Episode 8: Sensory Science and The Techniques We Use

[Music] Welcome to Food Products FAQ, a podcast where we discuss how we make food products and assess the takes to make yours. I'm Adam Yee, and I'm Branch Hyle. We're both independent food consultants who've been through the food industry for over a decade, and we're here to share with you how we make great food products that have been sold in thousands of stores. And today we're going to talk about what collectively makes a good food product. And we call this a sense of retesting. It's going to be fun. We're going to cover some scrappy methods. We'll go a little bit into detail of technical things, and we'll try to explain as best as possible, as well as drop some names in terms of who you can reach out to depending on your budget and things of that nature. Yes. Well, we are going to probably start from the very beginning of this. I guess at first of all, I guess how would you describe sensor testing? I think what we say sensor testing, we want to be able to know what it is, but it is not just about the science of tasting food. I would say this is the statistics of tasting food. Yes. A lot of this is statistics driven. It's figuring out qualitatively, so like the words, what they mean, and how they associate two flavors. So think of it a little bit like when people are doing wine tastings and things like that, the flavor wheel, how people describe flavor. But then there's also a little bit of quantitative. That's where we actually try to do scaling on a one to 10 or one to seven, one to five rating scale, and then try to associate the number based off of the question at hand and give us a little bit better statistics in how to interpret it. So there's a little bit of both. I will go into a little bit more detail into qualitative methods, quantitative methods, a little bit later, but essentially that's a very high level of understanding what people mean when they say certain flavors, what they like about flavors, and then basically sensory can be split into two parts. One is marketing, which is most of what sensory is, is marketing, branding forward. And then the other is quality assurance, quality control, which is what we covered in the previous episode. Essentially, it's like how do you maintain the quality or the flavor or sensory experience which concludes texture and visual from a QAQC standpoint as well? Yes, I want to just frame this. Sensory is a great way of quantifying how we evaluate anything in the food product. So this is important because when we give people a, let's just say cookies, I feel like cookies are the best example for any food product. If we give someone a cookie and they eat it, generally they just say it tastes good, right? It tastes good. I like it. I want to buy it. Those are good metrics, but they're not the metrics we look for when it comes to sensory testing. We get a lot more specific, and we also quantify that experience as well. So instead of like how tasty it's like on a scale of one to five how tasty. We're going to get to this in a lot more detail there is a lot of different things of all the sensory testing, but we're going to ease you whole into it. But in general, when we describe an attribute like tastiness, okay, it might be salty, then we describe salty, how salty. And then we quantify that for the scale of the one to five or one to nine or whatever. And then we gather that data and then we use some cool statistics to figure out what is perceived as that attribute. Yeah, let's dive into a little bit more of the scrappy methods. What are some scrappy methods that you have done at them in terms of just getting this type of information or data? Yeah, so if you are a food company who is just starting for instance and you just want to validate that your product is good, I actually would not recommend a sensory test because you can actually get a lot more information from your friends and your and your early consumers like a farmer's market or something like that. And there's a variety of ways of doing this. I would say though, however, that how I've built my two food companies is using these scrappy methods because for us, the data doesn't mean a lot when we get it specifically. Like it might have something about like, oh, this thing is too salty, but most of the times you can tell if it's too salty. Generally, how I usually view it is that when I work on a food product, specifically starting a company, I share it with two people who are my co-founders. And we have different opinions. Like I will think, because I work on it, I have an inherent bias on it because I think my product's amazing and it's perfect. My co-founder will not think that or they will like it, but they will say blah, blah, blah, blah, blah, blah, I'll fix a little bit and then we get to a consensus point. Then we share it with more of our friends because there needs to be like maybe investors, potential investors would be one of those categories, but also the smaller groups of people like at a dinner party and get their feedback too. And you know, most people who started food companies, at least what I've observed, started because they made it out of public and people really liked it and it thought they had a good idea and they went for it. And that's a decent method of validation. As you grow bigger and maybe you're in like maybe 20 stores and you want a consensus or feedback or even if you're just planning a launch, you can leverage those tools to make essentially what I do is a party and that party is like a giant sensory party where we show cool products and then we give them a form that quantitates out what I'm supposed to be looking for in a product. They market like let's say for a dumpling, we have a scale from one to five and let's just say we want this to be salty. We want the one to be not salty enough, three to be just salty or like the perfect amount of saltiness and then five to be two salty. And then we have tons of these attributes and they rank up, we gather data, we type them up in Excel and then we can see pretty clearly I would say just based on like maybe just 20 to 50 people if it is actually salty or not because well, we know from the assurgent test that we are targeting people. Generally, if you're a small business, you're doing this locally. We are targeting local people who are probably going to buy our products and therefore we have a decent idea that okay, these people think it's too salty. We should probably reduce the salt level. And is there a bias behind this? Yes, but in general, because you're quantifying it out, there's a lot less of a bias. Yeah, that's what's known as a jar type of test. So NUST about right is