your rights holders’ needs. To make things more interesting, imagine the variety of people and file naming conventions that have touched those assets in that time. Feeling anxious? If that wasn’t enough, now imagine commercial teams asking you for a clip of the winning moment with ‘that’ logo in the background. The panic, the time required to shuffle through assets is a reality for many. What value does a video have if no one ever sees it? Broadcast partners and rights holders need access the content quickly, at a time when they need it to curate their story, or the risk is they go elsewhere. Similarly, fans expect access to multiple streams of content whilst in the action.

• Most enterprise information is unstructured and therefore hard to search or operationalise. Industry analyses consistently estimate roughly 80% of enterprise data is unstructured, which means critical insight and value often remain buried in files, images, and video. (cdomagazine.tech)
• A large share of created content never gets used. The widely cited SiriusDecisions finding (now part of Forrester) estimates that 60–70% of B2B content goes unused—often because teams can't find it or don't trust it. (forrester.com)

Reuters Imagen video management system and our expertise offers powerful capabilities that allow you and your partners to instantly discover, licence and download high quality clips of specific ‘key’ moments from a larger library. Features like smart clipping enable far easier and quicker discovery, production and distribution workflows for key moments like winning celebrations, goals, wickets, tee shots, aces, players, overtaking, victory moments, capsizing and more. 

These capabilities are particularly valuable in sports and media broadcasting environments where real time is king, and the rapid provision of key moments is essential for production and distribution workflows to partners for licencing. Benefits that a successful metadata enrichment strategy can deliver include:  

1. Improved search precision enabling discoverability and accessibility. 
2. Enhanced governance and compliance. 
3. Reduced manual effort. 
4. Reduced reliance on tribal knowledge. 
5. Streamlined workflows and processes to maximise value from existing assets.  

Foundational to all of this is a strategy and consistent implementation of metadata enrichment that facilitates the discovery and utilisation of ‘key’ moments according to your preferred schema. Successfully enriching your metadata can have a significant and lasting impact upon your content management, distribution and ability to monetise your content.  

The Reuters Imagen approach 

Reuters Imagen have extensive experience and process in establishing metadata, annotation standards and schema across sports and media. We commonly work to a 5-step checklist to achieve a lasting solution for our customers, the objectives of which are: 

Our delivery approach follows these key phases:  

Discover and evaluate client content and identify enrichment approach opportunities for optimised and consistently structured annotations. 

2. Assessment and Recommendations:  
Define consistent data models and formats for annotations regardless of their source (manual logging, data feeds, or AI services). Identify gaps, inconsistencies, and opportunities. Make recommendations based on enrichment approach, considering outcome preferences.  

3. Solution Design:  
Create custom integration plans based on client needs, aligning metadata goals with business objectives. 

4. Implementation:  
Integrate, build and test the enrichment connections ensuring standardised approach and quality checks. 

5.Knowledge Transfer:  
Ensure client teams can maintain and leverage the enhanced system. Schedule periodic audits and updates to maintain metadata health, as required. 

Metadata enrichment 3 ways  

AI tagging uses machine learning and computer vision to automatically detect key moments (e.g. goals, celebrations, etc.) and generate time-coded metadata. The Reuters Imagen API provides integration options with the leading providers of AI generated metadata.  

Pros: 

• Speed & Scalability: Processes large volumes of content in real time or near real time.

• Consistency: Reduces human error and subjectivity. 

• Cost-effective at scale: Once set up, it reduces the need for manual teams. 

• Real-time applications: Ideal for live sports, news, and events. 

Cons: 

• Accuracy limitations: May misidentify or miss nuanced moments, especially in complex or less structured content. 

• Training required: Needs well-labelled training data and tuning for specific sports or formats. 
• Limited context awareness: Struggles with interpreting emotion, sarcasm, or subtle narrative cues. 
   

Production staff manually log events and add time-coded metadata during or after an event via a user-friendly tool that allows users (e.g., journalists, producers) to tag and annotate their own media content. Featuring one-touch metadata entry against a pre-defined schema for standardisation and timeline annotated chapter creation. 

Pros: 

• High accuracy and nuance: Humans can understand context, emotion, and subtleties that AI might miss. 

• Customisable: Easily adapted to unique editorial or production needs. 

• Editorial judgment: Useful for storytelling and narrative-driven content.  

Cons: 

• Manual: Relies on users taking the time to tag content. 

• Inconsistent usage: Quality and depth of metadata can vary depending on user diligence. 

• Costly: Requires trained personnel, especially for live events. 
  

3. Data Feed Integration & Parsing 
Integrates structured data from various third-party providers (like Opta for sports stats), analytics or user interactions into metadata, often in real time, into the Reuters Imagen system.  

Parsing processes a defined selection of this integrated data. It extracts relevant information from these data feeds, which could include keywords, timestamps, or other identifiers that are important in your metadata enrichment.  

Pros: 

• Highly structured and reliable: Professional data feeds are accurate and standardized. 

• Real-time enrichment: Enables live tagging and automated updates. 

• Rich contextual data: Adds depth (e.g., player stats, match events) to video content. 

Cons: 

• Dependency on third parties: Subject to licensing costs and data availability. 

• Integration complexity: Requires technical setup to parse and align data with media timelines. Integrating different sources of data feeds require standardisation.  

• Limited to structured events: Best suited for sports or domains with well-defined data models. 
  
  

Smart Clipping 

Smart clipping (and indeed many applications) functionality relies upon effective metadata enrichment strategies providing quality and structure of annotations associated with media assets. Annotations serve as the foundational metadata that enables precise searching, filtering, and clip generation.  

Smart Clipping is a feature designed to streamline the process of creating video clips from longer media assets. It allows users to quickly and intuitively select segments of video content, generate clips, and licence them without needing advanced editing tools. This functionality is particularly valuable for media, broadcasters, and sports federations who work with large volumes of video and need to extract highlights or relevant moments efficiently. By simplifying the clipping process, Smart Clipping reduces the time and technical skill required to produce key moment content, enabling faster turnaround for content. Its value lies in enhancing productivity, improving content accessibility, and supporting real-time content licencing across partnerships and organisations. 

Conclusion

The right metadata enrichment approach for you depends on your goals, scale, and content type. Whether you’re monetising historic sports footage, or managing large-scale, live video, investing in the right metadata strategy can turn passive storage and ingest into active value. We recognise that there is not often one size fits all and so offer expert consultations to partner with you to identify the right approach for your needs. Contact us to arrange a free consultation on your specific needs and visit our Smart Clipping page to see why content enrichment matters and what it can do for you.

1. Accessed July 2025. https://rivery.io/blog/big-data-statistics-how-much-data-is-there-in-the-world/