Ep. 2 Formulating and How to Find and Work with Food Scientists

Listen to "Ep. 2 Formulating and How to Find and Work with Food Scientists" on Spreaker.


In Episode 2 of Feast and Fortune, the Kevins, Co-founders of immi, share the learnings behind the tactical process they took to formulate early versions of their food product without any food science or culinary backgrounds, and the process behind finding and working with a food scientist PhD and culinary chef.

They cover various topics including: the spreadsheet method they used to run different formulation experiments and product versions, why they pivoted towards a new product direction, and the right way to discover and work with the right food scientists for your specific product.



KLee: Hey Everyone. Welcome back to Feast and Fortune Episode Two.

[00:00:05] KChan, what are we going to be talking about today?

[00:00:08] KChan: On this episode, we're going to be talking about how we began formulating our noodles as well as the process we use to formulate. And lastly, how we went about finding a food scientist to help take us all the way to the end.

[00:00:21] KLee: Awesome. So KChan, last episode we talked about how we were kicking off the process to formulate our low carb noodle. Can you talk a little bit about the first type of noodle that we tried to make?

[00:00:36] KChan: Yeah, so in the process of figuring out what we wanted to make, the first thing we did was obviously go to Google and search low carb noodle.

[00:00:45] And the first thing that turned up was this product called Shirataki noodle. So for those of you who aren't familiar with shirataki noodles, they're pretty commonly sold in most US markets, and it'll be under a brand called miracle noodle. It has this jelloish composition, which is, I think, a nice way of describing it.

[00:01:07] And it's often found in this plastic bag floating with water. Well KLee, what are your impressions of shirataki?

[00:01:16] KLee: Yeah, I think.. so I grew up eating a lot of Asian noodles. And Korean glass noodles are somewhat similar, right? So they're not the same chewy, gluten, like texture that, you know, a ramen noodle might have.

[00:01:29] But the thing that was the weirdest for me was when you opened the bag, it just has this really nasty kind of sour smell. And can you maybe talk a little bit about why it has that KChan? It has that.

[00:01:41] KChan: So, so shirataki is unique in that you actually use pickling lime to help gelatinize this liquid and mixture.

[00:01:50] There's a chemical reaction with the actual shirataki root, which is a Japanese yam. And when you add pickling lime to it, it sort of hardens or firms up, in order to create something that has a sensation of noodles. So with shirataki. we thought, somewhat naively that we could actually eliminate a bunch of those, those bad qualities.

[00:02:12] And make it less smelly, improve the texture. And the great thing about shirataki is that it was basically a blank slate nutritionally, it has zero caloies, it's almost all fiber, and the remainder is basically water and so KLee actually went about helping us figure out how do we go about working with a substance. So can you tell us a little bit about what those first steps were?

[00:02:37] KLee: Yeah, so it sounds super scientific, but the first thing we did was go to YouTube. We literally typed in how to make a shirataki noodle in YouTube, and there's a, there's like a bunch of videos that'll pop up. I think the one we follow the most was this Japanese lady who was making shirataki noodles at home, and she had laid out all these ingredients.

[00:02:58] It's actually, it looks simple, or we thought it was simple at first. You have to buy this type of a yam root powder or glucomannan powder, and then you effectively just mix it with water, let it rest, microwave it, and then mix in this pickling lime solution. And then you extrude it into boiling water.

[00:03:16] And that's how you get, shirataki noodles. So we thought, okay, this looks simple enough. It didn't look like you needed like a chemistry lab or anything. And I think we can buy some of these ingredients off the shelf. So. I think the first time we did this, I went over to your place and I had bought some glucomannan powder from Amazon and we like threw it into the water and we followed the instructions, literally like, we were like pausing the video, doing it, and it just, the, the glucomannan was not forming.

[00:03:47] So like, it wasn't settling at all. And I think, what did you do? You had to go down.

[00:03:52] KChan: So we actually ended up going to a pharmacy and glucomannan is actually sold as a dietary supplement. It because it's so rich in fiber, it helps you pass stools much more easily. So we bought a, we just bought a jar of powdered glucomannan from a pharmacy and we tried using that in the recipe.

