Troubleshooting Milk Flavor Problems for Small Processors

Have you noticed that your milk is not tasting as fresh and clean as usual during routine evaluations after production or during shelf life testing? Have you received consumer complaints that something is wrong with your milk? How do you go about figuring out and correcting the problem? Sensory evaluation! The first step in troubleshooting is to describe and define the undesirable flavor or texture accurately.

The purpose of this article is to help small and on-farm processors get started on the path to troubleshooting the most common flavor defects found in milk. The path begins by knowing the flavor of your best-quality milk, so you have a baseline for comparison. A simple, in-house sensory evaluation program can be used to routinely monitor quality and identify any issues before your customers do. Familiarity with the typical off-flavors and their causes provides a starting point for solving problems and preventing recurrences in the future – be it on the farm, in the raw milk supply, at the processing plant, in storage, transport, or elsewhere. 

Troubleshooting milk flavor problems is challenging. Off-flavors in milk can originate at many different places along the production and processing chain and may not develop until the end of shelf life. The same off-flavor can have multiple causes and sources. The perceived flavor may change during the course of shelf-life, although the initial cause is the same, as is observed with light-induced oxidation off-flavors. Light oxidizes both fats and proteins and initiates a cascade of reactions that create flavor compounds that are characterized as cardboardy, burnt hair or feathers, or paint-like, depending on how far the oxidation reactions have progressed.

Troubleshooting milk flavor problems is complex. This article will help small and on-farm processors set up a sensory program to monitor milk quality routinely and provide a general description of the characteristics, causes, and sources of the most common defects observed in fluid milk processing. A detailed discussion of the off-flavors, mechanisms, and suggestions for corrections is beyond the scope of this publication and has already been published by others. Readers are encouraged to seek these references for more information:

• Dairy Practices Council Guideline 38, Identifying and Preventing Off-Flavors in Milk and Milk Products (2023) 
  • Detailed description of off-flavors in milk, causes and mechanisms, prevention on the farm and processing plant
  • Basic information for setting up a sensory evaluation program
• The Sensory Evaluation of Dairy Products (2023)
  • Detailed description of off-flavors in milk and other dairy products, general background on causes and mechanism
  • Detailed information for setting up a sensory evaluation program

Setting Up a Sensory Evaluation Program

The concept of developing, implementing, and routinely conducting sensory evaluation on your milk and other products may sound overwhelming, but it doesn't have to be. A simple, effective sensory evaluation program can be implemented and used routinely by a one or two-person operation with a little planning and a commitment to producing high-quality products. Obviously, the more personnel available to develop, implement, and participate in the program the better, but the size of your operation should not be a barrier to using this important quality management tool. 

How Much Time Will It Take, and Is It Worth It?

Is it worth it?
Absolutely! What better way to guarantee your product quality than to routinely taste your products and know what your customers are getting. 

How much time will it take?
At the simplest, it should only take a few hours to set up your written tasting procedures and data recording forms and to organize supplies. Routine evaluation can be as quick as tempering samples by taking them out of the cooler for 30 minutes or so while you do other things and then taking 10 minutes to focus your head, taste a few samples, think, and record your evaluations. The more products you need to taste and the more people involved will add time for sample evaluation, discussion, record keeping, and data review, but the additional sets of taste buds and the different perspectives that more people bring are valuable information for your quality program. A more detailed scoring and record-keeping system may take more time to set up, train, and use, but it can be quite helpful when troubleshooting problems.

Training your evaluation team takes a bit of effort. The amount of time you commit to sample preparation will depend on how many defect samples you want to make and how many base milks you want to use (whole, 2%, 1%, skim). Some defects, like light-oxidized, need to be made a day ahead to allow for the oxidation reaction to progress and form off-flavor compounds, and thus require some planning. Training sessions should be limited to approximately 15 samples, maximum, to prevent tasting fatigue, and this can usually be accomplished in about an hour. Several initial sessions are recommended to get the team familiar with identifying the defects, and periodic refresher sessions are a good idea. There is nothing wrong with shorter initial sessions and 10-minute refresher sessions on two defects in one or two base milks – it doesn't have to be complicated – make it fit your schedule!

