Sensory Evaluation Methods for Dairy Products

How do dairy companies know whether a new cheese recipe will delight consumers, or whether a processing change has altered the flavor of their milk? The answer isn’t just chemical testing – it’s sensory evaluation. Sensory evaluation is the scientific discipline that uses human senses – taste, smell, sight, touch, and hearing – to measure and analyze the characteristics of food products. In the dairy industry, where subtle flavor shifts, texture changes, and consumer preferences directly affect product success, sensory analysis is an indispensable quality assurance tool. Several distinct methods exist, each designed to answer a different question – from whether a product has changed, to whether consumers will buy it.

Why sensory evaluation matters in dairy 🔗

Chemical analysis tells you what is in a product. Sensory evaluation tells you how a product is experienced. These two perspectives don’t always agree. A milk sample may pass every microbiological test yet still carry an off-flavor detectable only by a trained panel. Descriptive sensory analysis, developed in the 1950s, has gradually replaced older defect-based judging methods in published research because of its greater versatility, specificity, and statistical reliability. Today, dairy manufacturers use a range of structured sensory methods to cover the full spectrum – from detecting imperceptible formulation changes to predicting commercial success.

International standards published by ISO on sensory analysis include specific guidance for milk and dairy products, covering the recruitment, training, and monitoring of assessors, as well as requirements for the test room, sampling, and evaluation of appearance, odor, flavor, and texture. Following these standards ensures that results are reproducible and defensible across the industry.

Difference testing: detecting what changed 🔗

The first question any quality assurance team must answer is: has something changed? Difference tests – also called discriminative tests – are designed precisely for this. Their sole purpose is to determine whether two products are perceived as different, not to describe the nature or degree of that difference. If a further investigation is needed, difference testing provides the justification for proceeding to more detailed methods.

Paired comparison test 🔗

In a paired comparison test, panelists receive two coded samples and are asked which has a higher intensity of a specified attribute – sweetness, saltiness, creaminess, and so on. Each panellist is given two differently coded samples simultaneously, and the purpose is to identify which sample rates higher on the specified sensory attribute. This test is particularly effective when a known compositional difference exists between products – such as comparing two chocolate milks with different sweetener concentrations – because panelists can be directed to focus on that specific attribute. Between 10 and 50 panelists are typically recommended.

Triangle test 🔗

The triangle test is one of the most widely used difference tests in sensory evaluation. Panelists receive three coded samples – two identical and one different – and must identify which sample differs from the other two. Its key statistical advantage is that random guessing produces a correct answer only one-third of the time (33.3%), compared to 50% in a two-sample test, which means fewer assessors are needed to reach statistical significance. This makes it highly efficient for detecting subtle formulation changes, such as a 10% reduction in butter salt content. In practice, the triangle test has been used to assess differences between milk processed using high pressure versus conventional heat treatment, where nearly 90% of panelists correctly identified the odd sample – confirming a perceptible sensory difference.

One limitation worth noting is that the triangle test can create more sensory fatigue than simpler tests, since panelists must evaluate three samples. No more than six samples are generally recommended per session.

Scoring: putting numbers to sensory attributes 🔗

Scoring is the most frequently used method of sensory testing for food quality, especially in the dairy industry. It assigns numerical values to specific sensory attributes – flavor, appearance, texture, aroma – transforming subjective impressions into measurable data that can be tracked, compared, and statistically analyzed over time.

Scorecards based on 100 points are traditionally used for judging and grading dairy products, with more recent suggestions favoring a 25-point format. The American Dairy Science Association (ADSA) scorecard, which has been a cornerstone of the industry since the early 1900s, grades fluid milk on a 0-to-10 scale – placing products into categories from excellent (10) to unacceptable (0). Attributes scored include appearance, flavor, and texture, based on the presence or absence of predetermined defects.