[00:04:13] So it should have probably been a red flag to begin with when there is a recipe that you're using, which involves a microwave. But we, we didn't know, right? We were just getting started, so we thought, Hey, like, let's try it. Let's just see what works. And ultimately the shirataki performed, the way shirataki performs. So why don't you tell them a little bit about how you felt about the shirataki talky noodles?

[00:04:39] KLee: Yeah, so we I mean, we actually successfully made this shirataki noodle, and we probably spent a good two to three months actually going down this path where we started to tweak the formulation.

[00:04:53] We started to add in our own ingredients. We were, I mean, I as KChan mentioned earlier, it's a low carb, low calorie and noodle, but it doesn't have that protein. And so we were trying to add in our own plant based proteins to increase the macro nutritional profile, but every time we made this shirataki noodle, we just weren't that excited about it.

[00:05:15] It's, it's hard to say, but in episode one, we talked about how you really need to be passionate about passionate about the food that you're making because you're going to be doing yeah, 200 or 300 iterations of this before you get to your final formulation.

[00:05:29] And every time we made a batch, I think we would just stare at each other and be like, so who wants to eat this one? It wasn't like we were eagerly both, you know, chomping down in the noodles, getting really excited about it. And there was one specific day where I think we were both just like sitting on my couch and we just came to this realization together where we were like, look, we're not really evangelizing this because we're not really fans of it ourselves.

[00:05:53] We're not eating this for breakfast, lunch, and dinner. It was just this thing we would make and then throw in the fridge and then throughout the next day, so KChan, what, you know, how did we move forward from there?

[00:06:05] KChan: Yeah, so. Funny enough, when we, when we initially meet this, we, we got super excited cause we thought we had solved this problem and we were going to productionize or commercializes noodle immediately. But that bubble quickly burst when we gave it to other people to try.

[00:06:22] And there they were like, this tastes like jellyfish and this smells funny. So we did the thing that we did previously, which was, we went back on YouTube actually. But this time we actually had to do much more research on how do you create a low carb noodle, which still has a characteristic taste and bite that most people are familiar with.

[00:06:44] So something similar to pasta. And we went deep on Stack Overflow, Reddit blogs. We asked a bunch of questions to people in these communities just to figure out, "Hey, what have you experimented with? What can we try? Let us know. Given our lack of culinary experience, what we can do to move forward from here."

[00:07:05] KLee: Great. Did you, did you say Stack Overflow?

[00:07:08] KChan: There's a, there's, engineers, there's a, I'm sorry, it's not Stack Overflow. There's a similar cooking recipe version of Stack Overflow. So it's part of the umbrella community. So, no, not that. It's like nothing coding again,

[00:07:21] KLee: So you didn't code this into existence.

[00:07:24] KChan: Cool. So cool. Uh, KChan. Talk us through what you learned from visiting all these websites and where did you take us next?

[00:07:33] Yeah, basically we had to find a new binder to replace shirataki, which was giving the, the cohesiveness and, gelling capability of, of our new, our, of our noodles originally.

[00:07:46] And people basically gave us a bunch of recommendations for different things that we could try in order to stabilize the noodles and the next step was actually creating a spreadsheet.

[00:08:00] Just that was the easiest way to figure out what are the composition of ingredients that would result in the macronutrient profile that we wanted.

[00:08:09] Shirataki was special because it was a blank slate. Again, it was a zero calories, practically zero on everything besides fiber. Given that we're switching directions and not using shirataki, it means it meant that we had to be a lot more careful about what ingredients we were incorporating.

[00:08:26] So we had an idea of what we wanted macro nutritionally, and we set those parameters up front. So from there we then started pulling in a bunch of different types of ingredients. And slotting them, and it was kind of like a puzzle.

[00:08:39] You add ingredient A, ingredient B, ingredient C, and you look at the total of carbs and you see does that surpass what the allowable amount that you have, and then you switch it back up again.

[00:08:51] KLee: Actually KChan. I want to dig into that because I think there's probably a lot of people listening who hear like, "Oh, tally up a spreadsheet" and it's probably really nebulous to them. Can you help break down just a little bit more for our audience? Like, how did you even create that spreadsheet to begin with?

[00:09:05] What did the structure look like as we were experimenting, whether it was with shirataki or with this new direction?