How Much More Paperwork Will I Have To Do?

Just like with other procedures, having a written Standard Operating Procedure (SOP) for sample preparation and judging steps is a good idea and ensures consistency from session to session. The steps below can be used as a starting point and adapted to your situation, with appropriate modifications as you settle into your evaluation routine. 

The scoring and recording systems you choose will dictate the amount of paper, computer files, and time needed. Records should include the date, people conducting the evaluation, sample identification (type of milk, lot code, package size, etc.), scores, and notes. Assigning a numeric score will allow you to do a visual scan or graph the data to keep an eye on trends and catch anomalies. The recording system can be as simple as a spiral notebook with dated entries or as complex as a tracking system that integrates other information, such as raw milk analytical data and manufacturing information.

The most important thing about setting up a sensory evaluation program is to do it consistently! The sessions, scoring systems, data entry, and periodic reviews should be simple enough to incorporate into a daily routine for even the smallest-scale processor but detailed enough to collect the data needed for troubleshooting problems when they occur.  Don’t make it so complicated that you don’t use it routinely, and always be willing to modify your procedures or scoring as best suits your daily routine and product line. A little bit of data taken consistently is better than a lot of data taken sporadically for monitoring quality and troubleshooting problems!

How Do I Set Up A Scoring System?

It is a good idea to use some type of scoring system to make the process more formalized and facilitate troubleshooting. Assigning a numeric score to the samples will allow you to do a visual scan or graph of the data to keep an eye on trends and catch anomalies.

There are several scoring systems used for milk, which are discussed in more detail in The Sensory Evaluation of Dairy Products (2023) and Identifying and Preventing Off-Flavors in Milk and Milk Products (2023). Scoring systems can get complicated and confusing. The Collegiate Dairy Products Evaluation scoring system is presented in both of these publications as an example scoring system, but it is quite complicated and difficult for anyone but a dedicated college student to memorize and use consistently! 

Here is a suggestion for an easier system for a small processor to implement:

Step 1. Rank the overall flavor of your milk from 10 to 1, and

• 10 = excellent,
• 8-9 = very good
• 6-7 = good
• 4-5 = satisfactory
• 2-3 = poor
• 1 = very poor

Step 2. Record observations of off-flavor levels based on intensity.

• S = slight
• D = definite
• P = pronounced

Simple to remember, simple to record, simple to review, and very informative. 

Where and How Can I Do This In My Small Facility?

There is no question that many small and on-farm processors have extremely limited space available. Luckily, a short sensory session can be done in many places keeping the following things in mind:

• The space should be odor-free and well-ventilated. Off-flavors and odors in milk can be subtle and difficult to notice over other odors. Background odors to pay attention to in small and on-farm processing plants that can interfere with sensory evaluation include:
  • Product-related aromas from processing (cooked milk, flavorings)
  • Chemical aromas related to cleaning (chlorine)
  • Aromas from adjacent businesses (bakery, restaurant)
  • Exhaust odors from being close to external doors
• The participants should be as odor-free as possible. Scents that can interfere include:
  • Personal hygiene products (soap, shampoo, lotion, perfume, aftershave)
    * Hint: Keep unscented soap and lotion in the restrooms to use prior to sensory sessions.
  • Clothing worn in the barn or field can carry odors. Animal, barn, and feed odors from farm workers' clothes are often not noticeable to the wearer because we all get used to the aromas in the areas in which we live and work, but these aromas could be wrongly perceived as coming from the milk by others.
• The space should be well-lit for viewing the products. Color differences due to changes in fat levels in white milk are difficult to see without good lighting.
  * Hint: Consider purchasing an inexpensive desk lamp to help light the sessions.
• The space should be as quiet and free from distractions as possible. This can be tricky based on all the activity in the compressed area of small and on-farm processing operations. Do the best you can. Find an office or room where you can close doors or space away from heavy traffic patterns. Inform others of the importance of this activity and put a "Do Not Disturb" sign on the door for a short time. Another option is to choose a time of day when activity in the area is light, such as before or after the plant is done processing or before the store opens and customers are around.
• The area should accommodate the evaluators comfortably. Ideally, the option for all participants to sit is good, especially if there is a lot of discussion and note-taking. Standing for short sessions is fine, as long as people are comfortable and there is sufficient sanitary space for pouring and serving the samples, and participants have enough room for their evaluations.
  * Hint: If space is limited, consider purchasing a two-shelf cart with wheels and using the bottom shelf to move the samples from the cooler to the evaluation room and the top shelf for pouring samples.