Scoring is especially valuable for quality control and shelf-life studies. For example, if a dairy plant establishes that its premium milk must score between 7 and 8 for freshness, any batch falling below that threshold can be flagged for investigation before reaching consumers. For cheese manufacturers, trained panelists can score attributes like sharpness, nuttiness, and overall flavor intensity at different aging stages to identify the optimal aging window.

Ranking: ordering products by intensity 🔗

Ranking methods ask panelists to arrange multiple products in order based on the intensity of a specific sensory attribute. Unlike scoring, which produces absolute numerical values for each sample, ranking only reveals relative relationships – which product has more of a given attribute than another, but not by how much. In a ranking test, three or more samples are ordered, with one sample positioned as having more of a defined attribute than the others.

A ranking test can be used to assess noticeable differences between several products depending on the intensity of the difference. It is practical for competitive benchmarking – lining up your product against two or three competitors to determine where it sits on a specific quality dimension, such as saltiness in processed cheese or creaminess in yogurt. The limitation is that ranking doesn’t quantify the gaps between products, only their order.

Hedonic scaling: measuring how much consumers like a product 🔗

Technical quality doesn’t automatically equal consumer acceptance. A product can score perfectly on a flavor scorecard yet still disappoint consumers. Hedonic scaling bridges this gap – it measures how much people like or dislike a product, not just what its sensory attributes are.

For measuring product liking and preference, the 9-point hedonic scale is probably the most useful sensory method. Developed in the 1940s with contributions from the US Army Corps of Engineers, the scale runs from “dislike extremely” (1) to “like extremely” (9), with “neither like nor dislike” at the midpoint (5). Its endurance comes from practical strengths: it is easily understood by untrained consumers, results are remarkably stable, and product differences in liking are reproducible across different groups of subjects.

In dairy applications, a 9-point hedonic scale has been used to evaluate consumer acceptance of UHT milk varieties, finding that skimmed milk samples consistently received the lowest overall liking scores, while whole milk varieties scored higher – confirming that fat content directly influences sensory appeal. The overall liking question should always be asked first to avoid biasing responses with other attribute questions.

Variants of the standard scale – including 7-point and smiley-face scales – are used for children or non-English-speaking populations. For most standard research situations, the 9-point version remains the scale of choice.

Descriptive analysis: building a complete sensory profile 🔗

When you need to understand not just whether a product is different or liked, but exactly what the sensory differences are, descriptive analysis is the method to use. Descriptive analysis is a sensory methodology that provides quantitative descriptions of products based on the perceptions of a group of qualified subjects – a total sensory description taking into account all sensations perceived, including visual, olfactory, and textural.

A trained sensory panel should produce results analogous to instrumental data – precise, reproducible, and capable of being statistically analyzed. Panels typically consist of 6 to 15 well-trained assessors who agree on a defined vocabulary, or lexicon, specific to the product category. For dairy, this might include terms like “cooked milk flavor,” “whey notes,” “chalky mouthfeel,” or “barn-like aroma.” Assessors are trained to score each attribute consistently, regardless of personal preference.

Quantitative descriptive analysis (QDA) 🔗

Quantitative descriptive analysis can produce a full qualitative and quantitative sensory description. Assessors (8-15) agree on a list of attributes and then rate them individually on a line scale with anchored endpoints. Data are analyzed using ANOVA and presented as spider (radar) plots, making it easy to visualize the full sensory fingerprint of a product at a glance. This approach is especially useful for cheese flavor studies, comparing whey protein concentrates, or evaluating how fat reduction affects texture.

Flavor and texture profile methods 🔗

The flavor profile test evaluates flavor aspects including character notes, intensity, order of appearance, aftertaste, and amplitude (overall impact), using a scale ranging from 5 to 14 points. The texture profile test assesses mechanical and structural properties such as hardness, cohesiveness, adhesiveness, viscosity, and chewiness – attributes that are critical in products like processed cheese, yogurt, and butter.

