As AI advances, technology is improving itself through continuous evolution-including the enhancement of the master data that fuels it.
By Dave Wargo, Partner, Columbus Consulting
The retail industry is racing to integrate Artificial Intelligence (AI) and Machine Learning (ML) into business strategies. While the potential for transformation is enormous, scaling and operationalizing AI use cases remains difficult. Legacy systems, stagnant mindsets, siloed data structures, and unclear enablement pathways all stand in the way.
Adding to these obstacles is the one issue many do not want to confront directly: the data problem. Without clean, consistent, and connected data, even the most advanced AI initiatives are destined to underperform.
Several Challenges of Retail Data Today
Three key challenges exist for retailers to enable their data:
- Silos and Lack of Ownership and Governance: Many retailers operate with fragmented systems and siloed data spread across departments, leading to manual workarounds, duplication, and frequent errors. Business leaders often assume their data is ready, while IT teams are left managing chaos—resulting in misaligned expectations and stalled progress. Lack of business ownership and governance of the data is a common challenge.
- Reliance on Manual Efforts: Employees can spend hours each day attempting to detect or correct data errors due to conflicting sources, missing data elements or inconsistent calculations. These issues can cascade through analytics efforts, undermining trust in AI models and preventing enterprise-wide adoption.
- Mismatched Solutions: Many retailers skirt the importance of a data management solution – a mistake. ERP solutions are then tasked with requirements that do not fit within core capabilities (e.g. data quality, data syndication). Reliable data can remain a blocker.
Clean Data as the Foundation for Retail AI/ML Success
Although not exhaustive, retailers need to take inventory of their data and related processes exploring the following areas:
- Master Data Hierarchy and Attribute Standardization
- Ensures data is business-ready through consistent policies and rules.
- Supports scalable data modeling, parent/child relationships, and automated governance processes.
- Master Data Dictionary
- Improves clarity across teams with consistent naming, timeframes, and definitions.
- Intelligent Flagging
- Enables automation and control by assigning standardized flags (e.g. Active, Recalled, Disposition).
- Supports execution across merchandising, compliance, and supply chain systems.
- Unified Documentation
- Promotes alignment and accountability via consistent templates for vendor/customer agreements and reporting.
- Reduces confusion and manual errors in key business documents.
- Standardized Processes
- Streamlines master data workflows (item, vendor, location, etc.) across business silos.
- Reduces redundant work and improves onboarding speed and accuracy.
- Data Quality Management
- Drives trust in analytics with proactive monitoring for accuracy, completeness, and consistency.
- Supports operational agility with real-time data profiling and cleansing.
- Data Governance
- Establishes ownership, protocols, and approval chains for managing the data lifecycle.
- Integrates governance into cross-functional teams with executive sponsorship.
- Robust Data Solutions
- Enables data readiness through PIM, MDM, and PXM platforms.
- Ensures data can be sourced, secured, moved, and leveraged efficiently across systems.
Now that we know some challenges and requirements, how can AI technology help?
Is there a way for AI to support enabling AI? Can AI be used not just as the end goal but as part of the solution—automating the data cleanup, enrichment, classification, and governance work required to unlock its own potential? If retailers enable intelligent data management tools to elevate the quality of their data, they can better support a scalable, AI-powered transformation.
How AI Can Improve Data Quality
Most ERPs, as noted above, cannot manage complex domains, enrichment at scale, or syndication needs. Product Information or Experience Management (PIM/PXM) systems address rich product content, improving time-to-market and customer experience, while Master Data Management (MDM) takes a multi-domain view—ensuring consistency across products, vendors, locations, and customers. Both establish a single source of truth in support of scalable AI outcomes. Natively, AI plays a critical role in transforming data management solution platforms by automating data cleanup and enabling reliable data—essential for powering AI, analytics, and omnichannel retail.
Key AI capabilities include:
| AI Capability | Use Case |
| Automated Quality & Data Cleansing: Detects and corrects errors in naming, formatting, units, and terminology. | Equalizing Vendor names for analysis for trusted markdown or rebate compensation. |
| Auto-Classification: Classifies products into correct taxonomies using natural language processing (NLP) and image recognition. | Ensuring accurate taxonomy assignment to meet POS tax rules by product type and state. |
| Image Recognition & Tagging: Extracts features from images and verifies alignment with structured attributes. | Extracts color or fabrication features from product images and verifies alignment with governed attribute values. |
| Attribute Prediction & Completion: Fills in missing fields like material or fit by analyzing similar SKUs. | Rapid matching of child SKUs to Parents Styles and auto-tagging of product material based on Vendor, Country of Origin, and Taxonomy assignment |
| Workflow Automation: Prioritizes and routes records based on completeness, error likelihood, or urgency. | Prioritizes and routes records, requiring actions based on user responsibilities (e.g. Merchandising, Digital Marketing and Legal). |
| Content Generation: Creates rich product descriptions, SEO copy, and translations using generative AI. | Creating product descriptions, SEO copy, and translations for omni-everywhere needs; enabling automatic updates based on customer feedback and sentiment. |
| Deduplication & Golden Record Creation: Matches and merges duplicate records using fuzzy logic and ML. | Synchronizes product catalogues across multiple divisions or because of a merger/acquisition. |
| Syndication Optimization: Tailors content to meet channel-specific requirements (e.g., Amazon, Walmart). | Automated business rules to convert standard data model design to unlimited standards required by business partners |
These AI-powered features significantly reduce manual effort, improve data consistency, and accelerate time to market. Most importantly, they establish the clean, structured foundation needed to scale AI and ML use cases across the retail enterprise.