How Do I Prepare For The Sessions?

Preparation for sensory sessions includes sample prep and supply prep. Taking some time when you set up your initial sensory program to order and organize supplies makes it easy to grab what you need for routine evaluations.

Supply Preparation.
Here are suggestions for supplies to have available for each session:

• Scoresheets, a notebook, or other evaluation records. 
   - If multiple people are evaluating, make sure everyone has the ability to take notes or assign one person to take notes for the group. Don't forget writing utensils.
• Milk sampling cups 
   - 3-5 oz cup, clear or opaque white
• Spit cups, a bucket, or a trash can
   - styrofoam, opaque, something you can't see through and doesn't have an aroma
• Water cups
   - can be the same as milk cups
• Water 
   - preferably bottled to avoid background flavors from well-water or treated city water; room temperature
• Napkins or paper towels

The number of samples and choice of which samples to evaluate is something you will need to work out based on your product line, production schedule, and shelf-life of the products. Keeping a sample of your products, unopened, in a home-style refrigerator until the end of its shelf-life will give you a chance to see what the consumers see. It is a good practice to conduct shelf-life evaluations periodically on your products. 

Sample Preparation
Using a routine procedure for sample preparation will help to ensure consistent sensory evaluation data.

Step 1. Temper the milk to 50–60°F (10–15.6°C). Off-flavors will become more apparent when the milk is warmer, so tasting the milk too cold will not provide a good impression of the flavor.

• Try as best as possible to keep the temperature consistent between sessions.
• Take out larger containers earlier than smaller containers to warm up so that all samples are at equal temperatures before evaluation.
• Commercial coolers are typically around 35°F (1.7°C) and home refrigerators are around 45°F (7.3°C)
• Be aware of the normal room temperature and that your tempering times may change with the season.
• Verifying temperature with a thermometer and recording it on the daily session data sheet is a good practice.

Step 2. Agitate just before pouring. Mix the milk jug gently by inverting it several times to ensure the product is uniform before sampling. Be aware this is not a good idea if you are evaluating creamline milk and you will need to find a method of pouring that ensures all your evaluators are tasting the same product composition.

How Do I Evaluate Milk Like A Professional?

Professional judges are just regular people with a good set of taste buds, knowledge about how products are made, a lot of experience tasting different products, which brings familiarity with off-flavors, odors, and textures, the integrity to leave any biases at the door, and the patience to focus on what is going on in their mouths and noses and relate it to what we know about the product characteristics, manufacture, and handling, and then make an informed evaluation of what we are tasting and what may contribute to both positive and negative aspects of the sample. 

Learning about your products, setting your gold standard, knowing where problems can occur, and their typical characteristics are your first steps in objectively evaluating your products. The more you practice and conduct sessions, the more familiar you will become with your products and the routine sessions will move quicker.

