Solutions
GLP-1 flavor restoration
GLP-1 agonists modify taste perception systemically. Food companies reformulating for this market need constructs scored against post-drug receptor states — not against the general population baseline.
40M+
GLP-1 prescriptions (2024)
CAT-27
dedicated catalog
IFI-S + IFI-U
primary restoration axes
Free
catalog search
01
The problem
GLP-1 agonists — Ozempic, Wegovy, Mounjaro, Zepbound — are now taken by tens of millions of people and that number is accelerating. What they do to taste perception is consistent and documented: sweetness signals are dampened, protein-rich foods become aversive, umami and kokumi richness flatten out, and satiety signals arrive much faster.
For food companies, this is a reformulation problem without a molecular vocabulary. The standard approach — adjust sugar, swap fat, increase seasoning — doesn’t address the receptor-level reason the food stopped registering. Teams are working from tasting panels alone, with no way to specify what the receptor state actually needs.
Every flavor multinational has announced a GLP-1 taste modulation program. The question is what tools exist to pursue it quantitatively.
02
What the platform does about it
Scores against the actual receptor state
GLP-1 agonists act on gut-brain signaling pathways that modulate TRPV1 sensitivity, glucose-linked taste reception, and the gut-derived signals that govern palatability. IFI scoring accounts for where those signals are suppressed — a sweet receptor that is functioning at 60% of baseline needs a different construct than one at 100%.
IFI-S: sweetness restoration
Sweetness is typically the first axis affected. CAT-27 filters for constructs where the sweetness signal is concentrated at lower thresholds — engineered to register at the stimulus level that GLP-1-modified receptors can still detect. Not more sugar; a different signal architecture.
IFI-U: savory depth that lands
Umami and kokumi are the axes that make a food feel satisfying rather than merely present. When GLP-1 blunts that channel, protein-rich foods become flat — which is why users report protein aversion specifically. Constructs with elevated IFI-U and IFI-K scores are designed to restore that richness signal.
IFI-K: mouthfeel and continuity
Kokumi is the least-understood axis and the hardest to restore by conventional reformulation. It is a calcium-sensing receptor response that creates the perception of body, fullness, and long duration. CAT-27 includes constructs specifically modeled for GLP-1 kokumi restoration — the axis most implicated in why foods stop feeling satisfying.
Quantitative targets, not impressions
The IFI gives you a number for each axis. When a tasting panel says a reformulation is 'almost there but missing something,' the IFI score tells you which axis is still short, by how much, and which constructs close that gap. Sensory language maps to receptor math.
Start with what already exists
CAT-27 is a curated subset of the full 1,000-construct catalog filtered for GLP-1 restoration profiles. Many reformulation questions are answerable with an existing construct. The catalog search is free and is usually the fastest path to a specification.
03
A representative brief
- Product
- Protein beverage, RTD
- Problem
- Users on GLP-1 agonists find it flat and unappealing
- Axes affected
- IFI-U (umami/savory), IFI-K (kokumi richness)
- Catalog query
- CAT-27, IFI-U ≥ 0.70, IFI-K ≥ 0.55
- Existing matches
- 3 constructs in range
- Gap
- IFI-K 0.08 below threshold on best match
- Resolution
- Design revision or blending spec using IFI-K optimized construct
Three catalog matches is a good result — it means the direction is right and the work reduces to a small delta, not a full redesign. The IFI-K gap is the kind of thing that shows up in sensory panels as “something missing” with no clear candidate. Here it has a name, a number, and a construct that addresses it.
Deliverable is a complete specification: IFI score with per-axis breakdown, construct records, predicted receptor response, and a recommended verification protocol. A formulation team can take that forward without interpreting anything.
04
Getting started
CAT-27 is the fastest entry point. It is filterable by IFI axis, available as a PDF download, and designed to be browsable by a formulation team without a platform account. If the brief is specific enough to need a custom analysis — target receptor state, specific product category, confirmed IFI delta — the custom quote form takes that in and returns a brief within five business days.
Teams running ongoing GLP-1 reformulation programs typically move to an enterprise workspace with API access and direct construct licensing, so the pipeline can query IFI scores programmatically as formulations evolve. That tier is available on request.
Ready to run it
Try this in the Splice Platform.
CPU engines are free to run. GPU engines run on prepaid credits.
Related