Prodigy - Kentico Integration and Automation

Common Integration Use Cases Between Prodigy and Kentico

1. AI-Powered Content Tagging and Metadata Enrichment

Data flow: Kentico ? Prodigy ? Kentico

Content teams can send articles, product pages, images, and campaign assets from Kentico to Prodigy for annotation. AI and subject matter experts label content by topic, intent, product category, audience segment, or visual attributes. The enriched labels are then pushed back into Kentico to improve search, personalization, content recommendations, and campaign targeting.

Business value: Faster content discovery, better personalization, and more accurate content classification across large digital libraries.

2. Visual Asset Classification for DAM and Web Publishing

Data flow: Kentico ? Prodigy ? Kentico

Kentico-managed images and media assets can be routed to Prodigy for image labeling, such as identifying product types, scenes, logos, or compliance-sensitive elements. Once labeled, the metadata can be returned to Kentico to support asset reuse, automated gallery creation, and smarter media search.

Business value: Reduces manual tagging effort for marketing and web teams while improving asset governance and reuse.

3. Customer Intent and Feedback Labeling for Personalization Models

Data flow: Kentico ? Prodigy ? Kentico

Kentico captures behavioral data from forms, campaign interactions, chat transcripts, or page feedback. Selected text samples can be sent to Prodigy for annotation, such as intent classification, sentiment, or topic labeling. The resulting training data can be used to build models that improve personalization rules, content recommendations, and audience segmentation in Kentico.

Business value: Enables more relevant user experiences and better conversion performance through AI-driven audience understanding.

4. Product Content Quality Review and Taxonomy Standardization

Data flow: Kentico ? Prodigy ? Kentico

For organizations using Kentico for eCommerce or product content publishing, product descriptions, attributes, and supporting copy can be sampled and labeled in Prodigy to standardize taxonomy, identify missing attributes, and classify content quality issues. The validated labels can then be used to improve content templates and publishing workflows in Kentico.

Business value: Improves product data consistency, reduces publishing errors, and supports better merchandising and search results.

5. AI Training Data Collection from Campaign and Web Form Responses

Data flow: Kentico ? Prodigy

Kentico forms, surveys, and campaign landing pages generate large volumes of customer responses. These responses can be exported to Prodigy for labeling by category, urgency, lead quality, or issue type. Marketing, sales, and service teams can then use the labeled data to train models that automate lead routing, response prioritization, or content recommendations.

Business value: Speeds up response handling and improves lead qualification accuracy.

6. Active Learning Loop for Content Moderation and Compliance

Data flow: Kentico ? Prodigy ? Kentico

Kentico can provide user-generated content, comments, or submitted media to Prodigy for moderation labeling. Prodigy?s active learning workflow helps prioritize the most uncertain or risky items for review, making it easier to build moderation models for policy violations, inappropriate language, or restricted imagery. Approved labels can be used to automate future moderation decisions in Kentico.

Business value: Lowers manual moderation workload and strengthens brand and compliance controls.

7. Cross-Team Content Operations for AI Model Improvement

Data flow: Bi-directional

Content managers in Kentico can publish new content sets, while data science teams use Prodigy to label samples and refine AI models. Model outputs such as predicted categories, confidence scores, or content recommendations can be sent back to Kentico for editorial review and publishing decisions. This creates a continuous feedback loop between marketing, content, and AI teams.

Business value: Aligns editorial workflows with machine learning operations and accelerates AI adoption across digital experience teams.

8. Search Optimization Through Labeled Content and Query Data

Data flow: Kentico ? Prodigy ? Kentico

Search queries, zero-result searches, and content engagement data from Kentico can be exported to Prodigy for labeling by intent, topic, or relevance. These labels can be used to train models that improve internal site search, content ranking, and navigation paths within Kentico-powered experiences.

Business value: Improves findability of content and products, reducing bounce rates and increasing user engagement.