All judges follow these basic steps when evaluating milk, starting immediately after they are handed the sample:

1. Smell the milk. Place your nose directly over the container and inhale. Note any off-odors. Do not do this over the communal sample container for sanitary reasons.
2. Swirl the milk in the cup. Swirl the milk so that a thin film forms on the sides of the cup, and observe for any coagulated milk or other abnormalities.
3. Smell the milk again. Swirling the milk may release aromas you didn’t notice before.
4. Taste the milk. Take a decent sip and roll the milk in your mouth a few times, hold it no more than 4–6 seconds, and then expectorate. 
   * Hint:  Swallowing samples may make you feel full and make discerning flavor nuances between samples more difficult.
5. Breathe in through your mouth and out through your nose. This will give you a full impression of the milk flavor. The tongue only senses sweet, sour, salty, bitter, and umami – all other "flavors" are perceived in the nasal cavity.
6. Rinse mouth between samples. Some people choose to rinse with water, others use the next sample and take several sips to rinse any residual off-flavors from the previous sample. 
   * Hint: After objectionable samples, unsalted top "saltine" crackers, water, or going back to a good-tasting milk sample are all effective at cleansing the palate.
7. Think. Think about what you are tasting at the beginning, middle, and end of time the milk is in your mouth and what aftertaste persists after spitting out the sample. Is the sample good or bad? Are there noticeable off-flavors? What are they?
8. Score. Decide an appropriate score and comments and enter them into the data records.

Recommendations For Making In-House Training Samples

While there are quite a number of off-flavors listed below, a handful make up the majority of defects typically observed in milk. This will help narrow down the number of samples to make for any given training, and remember that not all defects need to be made for each training – train more on the ones you see more often in your facility.  It is also a good practice to make the defect samples in different base milks (whole, 2%, skim) so that you can learn how the background flavor of the milk affects the perception of the off-flavor. When conducting trainings, it can be helpful for the trainees to go from the mildest to the most objectionable sample.

Here is a suggested list of samples to use for in-house trainings:

S = 15 minutes, D = 30 minutes

¹S = Slight, D = Definite

Flavor Defects in Fluid Milk

How do you know when milk tastes bad? How do you describe what it should taste like? Describing the flavor of milk is difficult – it should have a mild flavor, with a fresh dairy-note, and be slightly sweet, with no objectionable off-flavors. Sometimes it is easier to describe the flavor of milk by what is not there! Once you know the flavor of your highest quality milk, then you can start to discern when off-flavors begin to appear.

All milk tastes different. Becoming familiar with the nuances in the flavor of your milk is an important baseline to have in your toolbox. Milk flavor is affected by species (cow, goat, sheep) and breed, animal health, feed, milking practices, and sanitation at the farm. Sanitation, temperature, and handling during transport from the farm to the processing plant can affect flavor, even if it is a short distance from the milking parlor to the pasteurizer or cheese vat. Nuances in milk flavor may be more apparent for on-farm and small processors if their herds are small, graze in the summer and provide rations in the winter, or have their herds on seasonal calving, compared with processors that have more consistent seasonal practices or that use commingled milk. The fact that milk flavor changes is neither good nor bad; it's just "different." It is to good to know if and how your milk flavor changes throughout the year. 

Categorizing Milk Flavor Defects

Traditionally, the dairy industry and literature have categorized milk flavor defects by their primary cause as absorbed (A), bacterial (B), and chemical (C). Some common off-flavors observed in milk, such as cooked and unclean, don’t fit within these categories and so the 2023 update to the Dairy Practices Council Guideline, Identifying and Preventing Off-Flavors in Milk and Milk Product, added two more categories called processing (P) and questionable (Q) to address these flavors, and provided the suggestion to "remember your ABCs and be aware of your Ps & Qs."

A confounding factor in troubleshooting flavor problems is that sometimes a perceived flavor can have multiple causes and sources, each with its own solutions. A general knowledge of the characteristics of common off-flavors, their sources, and mechanisms is a powerful tool for troubleshooting. Start by describing the characteristic flavor, and then see what the possible causes and mechanisms for that flavor might apply to your situation. Below is a general description of the major categories of off-flavors, followed by an alphabetical list of off-flavors to serve as an easy reference. 