the acronym and the idea is statistically you want to be closest to the middle. And these type of sensory test gives you direction. So if basically if it's too much or too little, those are the extreme ends and in the middle, in this case three, if it's a one to five scale, you're basically allowed a direction to decrease or increase so you can get into that middle number. So that kind of helps you as a founder to kind of say, oh, instead of saying, I just want more sweetness. It's like, well, how much more? And that's kind of quantified by the amount of people. So it's not just a standard, a sample size of one person or two people. If you get enough like 10, 2015 or even up to 30 people, then you have a better way of understanding the direction of how to improve the product as you're going through the R&D process. Yeah, and I do agree with that. There used to be a company called Tinker Kitchen out in San Francisco. And they partnered up with Kitchen Town, which is an incubator accelerator program in San Mateo, and they did those sensory testing with about 10 to 12 brands showcased different options and people were able to go and taste test as if it's a unique experience. So you could do this by yourself or you could partner with five, six or have any brands in your ecosystem to do a unique tasting where people would be able to kind of pay to kind of try this out. And that kind of helps with some of your sampling fees and just, you know, buying all those little materials. And you can create like an or derp type of networking sampling opportunity. And I think that's where you can become creative in that sense too. If you share booths at say, um, farmer markets or trade shows and things of that nature, I think there's opportunity there. If you're at a trade show, typically people are more interested in just buying the product. But if you're able to kind of incentivize people, I heard one time from a sensory analyst that if you provide a raffle where people would fill out a survey from a QR code, people are more likely to just try it out and keep it to minimal like five questions. That actually allows you to kind of build in not only sensory evaluation, but to be able to collect interest from people for your subscription lists. If you're doing newsletters and things of the nature, so you're able to build out your marketing, reduce your customer acquisition costs sooner and kind of tie it into your whole sensory evaluation or the marketing side of things as well. Yeah. And I would say that like those are all really good ideas. Demoing is also very strong indicator that demoing is really hard to gather quantitative data. It's very good at gathering like anecdotal data or what do you feel has been kind of a super feedback response to that, which are essentially sentences. Senses are very important as well. We'll get into a lot more detail. And yeah, I'll say that how I always rationalize it is that big companies do sensory tests to have assurance of a launch. Like they are doing this to make sure it's there a lot. They're gut checking specifically by a third party, but those are as internal that this product is right. already to launch and therefore they are willing to invest into launching that product. For a smaller company, you don't really have to do that because launching is probably as expensive as doing a SESR test, honestly. - Yeah, it can get up to, I mean, I used to work at the likes of Gary Jolly Chocolate Company. It can get up to hundreds of thousands of dollars depending on how many people you're asking to do the test, what you're asking them to do, and how many samples you're gonna prepare, where it's gonna be shipped out. It's a lot, it's a lot of work to do this type of testing. They reason it out because if you're that light of a company, they spend a lot of time in R&D and they can launch a lot of products. If it's not a good product, it could be like the Coca-Cola that was the new version that was thrown down the drain because people didn't like the flavor profile because of the changes that they've made. I'll plug in here that if you guys are having trouble, especially with a small business, trying to think of ways to gather data for sensor tests, you know, let me know. And I'm happy to at least structure away where you can gather pretty valuable data with the small amount of people. - Definitely, yeah. - We're now gonna talk a lot more about how the industry does it and you can take away from this as much value you want. But this is a standard that we do in the bigger and bigger your food company is. The more likely they're gonna do these types of tests. I find it quite fascinating about these tests and the data you get from these are pretty cool, I would say. So, you know, whatever budget you're on, like hearing this, one, you might find innovative ways to do this the same way big companies are doing, but this is what we've seen in the industry in sensor tests. So, Brian, how about you take it away in terms of like, how you've seen, I guess, sensor tests be approached or have been kind of communicated with brands. - Yeah, I offer the service of very basic sensory evaluation. If you have the budget for it, that kind of gets built in. It could be anywhere between five and 10K. But when I say basic, it's more for quality assurance, quality control. So, if I work on a chocolate project, for example, then a lot of it's just talking at length of, what do you mean by dark chocolate? Like, what kind of flavor profiles are you soliciting? - Any kind of a proctor? Is that a proctor role? - Yeah, it's a proctor role. It kind of goes into descriptive analysis, which has been established, I think, in Stanford and then transferred over to UC Davis as well. And it's a pretty standardized approach in understanding what you mean as a founder or founding team, or a decision maker when you're saying, it's too much acid or it's too much bitterness. I'm like, okay, let's break it down. What exactly do you mean by that? That really applies in more commodity or complex flavor profile type of products. Chocolate is one of the easiest. Chocolate slash cocoa, vanilla, strawberry. Those are, there can be pretty complex because of that flavor profile. But for those of you who are looking into just making bars and you just wanna know how to work with a food scientist and quality assurance, quality control standards, then we kind of look into the jar question or we try to look into triangle tests and dual tests, AB testing, if you will. And so for those who are more market forward, we look at AB testing, you have one or two options, you test which one is better between the two. They kind of helps facilitate in the conversation