Sensory Panels in the Time of COVID-19

Quality can be a very subjective aspect of food, based on each consumer’s perspective and expectations. For this reason, sensory panels are a critical component of new product development. But the COVID-19 pandemic has significantly limited the ability to conduct panels in the same way as in the past.

To determine the impact of this and what businesses can do to work around it, QA solicited input from three university sensory panel experts. Following are responses and recommendations from Mary Ellen Camire, Professor of Food Science & Human Nutrition, University of Maine; Julie Reiling, Senior Consultant, The Food Processing Center, University of Nebraska-Lincoln; and Wes Schilling, Professor, Food Science, Mississippi State University.

1. What impact has COVID-19 had on food/beverage sensory panels?

Schilling: It has made it challenging to conduct consumer acceptance tests with a large number of panelists. Our specific sensory laboratory has relied more on trained sensory descriptive panels to evaluate food products when we would have preferred to run consumer tests.

Camire: The loss of smell (anosmia) and loss of taste (ageusia) are common early symptoms of COVID-19. Potential research participants may not know that they are infected or that their chemical senses are impaired. People who cannot smell or taste food may still be able to evaluate appearance and texture, but the risk for the spread of the virus may negate any benefits of data collection. Also, fewer people can be tested at one time to limit their potential exposure to the virus, thus testing can take longer and be more expensive to conduct. We are still learning the efficacy of ultraviolet lights to disinfect testing sites and computers used to record data. At the University of Maine, we have modified graduate student research projects to be online consumer surveys so that students are able to graduate on schedule.

Reiling: We are a very small sensory lab. We don’t have a database of potential panelists that we can sort through nor do we really offer any QDA or focus groups. Being part of the University of Nebraska, we are subject to their rules and regulations. Classes were cancelled at the end of March of 2020. Soon after we were also informed that all in-person human testing was required to cease. Most employees were told to work remotely from home if at all possible. This ban was kept in place until in-person classes started again at the end of August. I had a sensory panel scheduled in early April that was postponed. That panel was finally run a couple of weeks ago. I have been fortunate to still have enough people willing to come in and taste as several of my regular panelists are still working at least part time from home and grad students are staggering when they are working in their labs.

2 How can food facilities continue to do sensory testing despite masks, social distancing, etc.?

Camire: Health screening for COVID-19 symptoms (preferably before they come in person to the facility) and an infrared forehead temperature scan are common ways to assess incoming participants. Can the traffic flow be modified to allow people to enter in one direction and leave on a different path to reduce exposure to infected individuals? Testing participants in smaller numbers and sanitizing booths or panel rooms between evaluations are two other strategies. I am still looking for a sanitizer that does not leave a lingering fragrance. Plexiglass barriers can be installed to protect staff who are greeting participants and/or distributing incentives. Portable, easily cleaned booths could be used in larger conference rooms if the food can be delivered easily. Bell Flavors & Fragrances delivered samples to in-house panelists in their offices. However, food samples may change in quality if a test is extended much longer than usual, requiring careful planning to make sure that all test participants have the very same food or beverage experience.

Schilling: There are a few options here. (You can) statistically model consumer panel data with descriptive analysis results so that consumer panels can be conducted on fewer occasions. Consumer panels can be done but there needs to be social distancing. At our sensory laboratory, not allowing walk-ins and only allowing a certain number of panelists in the facility at one time are important. It is also important for panelists to keep their masks on unless they are tasting. Use of Webex, Microsoft Teams, Zoom, etc., are also helpful because you can have the food prepared and served but have panelists in separate rooms on video. We have done this for product cuttings and descriptive panel training.

Reiling: Being part of the university, we have to follow the parameters that they put down on social distancing, mask usage, and flow of traffic. We are fortunate that we added an additional exit door from our panel room before COVID hit. With two entrances/exits to the panel room I can route panelists in through one door and out through another without them needing to cross each other. To allow for proper social distancing (at least six feet) I basically only run three of my eight stations during a panel which severely limits how many panelists I can serve in a given time. I normally have two one-hour blocks per day for panelists to participate in a given panel. However, because of the limitations of how many panelists I can serve at that time, we have moved to having the panel remain open from 10:00 a.m. to 4:00 p.m. Potential panelists can come when they have time available and don’t have to wait in line for a seat (or can come back when it is less busy). I have considered sending product home with potential panelists and collecting the sensory information over the internet, but we haven’t had to do this. I prefer to have control over the parameters of my panels so would really rather not rely on that.

