Every product you've tested leaves behind evidence about what consumers value, what sensory characteristics drive response, and where products succeed or fail.
Spark Digital Twins™ connect those historical relationships to build predictive models of product performance — creating a virtual environment in which new product possibilities can be explored before every one has to be physically developed and tested.
PRODUCT EVIDENCE
SENSORY × CONSUMER RELATIONSHIPS
PREDICTIVE PRODUCT MODELS
Historical products, prototypes, formulations and product characteristics provide the foundation.
Models learn how sensory characteristics and product differences relate to consumer response.
Those relationships are used to estimate how new product directions may perform — and which are worth taking forward.
Test more possibilities virtually. Make fewer and better prototypes physically.
PREDICT PRODUCT PERFORMANCE
EXPLORE FORMULATION SPACE
IDENTIFY DRIVERS OF LIKING
Estimate how new product or formulation directions are likely to perform before physical testing.
Simulate alternative sensory and product profiles to uncover promising directions.
Understand which sensory characteristics drive consumer acceptance and emotional preference.
PRIORITISE PROTOTYPES
OPTIMISE PRODUCTS
Explore how changing key characteristics could enhance overall product satisfaction and sensory resonance.
Focus physical development on high-potential directions to reduce unnecessary iteration and concentrate testing where it matters most.
The Twin learns from products you've already tested, models the relationships between product characteristics and consumer response, then uses those relationships to explore products you haven't made yet.
Test more possibilities before you make more products.
TRADITIONAL DEVELOPMENT
WITH SPARK DIGITAL TWINS™
Develop multiple physical prototypes
Explore many directions virtually
Test each with consumers
Predict likely performance
Analyse what worked and why
Identify likely drivers before prototyping
Reformulate and test again
Prioritise the strongest formulations
Repeat development cycle
Take fewer, better prototypes forward
USA · Singapore · Malaysia · India