of what do you like in version A that we give you versus version B, and then we can actually progress in that sensory parameter because now we actually have similar things to talk about in terms of changes that are necessary, in terms of quality standards that we are setting for the contract manufacturer, the golden standards that we are determining. And in that way, when you're at scale with us in the commercialization trial, we can taste test together at scale and say it's too salty or not salty enough because we have built in kind of like the sensory qualitative measurements together. If you want to go a little bit more technical, this is where you have descriptive analysis and focus groups come into play. Of course, when you hear where focus groups that means there's a lot more people. A lot of that becomes more expensive and that's kind of ties it itself into the whole CLT central location test methodology. This is where it can be hundreds of thousands because money is being put down to get people coming in on a consistent basis that is your target audience to be able to taste test and provide you enough feedback through understanding when they say grassy notes or when they look into evaluating earthy notes or bitter notes from pea protein, et cetera, you have a strong foundation of the answer in terms of direction, in terms of how to fix the overall problem. But for all intents and purposes, when I say basic, it's basically to be redundant, it's basically just trying to train you to understand how to speak this language and trying to understand when do you want to deploy a CLT test. It's typically if your very early stage and run very lean, you wait until you're in the millions or hundreds of thousands of dollars and have at least anywhere between 30, 50 or even 100,000 to deploy in terms of cash for this type of testing. It becomes an assessment about risk. And so if you're very early on, it's just, how do you understand what is a triangle test? What is an AP test? And triangle test is basically three options. And how you go about that is when you have already launched a product, we do triangle tests to figure out, especially if you're reformulating for cost of good solid reduction, does it taste the same? So basically it's like that sesame street thing or there's like one of these things is not like the other. So you're hoping that people pick out the bad option or the reverse switch. Or you're not hoping, right? Yeah. Exactly. A layer of reverse in which you hope that the change you've made is the same as what you have launched. So you can have a claim that's more fiber, but it has in sacrifice, it's sweetness level. Or if you reduce the salt that it's similar saltiness perception. So you're hoping those changes, subtle changes are maintaining the entire sensory experience for the end consumer. Yeah. So I think something to really touch up on, we're probably going to repeat this pretty often, is that sensory tests are designed to look at a very specific-- it's actually a change that you're going to make to the product. So if you make a change to a product, you either want a test that will highlight those changes or to essentially diminish those changes. So I would say one example is a triangle's test. One of the more common ones I've done, for instance, are try these three samples, which of the two tastes the same or rank which one you think is the best. Or there's several different statistical methods you can use. I always say these are all pretty simple statistical methods, as well, right? I think top tier sensory firms uses p-value and all that. Honestly, you can probably tell which ones are different based off of graph alone. And honestly, most executives only care about the graph, the pretty pictures about how big a spike, something is different or not different. Yeah, again, it goes back to the budget, right? It goes back into how much risk you want to take and how much risk is on the table. If you're a startup, you're already taking a lot of risks. So do you need to take on a budget for sensory evaluation? The answer is no. You are adapting in real time. Whereas if you already have products launched and you have a brand that you need to keep the reputation of, that's where you want to kind of mitigate as much risk. And that's where sensory evaluation comes in hand. Because the last thing you want to do is you launch a product where you already have brand reputation and then people don't like that product. And then now you lose brand reputation. And then now you have a lot of product that you're just trying to liquidate as quick as possible or just destroy. Because that's essentially how much risk is at stake at that point and how you can mitigate through sensory evaluation. And then a new test that people are looking into are quads. So now you have four options for the consumer. This is more complexity. It's about thinking about the frameworks. Like now you can have the opportunity of two. That could be the same and two could be different. It could be one is the same and the other three are the different. I'm sorry, one is different than the other three. And the framing of it is if you have fewer people to test, let's say it's like five to 10 people. Use a quad test because that's a stronger test to ask people. Because it's much harder for people in the smaller group. It's much harder for them to kind of just do a random guess of one out of three. So you kind of want to kind of balance that out. And that's really happening from a lot of flavor houses. And sometimes you could partner with a flavor house to kind of run those types of tests. So for example, McCormick Phona, which is flavors of North America, McCormick being the spice company that people might be familiar with, they do a lot of those tests so that you can actually hide their spices, seasonings or the flavors that they're extracts that they are selling into the market. Interesting. I think you and I both suffer through this where a lot of people A lot of the sensory tests can be kind of a 50/50 shot of the dark of success. And somewhat argue that's because the sensory test is only going to evaluate a product. It doesn't really evaluate anything else or it's very hard for it to evaluate everything else. So, you know, even if we do get perfect data, if it's a done in a sterile environment where all the factors are related to product and it doesn't translate well in the field, that brings up a lot of questions. Almost every company that has had a failed launch and have seen sensory tests look good, always have this kind of question. So, and I think there's a lot of tests that kind of fix this or they try to fix it or they try to