Accelerating Product Innovation for the Global Food Industry
Foodpairing AI leveraged GraphDB to unify 15+ years of ingredient, product, and consumer data into a powerful enterprise knowledge graph that enables faster, smarter, and more sustainable new product development for the global food industry.
30–70%time savings on market insights
70%reduction in team coordination time
25%reduction in R&D costs
The Client
Belgiandata and business intelligence company transforming product development in the consumer goods sector by empowering global food brands to create products that meet consumer demand.
The Challenge
Foodpairing AI had to integrate an enormous variety of in-house and third-party data into a unified system supporting deep analytics, collaboration and predictive modeling
The Solution
Foodpairing AI built a scalable semantic knowledge graph using GraphDB to link 20K+ ingredients, 3M+ products, 10M+ recipes, and billions of social posts
Technical capabilities
Integrated 15+ years of diverse data into a scalable enterprise knowledge graph
Enabled advanced semantic analytics and predictive modeling for new product development
Business outcomes
Accelerated product innovation with 80% faster novelty calculation for ingredient combinations and 30–70% time savings on market insights
70% reduction in coordination time across R&D and business teams and 25% reduction in R&D costs through ingredient targeting
The Challenge
To accurately predict successful product formulations and accelerate product development, Foodpairing AI needed to integrate a wide range of complex data sources:
Chemical, nutritional, and sensory profiles of food ingredients
Health benefits and nutritional information
Ingredient pairings and replacements
Recipes and consumer trends
Cost, regulation, and sustainability indicators (e.g., carbon and water footprints)
This diversity made it difficult to extract actionable insights quickly. The team needed a unified, scalable solution that enabled advanced analytics, supported reuse, and improved collaboration across R&D, marketing, and innovation teams.
The Solution
The team chose to use GraphDB to develop its own W3C compliant, enterprise knowledge graph with:
Several billion time-stamped social media posts, linking ingredients to dish types and other entities of interest (e.g. locations, target audiences)
20K+ unique food and non-food ingredients, categorized under 155 distinct categories organized within a 4-tier taxonomy. All the ingredients are accompanied by their associated aroma compounds, sensory and nutritional information, environmental impact, functional benefits and associated mood states according to literature, and much more
3+M SKU (stock keeping unit) products launched worldwide in the last decade, along with their ingredient lists, respective markets, companies, and brands
10M+ recipes from cuisines all over the world, categorized under a multi-level taxonomy of over 800 dish types
The Impact
The Foodpairing Knowledge Graph has delivered significant time and cost savings across the enterprise:
Lower R&D costs: Identifying underutilized ingredients cuts research spend by ~25%, balancing efficiency with creativity.
Faster innovation: Reduced ingredient novelty calculations by 80%, boosting productivity in recipe development.
Quicker insights: Improved data interoperability cuts market insight retrieval time by 30–70%, enabling faster, better-informed decisions.
Stronger collaboration: Centralized platform reduces coordination time by 70%, streamlining cross-functional alignment.
“By leveraging the unparalleled capabilities of GraphDB, more notably its W3C standards compliance, high performance, scalability, optimized resource allocation, and powerful query capabilities, we can now harness the full potential of semantic knowledge graphs in revolutionizing new product development. The semantic integration of vastly diverse data seamlessly into a unified framework empowers our company to efficiently predict consumer products that meet market demands, driving informed strategic decisions and fostering innovation.”
Stratos Kontopoulos, Knowledge Graph Engineer at Foodpairing AI
Struggling to integrate diverse data sources for successful product development?
Whether you're a CPG company, food manufacturer, R&D organization, or product development team, Graphwise can help you:
