Food scientist evaluating coded strawberry kefir samples beside a blank sensory worksheet in a bright dairy development laboratory.

Flavoured kefir prototypes: a fair sensory comparison

21 minutes ago

4 min read

Illustrative kefir evaluation scene; the image does not represent published trial results.

A better question than “Which one tastes best?”

Two strawberry kefir prototypes arrive at the development table. One seems brighter; the other seems smoother. Before selecting either, ask whether they were compared at the same age, temperature and handling conditions. Otherwise, a difference in sample history may be mistaken for a difference in flavour performance.

The useful R&D question is more precise: which prototype comes closest to the agreed product target, and what needs changing next? This guide proposes a practical internal screening worksheet for flavoured, milk-based kefir. It focuses on how to compare prototypes, rather than choosing a fruit concept or designing a new recipe.

1. Fix the question and document the sample

Write a short target before tasting: fruit character, intended sweetness–sourness balance, body and finish. “Fresh strawberry with a recognisable fermented dairy character” is more useful than “premium and delicious”. Agree what those words mean with the assessors.

Use the same kefir base and production lot where the purpose is to compare flavour systems. Keep fruit preparation, sugar or sweetener system, fermentation endpoint and mixing procedure consistent unless one is the variable under study. If several factors change, record the exercise as a comparison of complete formulations; it cannot isolate the flavour ingredient's contribution.

For every sample, record its formulation version, production time, flavour-addition stage, packing time, storage history and elapsed time since preparation. Include the plain base and an agreed reference when useful. A market product can be a benchmark, but differences in composition must remain visible in the interpretation.

2. Make the tasting blind and repeatable

Give samples unrelated three-digit codes and distribute them without supplier names or concept descriptions. Balance presentation order across assessors rather than serving every person the same sequence. The organiser should retain the code key until individual scoring is complete.

Set a serving temperature relevant to the intended chilled use, verify it, and apply the same target to all samples. Use equal portions, matching cups and a consistent pour-to-taste interval. Standardise gentle mixing before pouring where required; uncontrolled shaking can change the appearance and foaming being judged.

Taste in a quiet, odour-free setting. Provide water, a neutral palate cleanser suitable for the protocol, and a consistent break between samples. Collect independent observations before discussion. Keep sessions manageable, and repeat promising comparisons on another occasion and, later, another production batch. Agreement after an open group discussion is not the same as repeatable individual perception.

3. Separate fruit, sourness, balance and aftertaste

Sensory profiling characterises a product through defined attributes; ISO 13299 describes its general scope.[4] For this practical screen, use separate fields instead of one overall “flavour score”.

Field	Suggested record	Why keep it separate?
Fruit identity	Free descriptor, plus 0–10 match to the agreed target	Stronger fruit is not necessarily the right fruit character.
Fruit intensity	0 = absent; 10 = very strong	Separates strength from identity.
Perceived sourness	0 = absent; 10 = very strong	Describes sensation, not the pH reading.
Perceived sweetness	0 = absent; 10 = very strong	Makes the balance discussion explicit.
Fermented note	Named descriptor and intensity	Keeps characteristic kefir notes distinct from off-notes.
Sweetness–sourness balance	Too sour / on target / too sweet	Indicates direction relative to the brief.
Body and texture	Thin–thick scale; separate graininess/separation notes	Avoids hiding physical changes inside flavour ratings.
Aftertaste	Character, intensity and observed persistence	Distinguishes pleasant fruit persistence from bitterness or an unwanted lingering note.

Agree the scale anchors before testing. Record the aftertaste at the same agreed point after swallowing or expectoration. Do not add these different scales into a single total: intensity, target fit and persistence answer different questions.

4. Keep instrumental measurements beside sensory observations

Record pH and titratable acidity using consistent, documented laboratory methods, including measurement temperature and the stated basis for acidity. Add viscosity or another physical measurement when relevant, using comparable conditions.

Do not convert pH directly into a sourness score. Research on acid solutions found that pH alone did not explain perceived sourness; this is not a kefir-specific prediction.[3] Keep chemistry and perception as complementary records. If two prototypes seem equally sour at different pH values, retain both observations and investigate rather than overwriting one to fit the other.

5. Compare the storage journey

A prototype should not advance solely because it performs well immediately after preparation. Revisit it at matched storage ages chosen for the intended product and distribution conditions.

In one berry-flavoured kefir study, pH fell during refrigerated storage.[1] Another kefir study found no significant average pH decrease over its observation period.[2] Post-acidification is therefore something to monitor in the actual formulation, not assume to have the same magnitude in every kefir.

At each planned point, repeat the core worksheet and record actual storage and serving conditions. Watch for drift in fruit identity, balance, fermented notes, texture and aftertaste. Keep enough separately packed units for the plan; repeatedly opening one container introduces another variable. Unexpected changes should trigger investigation of the base, process and storage history as well as the flavour system.

6. Use the profile to choose the next experiment

Compare individual scores and their spread, not only panel averages. If assessors disagree, inspect their descriptors and scale understanding before declaring a winner. A hidden duplicate can help reveal consistency, but one small session does not establish statistical significance.

Write a decision with a reason and a next action: retain, revise one defined variable, repeat, or stop. For example, “fruit identity matches the brief, but the late-storage balance shifts away from target” is actionable. Simply asking for more aroma may miss the actual problem.

A technical internal screen does not establish consumer acceptance; ISO 11136 addresses consumer liking as a separate test purpose.[5] Samples must first be cleared under the company's food-safety procedures, and commercial selection still needs appropriate consumer, process and shelf-life validation. This worksheet is a development aid, not a release specification.

For a project discussion with Galbora, share the product brief, base composition, process, intended storage conditions and completed comparison notes. A clear record turns “we prefer sample B” into a focused next trial.