Know if your product will land, before it does
HaySense is Haystack's predictive analytics service that estimates consumer liking for any product in a category. Feed in a sensory profile, get back a predicted score. Then find out exactly what to change to make it better.
About the service
What is HaySense?
HaySense is a decision-support service that predicts consumer liking based on a product's sensory characteristics. By learning from historical sensory and consumer data - yours, ours, or both - it estimates how your target consumer will respond to any formulation, real or simulated.
Predict before you test
Enter a product's sensory profile and get a predicted liking score for the consumer segment you choose - by country, gender, age, or any combination. No consumer panel required.
Three things, one service
HaySense does three things: it predicts liking for existing or simulated products, maps the full sensory landscape of consumer preference, and identifies the reformulation path that moves a product fastest toward higher liking.
Why predictive sensory analytics matters
Consumer testing takes time. Reformulation is iterative. And every prototype cycle that goes to panel costs budget and slows momentum. HaySense gives your R&D and insights teams a smarter starting point. So you test fewer prototypes, make faster decisions, and get to your best product sooner.
Map the full sensory landscape of your category. See where high-liking zones are, where your products sit relative to them, and which directions move you closer to peak consumer preference.
Understand what drives liking in your category - and what holds products back. HaySense surfaces the attributes that matter most, so reformulation decisions are grounded in data, not intuition.
Simulate product variants without running a single panel. Tweak one or two sensory attributes, re-score instantly, and compare predicted liking across your full portfolio - existing SKUs, competitor benchmarks, and new prototypes side by side.
Start predicting what your consumers will love
HaySense turns historical data into a living model of consumer preference - so every formulation decision starts from the best possible evidence.
Ideal Partner
Why partner with Haystack Consulting for HaySense
HaySense is built on decades of sensory and consumer research expertise. We know that predicting liking is only useful if the model behind it is trustworthy - so every HaySense engagement starts with a rigorous data assessment and a validated model before any prediction is made.
We work with your existing data where it exists, and collect what is missing where it does not. Either way, the model is built for your category, your products, and your consumer segments - not a generic benchmark.
Unique process
Predict. Explore. Optimise.
HaySense works in three steps - each designed to move you closer to a product your consumers will love.
From sensory profile to liking score
Feed in a product's sensory characteristics - how it looks, smells, tastes, and feels. HaySense returns a predicted liking score for the consumer segment you select. Combine segments, compare products, or create simulated variants in seconds. No new panel. No delay.
From landscape to reformulation path
The sensory landscape maps every product in your category against consumer preference - showing hotspots, attribute directions, and the distance between where your product sits today and where liking peaks. From there, HaySense identifies the highest-impact sensory tweaks, in priority order, with an estimated liking lift for each step.
Brand aligned
A partner who keeps your best product in sight
We see predictive analytics as a tool for sharper decisions - not a replacement for sensory expertise. HaySense accelerates the process. Our researchers give it direction.
Hybrid precision
HaySense generates the scores. Our sensory consultants interpret them, challenge them, and translate them into decisions your team can act on.
Scientific foundation
Every model is trained on validated sensory and consumer research - yours, ours, or a combination. No generic assumptions.
Tailored to your category
HaySense is configured for your specific category, consumer profile, and portfolio - not a one-size-fits-all benchmark.
Ethically designed
We explain what the model predicts, how it was built, and where its limits are. You always know what you are working with.
How we build and run your HaySense model
Three steps from data to decision - fast, rigorous, and built around your category.
Align
We start with your business question. What decision does HaySense need to support - innovation, reformulation, competitive benchmarking, or portfolio optimisation? This shapes everything that follows.
Predict
We run HaySense on your product portfolio - scoring existing SKUs, simulating variants, and mapping the full sensory landscape of consumer preference in your category.
Optimise
We deliver a prioritised reformulation path: the sensory attributes to change, in order, with an estimated liking lift for each step. Your team moves forward with a clear, data-backed direction.
Build
We assess your existing sensory and consumer data, identify any gaps, and collect what is needed. The model is trained on your category and validated before any predictions are made.
Explore
We present the sensory landscape: hotspots, attribute directions, and where every product sits relative to peak-liking zones. You see the full picture before any decision is made.
Scientific foundation
We assess your existing sensory and consumer data, identify any gaps, and collect what is needed. The model is trained on your category and validated before any predictions are made.
Scientific foundation
Every model is trained on validated sensory and consumer research - yours, ours, or a combination. No generic assumptions.
Predict
We run HaySense on your product portfolio - scoring existing SKUs, simulating variants, and mapping the full sensory landscape of consumer preference in your category.
Explore
We present the sensory landscape: hotspots, attribute directions, and where every product sits relative to peak-liking zones. You see the full picture before any decision is made.
Optimise
We deliver a prioritised reformulation path: the sensory attributes to change, in order, with an estimated liking lift for each step. Your team moves forward with a clear, data-backed direction.
Don’t just take our word for it
"Great team to work with. Always innovating research agency for sensorial research!"
"Haystack brainstorms with us as a partner to create challenging, innovative, and custom-made projects."
"We have great professional experience with Haystack! They are reliable, dynamic, pleasant and highly flexible. They are specialised in sensory research among others."
"Haystack has a critical approach embedded in their company culture and methodologies. That is why we prefer to work with Haystack."
Ready to shape your HaySense story?
If you are ready to predict consumer liking, map your sensory landscape, and find the fastest path to your best product - our team would love to connect.
Frequently asked questions
HaySense is a new kind of service, and it raises practical questions. Here are the ones we hear most from R&D and insights teams exploring what predictive analytics can do for their category.
Start your HaySense journey today
With HaySense, you can predict consumer liking, map your sensory landscape, and find the fastest reformulation path to the product your consumers will love most.
If you are ready to bring predictive analytics into your innovation process, our team is here to help.