Common categories and mechanisms of dairy off-flavors are:

Absorbed (A)

• Milk has a delicate flavor and can easily absorb aromas and flavors from the surroundings. Off-flavors can be absorbed into raw milk prior to milking through animal feed and through the odors the animals breathe in the barns and milking parlors. Off-flavors can be adsorbed into pasteurized milk by storing milk near things with strong odors or from the packaging material.
• Off-flavors that are typically characterized as Absorbed include:
  • Barny
  • Cowy
  • Feed
  • Foreign
  • Garlic/Onion
  • Musty
  • Plastic/Package
  • Storage/Refrigerator
  • Unclean
  • Weedy

Bacterial (B)

• Off-flavors resulting from unwanted bacterial growth can occur in raw and pasteurized milk and come from a variety of places, including contamination, poor sanitation, and poor cooling on the farm, at the processing plant, and in transport.
• Off-flavors that are typically characterized as Bacterial include:
  • Acid
  • Bitter
  • Fermented/Fruity
  • Lacks freshness
  • Malty
  • Rancid
  • Unclean

Chemical (C)

• Some off-flavors are formed through chemical reactions that occur within the milk, such as light oxidation and rancid. Exposure to light can trigger a cascade of reactions with fats and proteins that have different perceived flavors at different times. Exposure to enzymes can trigger a different type of reaction with fats and release different off-flavors.
• Off-flavors that are typically characterized as Chemical include:
  • Foreign
  • "Light-Oxidized" – this is the name of the mechanism that causes the reaction, but is usually used to describe the defect. Characteristic off-flavors associated with light-oxidized are cardboard, burnt hair, burnt feathers, paint-like
  • "Rancid" – this is the name of the mechanism that causes the reaction, but is usually used to describe the defect. Characteristic off-flavors associated with rancid are blue cheese-like, baby vomit, soapy
  • Salty
  • Unclean

Processing (P)

• Processing-related off-flavors typically come from incorrect processing, handling, and storage of milk and dairy products.
• Off-flavors that are typically characterized as Processing include:
  • Cooked
  • Flat
  • Plastic/Package
  • Foreign
  • Storage/Refrigerator

Questionable (Q)

• These flavors are difficult to describe and may not have an easily identifiable cause. They are often used as catchall phrases, and their source and remedy can be difficult to troubleshoot.
• Off-flavors that are typically characterized as Questionable include:
  • Bitter
  • Foreign
  • Lacks freshness
  • Unclean

Common Milk Off-Flavors

This is an alphabetical reference list of common off-flavors in milk. 

