This is the Trace Id: e95f61e502421cdbf7b0711dd32590e9
12/16/2025

NIQ scales product coding globally, expands reach, and speeds delivery with Foundry

NIQ’s manual item coding—extracting structured product data from packaging images—was limiting scalability. The company sought an automated solution to accelerate insights and cut operational costs.

NIQ used Microsoft Foundry—Azure OpenAI in Foundry Tools, Azure Document Intelligence in Foundry Tools, and Azure AI Search—to build Capture as a Service (CaaS), automating item coding for faster, more accurate, and scalable product data coding.

With Foundry, NIQ cut item coding time by 90%, expands market reach and speeds client delivery, and launched services in 25 new markets. It now processes tens of thousands of products in hours, instead of weeks, to unlock real-time global insights.

NIQ

NielsenIQ (NYSE: NIQ) is a leading consumer intelligence company, delivering an unparalleled understanding of consumer buying behavior and revealing new pathways to growth for manufacturers, retailers, and various other consumer goods clients. With a holistic retail read and some of the industry’s most comprehensive consumer insights delivered with advanced analytics through state-of-the-art platforms, NIQ delivers the Full View™. The company provides solutions to its clients to help them make smarter, faster decisions on everything from product innovation, pricing, and promotion to marketing and distribution. 

One critical data point that NIQ relies upon is product packaging information. This information is part of NIQ’s proprietary database that holds 220 million products and over 9 billion product attributes. ​Historically, extracting product attribute data from product packaging was labor-intensive, so NIQ saw the opportunity to increase efficiency and scale. Dagan Xavier, VP of Product Content at NIQ, remembers his early days at Label Insight (acquired by NIQ in 2021) vividly. “I was in the supermarket with a Dictaphone, recording what was on packages. Then I’d go home, jump into Excel, and start coding a database,” he recalls. 

In later years, teams manually photographed samples, marked key elements—ingredients, brand names, nutrition facts—and used optical character recognition (OCR) plus machine learning to structure the data. This “item coding” process transformed raw packaging into searchable data, but with each item taking about four minutes to code, scaling thousands of products daily was challenging.

“It was a labor-intensive process that required deep market knowledge and meticulous attention to detail,” says Gabriel Harris, Principal Data Scientist at NIQ. “We saw an opportunity to eliminate the manual steps by using language models to automate the entire pipeline.” This shift to automation is a key part of how NIQ now delivers the Full View™, combining deep domain expertise with cutting-edge technology to unlock faster, smarter insights from every corner of the retail landscape. 

Dagan Xavier, VP of Product Content, NIQ

“Within two quarters, we designed, built, and released an MVP (minimum viable product) with Microsoft Foundry that met our goal of automating our collection of product packaging information.”

Dagan Xavier, VP of Product Content, NIQ

Building with Microsoft Foundry

As part of NIQ’s drive to innovate, the company introduced Capture as a Service (CaaS), a generative AI–powered solution that improved efficiency and unlocked new business potential. “Within two quarters, we designed, built, and released an MVP (minimum viable product) with Microsoft Foundry that met our goal of automating our collection of product packaging information,” Xavier says. “Four months. That kind of speed was unheard of before.”

To bring CaaS to life, NIQ turned to Microsoft Foundry. The team used Azure OpenAI in Foundry Models, Azure Document Intelligence in Foundry Tools, prompt flow, and Azure AI Search to automate item coding. “Grounded in NIQ’s data and decades of expertise and accelerated by Foundry, we created a solution that would have been impossible to deliver so quickly otherwise,” says Harris.

CaaS simulates the manual process with a panel of AI “experts” that extract, verify, and structure product data using a combination of Foundry Tools. Packaging images are processed with Azure Document Intelligence to extract structured data, while Azure OpenAI models interpret and contextualize the content. Azure AI Search validates the extracted information, and prompt flow orchestrates the entire pipeline to help ensure accuracy and consistency.

Automating a process like this requires an obsession with data accuracy and integrity. In item coding, even a small error can have big consequences. “Missing ‘contains peanuts’ or mislabeling fat content could mean the difference between a product being flagged as safe or not for consumers,” Xavier explains. “Foundry gives our data scientists the appropriate oversight capabilities to ensure our CaaS quality standards remain the highest in the industry—critical to our clients making strategic decisions based on our data.”

Dagan Xavier, VP of Product Content, NIQ

“Foundry gives our data scientists the appropriate oversight capabilities to ensure our CaaS quality standards remain the highest in the industry—critical to our clients making strategic decisions based on our data.”

Dagan Xavier, VP of Product Content, NIQ

Global reach, real-time insights

CaaS became the foundation for NIQ Product Insights (NPI), a new service that NIQ launched in 25 markets, with many markets previously inaccessible due to cost or staffing limits. “The solution dramatically improved speed and cost. In one project, we coded 32,000 products in 10 hours—a task that previously required 300 hours. Within two months of launching CaaS, we expanded it to Germany, France, Ireland, and Canada, giving clients unprecedented visibility into consumer preferences and product attributes,” says Andrew Criezis, Senior Vice President and General Manager at NIQ. 

“Even if we wanted to hire a team, it would take months—and they’d have to be native speakers,” Harris explains. “Now we can go to any market where a language model can understand the language. We’re moving fast without a physical footprint.”

Looking ahead 

From a business perspective, the results are clear. “CaaS is enabling cost and time efficiency in our ability to code products,” says Criezis. “We’ve seen over 90% savings in time. The speed at which we can launch product insights across the globe has jumped up drastically.”

NIQ’s journey with Foundry is a blueprint for transformation. By automating a critical process and enabling global scale, the company has turned a technical challenge into a strategic advantage.

Looking ahead, NIQ is exploring agentic architecture with Microsoft Agent Framework to scale item coding and insights, customized to every client based on its data sets and item coding. Clearly, NIQ is well on its way. 

Discover more about NIQ on Facebook, Instagram, LinkedIn, X/Twitter, and YouTube

Take the next step

Fuel innovation with Microsoft

Talk to an expert about custom solutions

Let us help you create customized solutions and achieve your unique business goals.

Drive results with proven solutions

Achieve more with the products and solutions that helped our customers reach their goals.

Follow Microsoft