explore it a little bit more. One example Brian, you put on our little cheat sheet is a HUT test. - Yeah, so those are home unit tests for short. So a bunch of acronyms, CLT, central library test, HUT is home unit test. What that is, is there are samples that you create and then you ship it out to anticipated end customers. Those end customers will then receive the samples, taste test that in their own home, in the right environment, and answer a survey in of itself. And then of course there's an incentive typically to hopefully get those people to actually answer and not just get a free sample for the sake of getting a free sample. So, there are companies that are, it's called less highlights. For example, that's out in the New York area and basically they've kind of provided a SAS model, if you will, for anybody who's like small all the way up to the large Fortune 500, to pretty quickly iterate or test in real time. They call themselves as more agile. Their surveys can be pretty cost effective and you can get 30, 50, sometimes 10 people, depending on what you're looking for. They do ask for a 10% extra in the case that, you know, some people, whether they get sick or they don't receive the product or they don't actually fill out the survey, that you're able to hit the extra numbers of people to get the actual data off of hand. So there are plenty of other companies that compete against less highlights that leverage the HUT model. The benefit there is you can actually go further and work with these sensory evaluation, marketing evaluation companies with an HUT to do Zoom calls and actually see how they eat the product or use the product. So there's longitudinal studies in which you can have these Zoom calls where you can actively see if it's like clothing, where you focus on food beverage supplements, but there are other CPGs where you can actually analyze the performance in of itself. That's not to say you can't apply that into how somebody prepares a meal if you are an ingredient in a meal preparation type of product category. - Yeah, I'll say like these types of very, very personal tests are really hard to do and really hard to glide data off of because the consumer has to be so involved in this. Even though they are getting incentive, it's still decently hard. But yeah, but the data's pretty clear in terms of like, how finicky your product is, like, where are they storing it? I mean, the data is really good, especially if they make the same mistakes over and over again. And different people with the same mistakes over and over again. So I like HGT's a lot. I've struggled a lot of them in the companies I've worked for just because it's so hard to get data from them because most people don't do it. Maybe incentives aren't that good for them, but it's something I've noticed specifically. It is quite a lot of laborious work to have someone. I've done it where they have it and they journal about it. And you know, like some people, yeah, you look at the journal, how you use the product and when you eat it and all that. And that's a lot of work. That is a lot of work. Yeah. And also, if you look at the supplements or function of food space, there's another company I sit as an advisor to in the Cupertino, California. It's called Tastermonial. So they part it with glucose monitoring systems. So if people are diabetic or people just want to look at how much their glucose spikes when they're eating food, Tastermonial will basically run an HUT type of test and synchronize that with the app that your wearable device is, so that you can actually see as a consumer, the CPG product that you're eating if you have any spikes in glucose. And to see if that extra functional claim of low glycemic or keto is actually valid for for yourself. And on the back end, if you are the brand owner, you can see in real time, you don't have to go through clinical trials. This is just raw data for you to be able to present. It's like, look, this is how many people that we tested and we actually see the biometrics involved in that. So if you're in the supplement space or you're in the functional food space, if you're trying to go low glycemic, low added sugar, this is an alternative to take a look at data and say, yeah, this is marketing claims that we can present and or some sort of scientific data, not necessarily a full clinical trial, but we may not need to go through a full clinical trial to prove this. - Yes, indeed. I think that there's a lot of innovation going to this brand. I think there's another one for our palette insights. I think it's starting to get some buzz in California, for instance, for kind of this scrapier way of doing such. Basically, I think a lot of you'll do this, but I think they're doing this more focused, doing these specific at restaurants and hosting events at restaurants. They get people to try food products over there and rank them. And there's certain ways to do this as well. Emote, if we did some launches that involved ranking as well. But honestly, for a business to do a sensory test, there's a lot of bias. And we use those metrics from marketing purposes, one that we do, anything else. - Yeah, I think at the end of the day, if you look at how sensory evaluation marketing firms that offer these type of tests, it's really geared towards the marketing and the elements side of things. Some of them are adding the value of, here's raw data that you can use from scientific means, but at the end of the day, it's really most of the companies that have the budget, it's coming from the marketing side of things rather than the scientific R&D side of things. - Yeah, and I'll say there's, oh, we forgot to mention one other test that, do we mention specifically descriptive where they kind of case they product and then they rank, how spider plots, right? Do we mention it? - So yeah, going back to the descriptive analysis, usually you'll get panelists that, after the same panelists usually are in the focus group that identify attributes like grassy nodes or it's fluffy texture, et cetera. And then those same categories of attributes would then be tested every single week for X amount of weeks so that the panelists would then, over time, have a stronger statistical understanding of what they are tasting, so that they'll rank it, so let's say it's a one to eight or one to 10, one to 10 is easier. They'll know if it's a very strong attribute of fluffiness or strong attribute or weak attribute of spiciness, things of that nature. And the reason why it's on a weekly basis is, you don't want, you want to prime and you want to teach, educates that panelists to understand your product very intimately. Otherwise, it's just gonna