Most solutions can be run with minimal input. In general the platform requires the following input:

• Consumer target
• Product category

Foodpairing has a large data-set readily available in a range of countries. The Foodpairing platform leverages both open- and proprietary data. The ingredient flavor database and the consumer flavor preferences are the main proprietary data sets.

Line extensions do more than just grow a brand’s product portfolio and give consumers more choice. Line extensions have been shown to lead to higher brand equity. Diversifying product lines allows brands to connect and engage with a broader spectrum of consumer preferences, leading to greater sales of core products and wider brand recognition. Research shows that flavor-driven innovation is an essential part of effective line extension strategies.

The potential rewards are much larger for more established brands; however, the risk of launching the wrong product could also have significant consequences. New product launches that fail to meet market demands could cost millions in development costs, sales cannibalization, and broken trust. 

80% of our taste experience is due to aroma. This is why we put an emphasis on aroma. To increase accuracy, we have also added taste data to our ingredient database. We have pairing algorithms that take aroma, taste into account. 

The quantification of aroma and taste (actually ‘flavor’ is the correct description) is executed through analysis like:

• GC-MS (aroma)
• HPLC (e.g. sugar)
• Spectrophotometry (e.g. alpha-acids)

Our models aim for an accuracy of up to 70% when it comes to predicting what would happen in the market once the product is launched. This target is not something we always reach within a first project. Most of the time multiple iterations are needed.

Reliability comes within the model simulation: since our data and models are precise and robust, the concept generated is likely to be successful.

Yes, our Core Product Insights and Core Product Upgrades solutions allow you to test or simulate your existing product in our virtual world to predict the success rate.

Flavor houses produce and sell the flavor compounds and delivery technology used in food and beverage development. They provide information about flavors and fragrances based on a list of trends and other factors. This list is not client-specific or customized based on your target audience. 

Foodpairing, on the other hand, uses prescriptive analytics to give you insights on not only what flavors to launch, but also how likely these flavors are to succeed in the market.  Our models are customized to both your product and target group to give you the most appropriate insights. Our CFI technology creates a virtual world and combines food data with consumer data to predict the most successful formulations. Each concept is supported by: the trial prediction (how interested your target audience will be in buying the product), the average liking (how much your target group likes the product), and the reasons for the product’s success (supporting statements on why this concept will be successful based on insights retrieved from the customer-specific model). 

The success of this technology depends on two main categories of data – digitized food data and digitized consumer data. 

Our algorithmic models combine these two data categories to prescribe the best flavor concept that is most appropriate for your target market, as well as the likelihood of its success. We also work with flavor houses to provide a finalized recipe or formulation for the ideal product type. 

We help you speed up the product development process as our data-backed insights ensure you only need to test a few formulations, thereby reducing the choice dilemma.

No, this is not  a service we currently provide.

No, this is not something the current technology can help predict.

Flavor can be described as the combination of taste, aroma, texture that create the overall experience in food and beverage. At Foodpairing, we specialize on flavor and more specific on aroma and taste attributes.  We do have  understanding of texture and trigeminal sensations and can incorporate it into our modeling, but we require an expert sensory data set from our clients as input for the model training.

Foodpairing helps provide a good basis for product development. Our process involves suggesting the flavor concepts that can be launched and creating a technical flavor brief that includes the exact flavor formulation needed for the final product.

Additionally, we also provide recipes as a reference for R&D teams. However, since we work with a variety of food products with different production processes, we do not provide instructions on how to translate this recipe and formulation to the factory floor.

For example, we do not provide input on what stabilizers, preservatives or emulsifiers are required in the production process. This is something the R&D teams would have to work on internally, depending on their company’s own production processes.