Why a Promising Flavour Sample Can Disappoint in the Finished Product

12 minutes ago

6 min read

Illustrative image: application evaluation should include the intended product and its processing conditions.

The first sample is a starting point, not a finished-product approval

A peach flavour may seem fresh in a simple tasting base, yet appear weak in a protein beverage, unbalanced after processing or less convincing several weeks later. The useful question is not simply, “Is this a good flavour?” It is, “Is this flavour suitable for this formulation, process, pack and intended shelf life?”

An early tasting helps select a direction. It cannot resolve every application question. This guide offers a practical checklist for food and beverage R&D teams to move from a promising sample to a decision supported by relevant trials. It connects the formulation, manufacturing and sensory questions that should be answered before commercial approval.

1. Define what “good” means before changing the dose

Write a short target profile before the next trial. “More peach” is less useful than “ripe peach character, limited green note, balanced with the tea base, without a lingering perfumery impression.” Agree on the reference product, intended serving conditions and the attributes that determine acceptance.

Separate three questions: Does the candidate resemble the target? Is it stable in the application? Do intended consumers like it? An internal screening panel can help with the first question without establishing the third. ISO 6658 distinguishes objective sensory analysis from preference-oriented testing; the method should follow the decision required. [5]

Record both desirable and undesirable attributes. Increasing dose may lift a fruit note and an unwanted note together. A different profile or formulation adjustment may be more useful than a stronger addition.

2. Check the complete application, not only the flavour name

A flavouring operates inside a food matrix: the product’s composition and physical structure. In original research on ripened cheeses, changing composition affected aroma release and sensory attributes. That finding supports application-specific testing; it does not predict the size or direction of an effect in a particular beverage. [1]

Use the following matrix at the briefing stage and again before pilot production. Record actual values where available; mark missing information rather than guessing.

The sample-to-product decision matrix

Checkpoint	Information to record	Question for the next trial
Product base	Ingredients, fat, protein, dry matter, pH, acidity and sweetener system	Does the tasting base represent the intended formula?
Dose and dispersion	Addition by mass, batch size, carrier, dilution and mixing sequence	Is the candidate distributed consistently, with no visible separation?
Process	Time–temperature history, homogenisation, cooling and permitted addition point	Is the trial representative of the intended manufacturing route?
Packaging	Food-contact material, closure, fill volume, headspace and relevant barriers	Has the complete pack been tested with this product?
Storage	Intended conditions, target shelf life and evaluation dates	Is the profile acceptable beyond the first tasting?
Sensory comparison	Reference, descriptors, serving temperature and sample age	Are differences due to the product rather than the tasting setup?

Choose the fruit profile, then test it in the product

T. Hasegawa presents HASEAROMA™ as a flavour series recreating detailed ingredient aromas. Its US portfolio distinguishes white and yellow peach: “peach” alone leaves the sensory target incomplete.[7,8]

T. Hasegawa USA also publicly lists beverage application capabilities including hot-fill, HTST and UHT, alongside analytical checks such as pH, titratable acidity and viscosity.[6]

Consider a hypothetical peach tea whose fruit note weakens after processing. Agree the desired peach character and its balance with tea. Then ask the supplier: Which candidate fits this base and process? What conditions support the recommendation? Should the next comparison investigate profile, dispersion or dose while holding the other factors constant?

Evaluate the selected candidate in the actual base, through the intended process, in the intended pack and at planned storage checkpoints. HASEAROMA identifies a flavour series; the name alone does not establish heat resistance or finished-product shelf life.

3. Reproduce the process before judging heat stability

“Heat stable” is an incomplete specification without the application and processing conditions. Research on milk found different volatile profiles after the heat treatments studied. An apple-juice study also found processing-dependent volatile changes and further changes during refrigerated storage. Neither study establishes the performance of a specific commercial flavouring in your product. [2,4]

Give the supplier the relevant heating and cooling stages, residence times and addition point. Include any holding period before filling. Ask which conditions informed the recommendation, then test under the intended route.

A laboratory hot plate and a production line need not create the same product history. If a bench method is used for screening, document its limits and confirm the selected candidate at a representative pilot scale. Any proposal to move flavour addition to another stage must fit a validated hygienic process; a sensory improvement alone is not a reason to change the production sequence.

4. Include the package in the formulation discussion

The commercial pack is part of the application, not a final visual choice. Original work on model solutions and orange juice examined different aroma absorption behaviour with different packaging polymers. Results depend on the tested material and food system; they do not justify a blanket claim that one material is always preferable. [3]

Specify the intended bottle, carton, liner and closure as accurately as possible. Where a packaging interaction is suspected, compare equivalent product from the same batch in the intended pack and a justified control pack under matched conditions. Record fill volume, headspace, opening dates and storage exposure. A material name alone is rarely a complete packaging specification.

5. Make the sensory comparison fair

Use samples prepared to a written protocol. Weigh additions, record the flavouring lot, identify the base version and confirm that the product is suitable for tasting. Concentrated flavourings should only be handled and diluted according to supplier instructions.

For an internal screening session, use anonymous codes, a balanced or randomised serving order, consistent portions and a defined serving temperature. Keep sample age and time after opening comparable. Ask assessors to record individual observations before group discussion.

Choose a short attribute list relevant to the brief: fruit identity, freshness, sweetness–acidity balance, mouthfeel, unwanted notes and persistence, for example. Define what each scale means. A higher intensity score is not automatically a better product, and a score from a small internal group is not a consumer-preference result.

6. Diagnose the change with a small, controlled comparison

When performance disappoints, change one suspected factor at a time for diagnosis. Start with a reproducibly prepared reference, the unflavoured base and the candidate in the target base. Comparing the unflavoured base helps identify whether an unwanted character already exists before flavour addition.

If the concern appears after manufacturing, compare suitable retained samples at defined stages, with the process history recorded. If it appears during storage, compare unopened packs of the same formulation and batch across planned evaluation dates. Do not confuse a newly prepared reference with a stored product without accounting for their different histories.

One-factor comparisons are useful for locating a likely issue, but they do not reveal every interaction. Once a hypothesis is credible, a designed experiment can test interacting formulation or process variables more efficiently than repeated informal adjustments.

7. Turn observations into an actionable supplier brief

Replace “the sample did not work” with a compact report: formula version; relevant product measurements; flavouring code, lot and dose; process route; packaging; storage age; serving conditions; reference; and specific sensory observations.

For example: “In the current dairy base, fruit identity is acceptable initially, but the cooked note becomes more prominent after the recorded process. We need a less cooked, more recognisable banana profile at the same serving temperature.” This is an illustrative brief, not an experimental result.

Agree on the next change, the person responsible and the evidence needed for the next decision. Before commercial approval, complete the product-specific shelf-life, microbiological, physical and regulatory checks alongside sensory validation. An acceptable first tasting is one milestone in that process.

Share the target profile and the stage where it changes with Galbora to define the next comparison.