make random guesses and then your data is like, just doesn't make any sense. Sometimes when we get all those data points, we can cluster them and see, oh, is it gonna move in one direction and we try to take an average across the board and see where does that stack up against other panelists? So it helps people from an R&D in very technical sense. This is a very expensive process by the way, that's why you get hundreds of thousands of dollars put into this. A lot of software, a lot of statistics, and a lot of preparation as well as providing this panelists like $100 after 10 weeks in a gift card, just for them to come back every single week for 10 weeks just to taste the same cracker or cookie or bar or juice. And it gets boring for the consumer. It sounds fun at the end of the day, but if you are the actual panelist, you have to commit every single week to get this $100 and you have to taste what could be like, a very sour cracker, you don't want to eat the same sour cracker, but your results are interesting for the vinyl product output. - I have a funny anecdote story about that with ice cream. It was a summering in my university, Cal Poly, and they have a good sensor program, but they had these tests where I think they would pay you, a good amount of money to try 15 sales of ice cream. And I was like, yeah, this is great. And then at the point, there was a point where I was like, this is so hard because they were all really good, actually, but it was just so much ice cream. I got so much fatigue on ice cream. It's just, yeah, this is a note for, I think there's two notes on here. You can use your university, but two, the expertise of the questions ask really matter. There's just a lot of nuances on the questions that you have to ask to get a straight answer. And there are plenty of design structures, experimental design structures that just may, you'd be surprised don't work. One example is overcriding with too many samples in a test. Another is just not asking the right question, like directly. Most common one is probably ranking favorability, like some people just completely forget. To rank like, what is your favorite product out of the ones tested? And actually, gives you a very strong indicator of what to do next. Because for us at mobile foods, that has been actually the most important question that we've asked our small data set is what product you're looking forward to the most. And then when they try it, what product was the best here? Those simple questions actually do have a lot of impact. So sometimes with sensory test, you can get two in the weeds and you can have certain types of data that might be interesting to you, but may not be interesting to a lot of other people. It may be interesting to do an ingredient supplier, it might be interesting to do an investor, or might be interested as scientist or even a consumer. And so being aware of that, I think is really important as you design these tests because honestly, every question that you try to answer will cost money. - Yeah, it takes a lot of thinking to get the right answer. And they also have to refine them out of questions. If you don't want to present a survey that's 50 questions long or takes an hour because people lose interest. I mean, I remember when I was in school and I did the whole descriptive analysis and panel attribute analysis as a panelist. And I was eating bars, one had spirulina in them and I just didn't want to eat spirulina after a while. So I had to bear through 10 weeks of eating spirulina bar because I was the most disgusting, but that's what they needed. And so it may not be useful to think of, you have to think about what the consumer is interested in and how long they can actually do this type of test. You don't want a consumer to do a test that's gonna take more than 30 minutes. To be honest, for a lot of these type of, whether it's H-U-T, CL-T, whatever it is, you don't want to spend more than 30 minutes. Actually, for most part, the early stages, you don't want them to spend more than five, 10 minutes to be honest. You want enough short answers for you to get going in a directional manner for R&D purpose, sales purpose, marketing purposes. - Definitely. So they also want to test them on, especially even the kind of, if you can't afford sensory testing. And honestly, most of the younger brands are going to focus more on direct feedback. And one thing I found extremely strong in almost every single category, whether it's selling more product, whether it's exposure to brand, and specifically, very basic sensory testing is demoing your product into stores. And I've recently gotten a few clients that are interested, or I've done demos for them. One, I like it. I'm an extrovert. I actually do like talking to people. And so there's even not my product, I've noticed. I think you have sometimes a bias, right? And you get kind of offended. At least I do, I don't know about you, but if people don't like it because I know it's good and other people said it was good, that's a different story. But point of view is that, I've had some friends that have asked me to demo for them. And usually what I've been noticing, demoing has been kind of very strong indicators of purchase intent. And pretty much I've been writing reports for them, specifically about hooks and what sells, and what are some complaints and all that. And I don't think a lot of people who do demos do that, usually what they do is hire high schoolers or college students to sit around for three hours and try to sell something, but they have really no incentive, except for waiting for the amount of hours it takes, right? Like, most brand ambassador paid it hourly. I've only seen a few of them actually get paid for amount of sale, right? Like our kind of sales commission, that's what's called. Yeah, it really does have an amount of some people. But I turn it as a game, I like selling product. And you know, you'll understand very quickly, especially when you do a good amount of them, what sells and what doesn't sell, and why they don't sell. And when you're saying every three hours, trying to convince them to buy something, you just learn just on the fly through just a lot of touch points, what people want, where they want it, why they want it, what time they want it. These are all a lot of things that you can kind of tracking your head to make a consensual story about it. Not a consensual story, a consensus of a story about it. One example I did literally yesterday was a company called Game, it's our friend Candace Troy. Did you meet her in Chicago? I was small, yeah. Her son. I don't think in person, but I've heard the seaweed snack that she does. I've been introduced