#FutureReadyHealthcare
Indegene’s AI solution automates content deconstruction and tagging for a global pharmaceutical organization, resulting in an 85% reduction in efforts
The Customer
A global pharmaceutical organization with more than 20 brands spanning more than 150 markets was looking to launch a new modular approach to content creation. They planned to build a library of atomized components from existing content. However, their existing manual content deconstruction and tagging process faced limitations, thus causing delays.
Challenges
Inconsistent deconstruction and tagging: The manual content deconstruction and tagging process, coupled with manually entering data into systems of record, led to inconsistent tag generation. Moreover, differences in the interpretation and taxonomy of tags also created inconsistencies and confusion throughout the workflow.
Poor scalability: The organization’s manual process struggled to scale-up to meet deadlines and was inadequate for accommodating future content growth.
Resource-intensive operations: Manual content deconstruction and tagging required the increased involvement of medically trained personnel, making it difficult to staff teams based on varying demand.
Handling content in multiple languages: Tagging assets in multiple languages required recruitment and extensive collaboration among language experts and medical specialists.
The Solution
Taxonomy and deconstruction strategy definition: A key step to modular content activation is deconstruction and tagging based on a taxonomy that is comprehensive and granular, suits content nuances, and enables derivative content authoring with minimal efforts. Indegene worked with the organization to define and align their taxonomy and deconstruction strategy.
NEXT Commercial Content Intelligence (NCCI) deployment, customization, and model development: Indegene’s proprietary artificial intelligence (AI) platform for content deconstruction and tagging, NCCI, was introduced and customized to align with the defined taxonomy and strategy. Six new machine learning models were custom-built to enable automated deconstruction and tagging. NCCI was also configured to handle content from multiple asset types, languages, brands, and markets.
Library enablement for deconstructed content: NCCI was utilized to automatically deconstruct and tag content, which was then integrated with the organization’s digital asset library application to build an easy-to-use content library for multiple markets.
Completeness and reusability: NCCI’s source file processing capabilities generated high-quality and reusable components in multiple formats, including images. These components were then linked to the necessary references, footnotes, and glossaries to ensure completeness and referentiality for modular content authoring.
Workflow automation: The automated process significantly reduced manual efforts, improving the efficiency, usability, and reliability of the organization’s content production process.
Unified global strategy enablement: Implementing a unified and user-friendly solution across markets and different asset types created uniformity and consistency with the organization’s global content strategy.
Downstream integration: Seamless integration with downstream applications ensured that the organization’s modular content strategy was implemented without extensive change management.
Outcomes
After a successful pilot in a single market, the organization expanded NCCI-powered automated content deconstruction and tagging to multiple key markets and languages. NCCI was also integrated into their content creation process to tag newly generated content, enhancing the discoverability and use of content by teams around the world and enabling them to create engaging new content.
80%
reduction in time required to extract and tag content
<1%
client change requests
<5%
human intervention in the automated workflow