And leading data management solutions are helping retailers with advanced techniques.
For more information and insights on Columbus Consulting’s data management and AI enablement services: https://www.columbusconsulting.com/service-spotlight-data-analytics/
Case Studies
Pimberly https:/pimberly.com), a leading SaaS-based PIM/DAM solution in the US, has seen firsthand how embedding AI within a PIM platform transforms retail operations by removing the friction from product data management
Customer Case Study Example
The Challenge: Absolute Snow, a leading outdoor sports retailer, had thousands of SKUs across multiple brands and seasonal variations with product data that was complex and constantly changing. Using Pimberly AI, they streamlined their entire product content lifecycle.
The Solution: Using Pimberly the retailer embedded AI within a PIM platform to remove friction from their data management. The solutions addressed:
- Automated product categorization: AI instantly sorts products into the right categories, saving merchandising teams hours of manual work.
- Automatic attribution: Missing details such as size, material, or usage are instantly filled by analyzing existing data, product images, and brand rules.
- Content creation/translation: AI generates brand-consistent copy, SEO-optimized product descriptions, and localized product pages to accelerate global launches.
- Workflow automation: Smart prioritization ensures urgent updates (e.g., compliance or seasonal shifts) are surfaced and approved quickly.
- Channel-specific syndication: AI tailors product content to meet the unique requirements of each marketplace—from Amazon to direct-to-consumer sites—without duplication.
- Data quality monitoring: Companies can use Pimberly AI to flag gaps, inconsistencies, or errors before they impact conversion rates.
The Result: Richer, cleaner product data, faster time-to-market, and a measurable uplift in conversion rates. By integrating AI directly into the data lifecycle, retailers like Absolute Snow can not only keep pace with change but also turn their product information into a competitive advantage.
About Digital Wave Technology
Digital Wave Technology believes clean, connected data is the foundation for unlocking AI’s true potential. Built on a foundation of Master Data Management (MDM) and powered by embedded GenAI and Agentic AI, its AI-native ONE℠ Platform unifies product, product content, merchandising, pricing, marketing, and supply chain solutions on trusted, enterprise-wide data. Digitalwavetechnology.com.
Customer Case Study Example
A national specialty retailer offering a wide assortment of trend-driven products across categories began their AI journey with a focused, high-value project: optimizing product onboarding. Using Digital Wave Technology’s GenAI Product Attribution and Copywriter within the ONE℠ Platform they:
- Enriched and validated attributes
- Generated SEO-optimized, on-brand product copy
- After launch, they adopted Product Review Analysis to infuse customer sentiment back into product descriptions while also sharing insights with merchants and vendors for continual improvement.
The Results included a 7%+ lift in conversions, 5% sales and margin growth, 40% of new attribute recommendations generated automatically to fill data gaps, and a faster, more efficient path to market.
AI Relies on AI, supporting Accurate, Consistent Foundational Data
By integrating AI directly into the data lifecycle, retailers can not only keep pace with change but also turn overwhelming levels of product information into a competitive advantage.
Retailers must invest not just in AI/ML solutions, but in the foundational data that powers them. And AI within Data Management solutions can help. Regardless of your use cases, Data Management with support by AI is not an option, but a must for enabling Retail AI. Retailers who treat data as the backbone of AI—not an afterthought—will win in the next era of retail. Those who do not risk wasted time, efforts, investments, and stalled innovation.
Find out how Columbus Consulting can help you with your data/AI: https://www.columbusconsulting.com/service-spotlight-data-analytics/
ABOUT COLUMBUS CONSULTING
Columbus Consulting delivers solutions that drive true value and have been transforming the retail, grocery and CPG industries for over two decades. We are a retail consulting company of industry experts. Our approach is simple, if you do it, we do it. We are more than consultants; we are experienced practitioners who actually sat in our clients’ seats. We understand the challenges, know what questions to ask and deliver the right solutions. Columbus offers a unique, consumer-centric approach with an end-to-end perspective that bridges functional & organization silos from strategy to execution. Our specialties include: unified commerce, merchandising & category management, planning & inventory management, sourcing & supply chain, data & analytics, accounting, finance & operations, people & organization and information technology. Let us know how we can help you. To learn more, visit COLUMBUSCONSULTING.COM.