to her by email many times but I haven't heard. She hasn't ever replied. She's a busy person as a busy person. Yeah, well, she finds me. Not to say. I mean, we help each other out quite a bit. She doesn't want to fly from Chicago to-- so she just asked me to demo it. And I demoed it. It's a store in San Francisco called Buy Write, which is kind of this-- I would say it's a premium store. Like a lot of early CPG brands go to the store. And I sold the product there. And I demoed the product there. And you can tell that when people try it, they really like it. They find a unique experience when trying that. But people have a very hard time understanding what it is on a packaging perspective. And there's many ways about it. One, see what snacks-- yeah, people know what they are. But since they're so different, you can't really decipher that in the package. Two, that the flavors are also very hard to justify. For example, there's a spicy flavor. But when we say the word spicy on a package, people have a lot of different metrics on the word spicy, right? And I will say, these CV chips, though they're not that spicy, they are the best seller when people tried them. And when people tried them, they picked them up and bought them instantly. So that tells you a lot just off the bat about your consumer's interaction with the product. They also ask you questions like, why is this product salty? Well, let's see what it comes with the sea. And you can't say that I'm from the face. You just say, have a nice day. And you're going to have some apprehensions, for example, with so-called foods, plant-based meat is polarizing. And some people don't like plant-based meat. And they'll either say to your face or be extremely aggressive about it. So in my opinion, demoing has been kind of a surefire way of having small brands understand directly how good a product is. And because you're targeting literally a store, which has a demographic itself, it becomes a very strong data point when you take the time to analyze how well a demo went to see how you can position it or improve the product going forward. Small story. Small story, though. Well, it's a very powerful story. And I think that's that should be recognized and should be understood that in-person interactions is a very powerful way to make the connections and build up the rapport with more customers to buy your product. Whether it's in-store or if you're just tabling and recommending D to C, that allows people to start buying your subscriptions, because now they understand how to use the product or give you enough good feedback on the packaging design or shipments and things of that nature. If you wanna do it yourself, you start to see that there are people that do sample packs, sample kits online. And I think that's where if you're able to figure out a way of getting people onto your newsletter and then having them request a sample, sample kit, and then fill out via QR code, et cetera, and build out more incentives, you keep them on the hook through that newsletter, you keep them on the hook for the subscription process or getting them to a subscription process from a D to C level. - Yeah, so there are tons of ways to get this feedback. And I will say, I think a big folly of a lot of small brands, they don't know how to capture that as valuable data. And sensory essentially is that you're capturing consumer feedback through data to make your own decisions. And the data doesn't be complicated. It can be a simple Excel spreadsheet with a certain number on it and then you can play like a mean, median average, whatever. But it is important to capture. If you can't use it now, you'll probably use it eventually. Sometimes anecdotal data is not enough and sensory data, whether you could spend millions on it or a hundred thousands on it, but in general, it is just how the industry progresses in terms of, it is how the food industry makes things, makes things work and it alladates things and markets things. They do use sensory data. A lot of people just don't know this because it's such a really, really, I don't wanna say complex. I don't know what you wanna call this type of industry, but it isn't industry that is actually pretty important to it. - Yeah, it is very complex because it's about decision making. It's about the economics. It's something so subjective per person that you need to quantify and objectify so that you can actually make a decision yourself. And that's the thing is flavor sensory experience that's so personal to the individual. But there are group of individuals that have similar purchase intent or purchasing power that you want to be a bit adequate. And I think as we look into the power of AI and the power of improving logistics and the power of improving market analyses in survey, I think these HUT type of sensory firms or companies as SaaS models, they'll start to lower their pricing so that it's easier to onboard a lot more of these customers. And I think there is an opportunity that sets to be able to say, well, I can only do this. And then just asking somebody like let's highlight in their emerging brands division to see, Is there something that you can do? something within budget or something that you can work together on. And then also taking a look at, again, incubator and accelerator programs, I'm sure there's different incentives because this is very important for you in your marketing, your branding, your R&D, and your sales. That's something that is more that creativity side and just asking people, like, can we do something in this timeframe, in this budget? And I think there will be a reception because there's innovation in this space, especially on the tech side. There's a really popular AI platform going on right now with sensory testing and marketing analysis. I'm trying to find out my emails, actually. Do you recall what it is? It's when I hate about these companies, it's just a mate. So you never know what it is. Yeah, I get a bunch of them, so I don't, I don't know. Suzy. Have you heard of Suzy? No, I don't think so. Okay, I actually think it's actually really good. I'll surprise about how good it was. Yeah, if you go to like Suzy.com. There's a bunch of them, most of them are probably tractionously, but I'll just say, for example, I've been exposed to this one. BC used it and a friend of mine found it very valuable. It's called Suzy. Suzy.com, it is a pretty good, I would say, a visualization tool. So it has the ability to not only do like very quick sensory tests, and these are mainly surveys, but they're able to visualize your surveys off the bat, too. And it uses AI, whatever. What I'm trying to say is like, there's a lot of technology