3. Can you discuss some remote options?

Camire: Fortunately, major sensory evaluation software companies had already implemented testing in the cloud prior to the pandemic. Online focus groups were widely used already, so participant recruitment was both easier and more challenging. With more people confined to their homes, some individuals were more likely to take part in focus groups. Recruitment of consumers with specific demographic traits might rely more on external recruiters since fewer people are available in-house. Home use tests (HUT) may be more expensive and slower than laboratory or central location tests but offer researchers the opportunity to mine more data from the target person and their household members. Drive-up testing is also being used with test participants rating samples in their cars using smartphones. I have to think that trained descriptive panels are significantly affected by the pandemic. Our conference room went from a capacity of 24 to six persons, for example. How does a panel leader control the evaluation experience if panel members are all at home with children, parents and/or pets to contend with? The levels of sample control and presentation may be compromised with home testing.

Schilling: This would be similar to Zoom focus groups. This can be done and is very helpful with the right consumers. However, it is still not as good as in the same room because it is helpful to see the panelists interact with the product and each other. In-home use testing is very beneficial and should be increased during a pandemic. The main issue here is that people using the product at home brings variability in the data since one person or one group did not prepare the samples. Here, it is important to use statistical modeling to look for relationships between descriptive data, consumer data, and in-house testing.

Reiling: Although that is not something that I normally do, I have worked with a client in the past that contacted their customers about participating in a specific panel. They took care of mailing the samples to their customers and emailing them with a link to the panel ballot that I generated. I collected that data (via internet), analyzed it, and presented a report back to them. That worked very well, and I am sure could be done on a larger scale. You lose control over the samples and testing environment, but for basic consumer panels, I think that would be fine.

4. Why is consumer perception of food quality difficult — but important — to assess?

Camire: Consumers have very different opinions and those opinions are malleable. If your favorite brand is not available in the store due to pandemic shortages, do you become accustomed to a competitor’s product that is available? The importance of product attributes, such as (being) sustainable, local, and healthy, changes as consumers’ own economic and health situations change. The nine-point hedonic scale is now more than 60 years old. Is it time for a new paradigm for measuring food quality and acceptance?

Schilling: It is difficult because consumers are highly variable in what they like and their preferences. It is not possible to understand what consumers want without both marketing and consumer-based sensory testing. Descriptive analysis and home-use tests are valuable tools to help understand why consumers like or prefer certain products, but consumer testing is where the rubber meets the road.

Reiling: I think it is so difficult because everyone has their own definition of good and bad. Everyone has their own preferences and perceptions. I will always have someone that doesn’t like the samples that I serve and others that just love them. I think what we are really looking for is a consensus that a sample is liked by most people. We want to make sure that people like our food and will want to buy it before we take the chance of putting it on the market. A bad product released to the market could not only be a huge loss in money for a failed product launch, but could also potentially hurt the brand integrity.

5. Other thoughts/advice for the food industry?

Camire: We will need to evaluate how working from home will influence shopping and food preparation trends through the next few months of the pandemic and then as businesses begin to return to prior activities. Will the increased interest in from-scratch and semi-homemade cooking continue? How will consumers react to significant disruptions in the food supply chain?

Schilling: Conduct as much sensory testing as possible with respect to both consumer and trained descriptive analysis and use statistical modeling to understand the relationships between these data. Also, make sure that there is someone who is skilled in data analysis that builds a database with this information in order to have as much information as possible about your products when an emergency such as the COVID pandemic occurs.

Reiling: I think patience and flexibility is called for right now. Look for opportunities that may not have existed before. Stay optimistic that “this too shall pass” and that things will get more back to “normal.”

The author is owner of LJ Writing Services and a member of the QA Advisory Board.