Descriptive analysis has a key advantage over traditional dairy judging methods. Descriptive analysis was shown to be more sensitive to differences than traditional judging panels given the same amount of training – and it is more likely to correlate with actual consumer liking scores.

Preference and acceptance testing: the consumer verdict 🔗

Preference and acceptance testing moves beyond technical sensory measurement to directly capture consumer behavior and purchasing intent. These tests address questions like: which product do consumers prefer, and will they actually buy it?

Preference testing typically involves side-by-side comparisons where consumers choose between two or more options. A dairy company might present two ice cream formulations and ask which the consumer prefers. The results deliver clear, actionable guidance – one product wins, and the margin of preference can be quantified.

Acceptance testing goes further by measuring degree of liking and purchase intent. Hedonic tests in an acceptance context are mainly used to compare a product with competitors and to optimize a product to be liked by the largest number of consumers. These tests typically use the same 9-point hedonic scale, but the analytical focus is on whether products clear a commercial acceptability threshold.

Research using hedonic scales for yogurt evaluation has shown that while scales can discriminate effectively between different formulations – Greek, drinkable, soy, coconut – they may not always capture cultural differences in consumer response. This finding reinforces a broader principle: for high-stakes consumer research, combining multiple sensory methods gives a more complete picture than relying on any single test.

Choosing the right method for the right purpose 🔗

No single sensory method covers every situation. The basic goal when selecting a sensory evaluation method is to match the right test to the right question. Quality control applications typically rely on difference testing to catch unwanted batch-to-batch changes. Product development projects use descriptive analysis to understand how a formulation change affects specific attributes, followed by hedonic testing to confirm consumer acceptance. Competitive benchmarking calls for ranking, while shelf-life monitoring is well served by regular scoring against established standards.

Many dairy companies use methods in sequence. A manufacturer introducing a new processing technique might first apply a triangle test to confirm whether the change is detectable, then use descriptive analysis to identify exactly what changed, and finally conduct acceptance testing to determine whether consumers prefer the reformulated product. This layered approach ensures that both technical performance and consumer satisfaction are validated before a product reaches the market.

What do you think? If you were launching a new reduced-fat dairy product, which sensory method would you prioritize first – difference testing to confirm the change is detectable, or hedonic scaling to measure whether consumers still enjoy it? And do you think traditional dairy scorecard scoring still has a place alongside modern descriptive analysis in today’s quality assurance programs?

References

1. https://www.sciencedirect.com/article/pii/S0022030207719604
2. https://www.sciencedirect.com/article/pii/S0022030217310536
3. https://www.sciencedirect.com/topics/agricultural-and-biological-sciences/sensory-evaluation
4. https://foodsafety.institute/food-fundamentals-chemistry/difference-tests-sensory-variations-food-products/
5. https://www.foodresearchlab.com/blog/rte-rtc/sensory-evaluation-methods-for-food-and-beverage-products/
6. https://www.mdpi.com/2304-8158/11/9/1233
7. https://www.sciencedirect.com/article/pii/S0022030281828470
8. http://ecoursesonline.iasri.res.in/mod/page/view.php?id=6055
9. https://www.agrimoon.com/wp-content/uploads/Judging-of-Dairy-Products.pdf
10. https://www.foodandnutritionjournal.org/volume8number3/implication-of-sensory-evaluation-and-quality-assessment-in-food-product-development-a-review/
11. https://pmc.ncbi.nlm.nih.gov/articles/PMC7922510/
12. https://www.sciencedirect.com/topics/agricultural-and-biological-sciences/hedonic-scales
13. https://www.mdpi.com/2304-8158/11/9/1350
14. http://ecoursesonline.iasri.res.in/mod/page/view.php?id=147827
15. https://www.foodresearchlab.com/blog/rte-rtc/sensory-evaluation-of-food/
16. https://www.journalofdairyscience.org/article/S0022-0302(17)31053-6/fulltext
17. https://pmc.ncbi.nlm.nih.gov/articles/PMC8227163/