actually going through this point of the industry. This is an industry that's really ripe for AI, because it requires a ton of data and it requires a ton of visualization and requires a ton of work for lack of a term, right? Yeah. And you need expertise to have really strong knowledge to, I don't want to say understand it, but to give you tools to understand it. And that is useful and who knows where is going to happen when we look into these types of new tools. But this is coming. And this might be more accessible to you if you have the budget to use something like as Suzy.com. What I'm trying to say is that there's just a lot of ways you can utilize what we know for your products just on your own. Yeah. At the end of the day, when you look at sensory evaluation, again, we throw you a lot of different terms. This is more to empower you, not to scare you off. And this is really for you to ask the right questions for any type of probably developer R&D person or anybody. And that's a fruit scientist in their field. Fruit science is multifaceted, has different sub disciplines. And so what you want at the end of the day is to be able to ask the right questions and see if that particular individual or firm can help you in XYZ division. So some could be very specialized in just R&D. And that's okay. And some could be very specialized in sensory science. And that's okay. And those are things you learn typically from a food science program. And sometimes you may have journalists that can do a little bit of everything. So at the end of the day, these are just ways to inform you and to ask questions. How did I look into sensory evaluation? If a food scientist out there is kind of scaring you to getting into sensory evaluation from the get go, you may not have the budget for it. I think you should just take a step back and say, is this the right time and place to hire this particular individual? And if not because of your budget or where you are as a startup, then I think you just keep that person in mind for the future and go about figuring out creative ways to getting the data that you need. Yes, indeed. Let's go wrap this up with some really, I want to say framework questions that you would ask a sensory person when you design these types of tests. And what I mean by that is like, if you are like, let's say a low budget brand and you would ask either me, Brian, or a sensory firm, how would you approach them? Yeah, I think it depends on the stage of R&D. I think if we're near the end or even just very beginning stages, one of the questions is about purchase and tense. So I'd actually ask about like, what's the likelihood of you buying this particular product and/or what's the likelihood of buying this product at X price point? What's the MSRP? I think that helps drive a lot of conversations with product development or R&D in general to look into the cost of its sold that you want to operate and the margins you want to operate. It sets realistic expectations on both ends. And it also sets expectations on where is the demand for this type of product and product category? And how do you want to position this product? So I think that's one that's pretty important that's not necessarily asked early enough. And then the others are more around the sense of a flavor profiles or sensory experience. If you're in the alternative protein space or protein space in general, often something that's overlooked is texture. In terms of let's say even a protein bar, like what's what is the texture that people are interested in? Because essentially a nut butter bar in the refrigerator is a different texture than what you get from a shell stable protein bar. And so that also kind of helps in terms of setting realistic operational process, the parameters in the discussions in the R&D process. So that as the food scientists or product developer, we'd be able to say, well, we understand where you're trying to go in terms of innovation. But we should also take a look at how you should frame that question of the texture or the visual or the flavor profile in realistic terms. We can try it out. But we should also kind of tie it to what is the cost associated with it? What's the feasibility from a commercialization standpoint? And what is the feasibility from an ingredients standpoint? So those are kind of things to kind of high level type back into the R&D process. Yeah, no, no texture. I got grilled from this for my manager, Motie, for names, Dylan. And I just have to give her credit for really grilling me on sensory, entropy, specifically on texture because texture is really complicated. And it is not a lot of like lexicon and literature of it. And you kind of have to define your own. So one example is like crutchiness or springiness or guminess, right? These are all like the very specific attributes for a specific texture. You can't say like how good is the texture of this product, right? You should be pretty specific on the texture you want to achieve. And generally you would have to compare multiple different types of bars and figure that out. Some have some more positive connotations than others, but in general, texture, what we've noticed is that different people have a lot of different understandings about texture of attributes. And you have to as kind of the main scientist of the main communicator of these attributes really, really compare or like set the tone about what this certain texture is. Sweetness, all these people can generally understand what those are, but when you say like gum iness or springiness, that gets a lot harder to have preconceived notions of. So therefore, when we make these attributes or we look at these attributes or we redesign these tests to focus on these attributes, it is our job to communicate what specifically this attribute means. Yeah. Going back to texture, I mean, Adam and I are East Asian and there's a Chinese term called QQ. They can't really tell, get they? That's true. But like our cultural heritage is of Asian descent and there is a term called QQ, right? And what does QQ mean? It's a texture. It's a bounceness of boba balls of certain noodles and things like that, which is very hard to translate over from a Western audience is it's like, what is bounceness? What's QQ? Like, how do you identify that? And unless you are actually tasting something, it's very hard qualitatively to translate that into something from a Western audience perspective. You know, I guess the same would be said about al dente. What does al dente mean? Italians know it and Western audience comes somewhat knows it, but al dente could be very different based off of how long you cook that pasta itself. And is it only applied to