• Acid. A sour flavor is one of the basic tastes perceived on the tongue. Lactic acid provides the typical tart flavor of fermented dairy products like yogurt, but acid flavors are undesirable in milk. Sour flavors may be characterized as a vinegar-type flavor (acetic acid) or a lemony, citrusy flavor (citric acid).
• Barny. An unpleasant odor and flavor of a poorly maintained barn, milking parlor, or unpleasant feed. Barny flavors may be transmitted through the animal from the environment or absorbed directly in stored milk.
• Bitter. Bitter is one of the basic tastes perceived on the tongue. Bitter flavors come from small fragments that are released when enzymes break down long proteins.
• Cooked. Cooked flavors range from rich, caramelized, to sulfurous and eggy or burnt. They result from sulfur being released from proteins when higher pasteurization times and/or temperatures are used. Cooked flavors may be desirable or undesirable, depending on the product.
• Cowy. An unpleasant odor of cow’s breath.
• Feed. Feed flavors vary from sweet and grassy to dry hay-like to bitter and unclean, depending on the source of the feed. Feed flavors are acceptable in milk from grass or pasture-fed animals. Garlic, onion, and weedy flavors are also pasture/feed-related.
• Fermented/Fruity. Fermented/fruity off-flavors are often discussed together, but have different flavor characteristics.  Fermented flavors tend to be sour, like sauerkraut or vinegar, and fruity flavors are described as pineapple or apple. In some cheeses fruity flavors are desirable, but they are considered to be a defect in milk.
• Flat. Flat milk is described as having a watery taste and mouthfeel, and is less sweet. Flat milk can be due to low milk solids or from water in processing lines that dilute the milk.
• Foreign. Foreign is a catch-all phrase used to describe odors and flavors that are not commonly associated with milk. Descriptors include medicinal, chemical, sanitizer-like, and fruity. Foreign flavors can be absorbed in milk when stored near chemicals, in coolers with other products, or when product changeover from cleaning solutions or one flavored milk product to a different flavor is incomplete. Foreign flavors can be very difficult to troubleshoot.
• Garlic/Onion. Wild onion, chive, and garlic are natural plants occurring in some pastures, and their flavors can be absorbed into the milk.
• Lacks freshness. A stale or old product that lacks the sweet, fresh dairy notes of products made on the first day of production.
• Malty. A flavor similar to that of malted milkshakes due to bacterial growth in poorly cooled milk.
• Musty. Musty aromas and flavors may be absorbed in the milk from poorly maintained barn and storage environments.
• Oxidized, Light-induced. Light-induced oxidized flavors are described as cardboardy, burnt hair or feathers, cabbage-like, potato-like, and plastic. They result from exposure to light that causes a cascade of oxidation reactions with fat and protein. The flavor will change at different stages of the reactions, and can be perceived differently in milks with different fat levels.
• Oxidized, Metal-induced. Metal-induced oxidized flavors are described as tallowy, metallic, and old oil-like. Metal-induced oxidized off-flavors are no longer common in dairy products and were caused by exposure to metals with copper or iron.
• Plastic/Package. Off-flavors may be absorbed from paperboard and plastic containers and described as plastic-like, carton, and wet paper.
• Rancid. Rancid flavors are described as bitter, soapy, blue cheese, or baby vomit-like and are due to fat breakdown by bacterial enzymes.
• Salty. Salt is one of the basic tastes perceived on the tongue. Milk can taste salty from animals with mastitis and other health problems.
• Storage/Refrigerator. Characteristic storage or refrigerator flavors can range from stale to a general dirty refrigerator smell to whatever foods with strong aromas are stored in the same cooler with the milk. This is more noticeable at the end of shelf life.
• Unclean. Unclean is a catch-all phrase used to describe unpleasant flavors in milk that are not otherwise categorized. The flavor profile can range from a dirty dishrag or dirty gym socks aroma to an undefined character. These flavors are difficult to troubleshoot, but the first place to check is with sanitation from the farm to the end of the processing line.
• Weedy. The flavor profile is dependent upon the types of weeds that are in the pasture. 

Texture Defects in Fluid Milk

While the focus of this article is on flavor defects, there are two texture defects commonly seen in milk that are worth noting to help small and on-farm processors troubleshoot problems:

• Coagulated. Small bits of milk protein may destabilize and coagulate and form visible particles in the milk. This can occur either due to high acid production from bacterial growth or by enzymatic action, which is called "sweet curdling" because there is no acid production.
• Ropy. Some spoilage organisms can produce gummy substances called exopolysaccharides that make the milk stringy or ropy when poured.

References

Identifying and Preventing Off-Flavors in Milk and Dairy Products, Dairy Practices Council Guideline 38, 4th Ed. 2023. Kaylegian, K.E., Carey, N., Murphy, S., Wilson, S., and Traas, L.M. Dairy Practices Council, Pandora, OH.

Sensory Evaluation of Dairy Products, 3rd Ed. 2023. Clark, S., Drake, M.A, and Kaylegian, K.E., (eds.). Springer Nature, New York, NY.