pasta? Could we apply the al dente framework or thinking to other textures? There's not a lot of understanding that. And I think that kind of broadly encompasses sensory in general and going back to the whole subjective versus objective. It is very much subjective objectiveism. If you kind of think about it, we're trying to take our personal experiences around flavor, around what we smell, you know, or what we see. Like we know, even unless your color blind, we know what red looks like. But the hue of the red and describing red is very hard. And then if you're looking at smells of things, like it smells floral, what does that even mean? Like it's flowery, but what kind of flower or like mouth feel? And so there's a difference in terms of texture and what mouth feel could be. Like how do you classify spiciness? Because essentially that's a feeling because of how you eat something. Chemically, your brain thinks your mouth is on fire because it's from a chemical sense. It's literally not on fire. That's more of like a mouth feel, like your mouth feels that. And the same could be said with mental and mintiness, like your mouth feels cool. But that's It's not necessarily a texture. And so we have to kind of think about how do we classify these words so that all of us can somehow be aligned on the language that we use. Fundamentally, it's basically communication at the end of the day. That's effectively what we're trying to get across is a lot of this is communication, a lot of this is thinking about how do we communicate with each other. There's a lot of ways to miscommunicate. So we want to make sure we try to minimize that so that your marketing is appropriate. Your sales is appropriate. Both of those are communication methods or around communication as well as R&D. Because we have to communicate with you as much as you are communicating with us. We cannot read minds and we need to be able to make sure that the sensory attributes are similar within the realm of what we both think is appropriate for launching into a brand. Yeah, basically, sensory science is the Rosetta Stone of Taste. Yeah, of Taste and Flavor. That's a good one. Am I trademarked that? Sensory science is the Rosetta Stone of Taste and Flavor. I like that. Anyways. But yeah, it allows us to translate these complex terms. Oh no. It allows us to translate feelings and positive negative feelings. It allows us to translate positive negative feelings into words, scientific words, then we have to translate that into commonplace words that we can, yeah, math. Oh, yeah. First and it's math, which is the language in itself. Yeah. Language in itself. And then data science and visualization, which is another language, and then eventually translated into something simpler for kind of the wider audience of our company, which then they'll translate to consumers. Exactly. Another translation happening. I mean, I think every single episode, there's some sort of translation, translating of the quality assurance, translating of how manufacturers operate and translating how suppliers operate. A lot of this is knowledge that you may not be aware of. And being able to translate it to something that you're able to digest in your mind and then be able to translate that over to the remaining team or other stakeholders or investors so there's a lot that's happening. And I think that's the importance, but also to realize that just because you're scared that there is an importance in sensory evaluation, doesn't necessarily mean you can't be creative in finding a solution within a small budget or no budget. Definitely. There's a lot of isn't out. There's a lot of complex stuff. We gave you the whole dammit of what is available out there and what you can use, whether you have no money or a lot of money. These are the options available and there's tons of ways to get consumer feedback and you can do it in a variety of different approaches that we've aligned with you today. Hope you all enjoyed this episode. I know I did. I actually love talking about sensory science. I hate doing it. I'll tell you that. The only fun part is if you are tasting tasting products and if you're not being grilled for, if you're not actually doing the test, if you're kind of like, I'm just doing it for fun, that's kind of like the fun party trick is to be able to taste it. Yeah, I know I get the creaminess or whatever people are, co-coumine or whatever. So one another anecdote, I'm full of stories today. So specifically, my favorite professor in Cal Poly is a sensory scientist and she was like saying, "Oh, why aren't you my grass dude?" I was like, "I really hate sensory." Everything about setting up a sensory test and just about like, because you have to do the hard work as a sensory like grass dude. You have to like make the sound because you have to make the little three digit number. You have to put them on the little cups and you have to like schedule everything like that. No, I don't want it. I'd rather make cool products and then talk to a sensory scientist doing that. Yeah, no. For the first two years at Davis, doing sensory work just to see, because it sounded fun, right? You're going to just plug in numbers and everything. But once I was actually doing it, first year was like being the panelists and understanding that side. And then the second year was being in the opposite actually setting up all the tests and everything. I feel like at some point I was like falling asleep just looking at the data. And then at other points, I had to translate literally every single word that was said from a focus group. This is before the technology of trying to spread everything. And I was just crying. I have finals I need to study for. And I had to figure out how to describe all this. That turned me away from sensory. Yeah. I mean, some people just love the end goal of great data. I think that is the benefit of sensory science, right? You really get a very satisfying feeling when you get this angle of great data. That's just not me. Anyway, I appreciate all the sensory scientists out there. Don't get me wrong. I'm like, you're as hard as you are amazing. You are necessary. So keep up the good work. And it helped me. I make you draw easier one day. Yes, hopefully. All right. Well, next week or next episode we're going to talk about shelf life fun stuff in there. That's also another winning game. But I hope everyone has a great rest of their week day or whenever you listen to this podcast. And take care. Stay tuned. And we're both the independent food consultants who've been through the food and we're both food. No, wow. What am I doing today? Got to pitch our serves. That's a guy that put my up. All right.