[00:09:10] KChan: Yeah, so, so the basics of it is you first create a sheet with just the list of all the nutritional profile of the noodles. So things like carbs, dietary fiber, protein, fat, calories, sodium, potassium, all that jazz that you normally see on the back of a food box in the nutritional label. So under the nutrition facts section.

[00:09:34] The next piece is actually creating a lookup, which says, you know, input ingredient name, and then it goes and it fetches from your ingredient database and says, okay, if you are using one gram of, I'm just making something else, something up right now. Citric acid.

[00:09:53] It will then go and create a calculation saying, here's what one gram of citric acid adds in terms of sodium to, to your formulation. So you can imagine just a simple spreadsheet that pulls all that information based on the ingredients you input, as well as the volume of ingredients that you put in there.

[00:10:11] KLee: So breaking it down one more time, KChan effectively created one. It was a single spreadsheet.

[00:10:16] One tab was the ingredients database, and it just listed all of our ingredients along with each column for, you know, like calories or fat proteins, so on and so forth for each of those ingredients.

[00:10:27] And then we had a main spreadsheet, which was the formulations spreadsheet.

[00:10:32] And the formulations spreadsheet had like experiment one. And it would pull the five or six ingredients that we wanted to experiment in that formulation. And because he was using like a lookup, you can look that up on your own time. It would pull in all of the macro nutritional profiles for that ingredient, and then it would give us a total.

[00:10:48] So for that formulation number one, what was like the total nutritional profile? What did that look like?

[00:10:54] KChan: So that was great because what it effectively does for you is, or what it did for us was that it created some guardrails for possible recipes we could use because we already knew that using wheat flour, for example, would break our macronutrient parameters.

[00:11:12] It was just way too much carbs, for example, and then that enabled us to quickly pull together 10 20 30 40 all the way up to a hundred different formulations that we wanted to test. And then we could very strategically test out specific parameters or sets of ingredients that would effectively form into a dough flour.

[00:11:38] And then we would just taste and test and see what worked or what didn't, and then create new formulations based on that feedback. I'd like to say it was pretty scientific, but in reality it most definitely was, was not.

[00:11:52] KLee: Yeah. I think, you know, it's funny, a lot of people often ask like, well, did your previous jobs have any skillsets that translated and KChan and I, you know, we talked about this on episode one, we came from the tech industry.

[00:12:03] We also spent some time in the finance industry before that. And so we are naturally like somewhat quantitative and we have played around a lot in Excel before doing models. So it just felt like the natural next step here was to build a spreadsheet and treat each formulation like it was its own model in a way with inputs and outputs and

[00:12:23] KChan: Yeah, but the truth is that the entire process was, it was incredibly grindy.

[00:12:28] And it's one of those things that you just have to do because you just need to go through hundreds and hundreds and hundreds of formulations in order to land at something that is pleasant to taste to smell, texture to smell. And it, it wasn't sexy at all. It was really just us sitting in a kitchen for hours on end, mixing bowls of noodles, waiting for it to dry, forming it by hands.

[00:12:57] And. That was, that was a grind. But it was, it was super important because it put us in a position where we felt like we intimately knew how to make noodles and we, we knew what the dough should feel like. And it's gotten to a point now where KLee and I can actually look at dough in a mixer like a stand mixer as its spinning. We look at each other and we can say uh yeah, that's going to suck.

[00:13:22] KLee: Okay. Yeah. That's just sheer brute force. So KChan, probably makes this sound a little bit more impressive than we really were, and it's, it's easier connecting the dots, moving backwards then forwards.

[00:13:35] You know, in the beginning when we were doing all of these experiments, we didn't actually follow that much of a rigid process while we were tweaking from like one experiment to another.

[00:13:45] For example, I used to make the mistake of trying to change two variables at once. So I would swap out two ingredients, or I would swap out one ingredient and then also changed the method.

[00:13:55] So I might, you know, microwave something at a few more seconds than I did last time. And I think every time you change more than one variable, you're going to result in a completely different result that isn't really replicable, I guess. And so KChan, how did we actually even learn how to run the right process of experimentation?

[00:14:17] Where did that come from?

[00:14:19] KChan: I think that's a, that's a great transition to the process of how we ended up finding food scientists. So it was actually our food science advisor who we showed our methodology too, and he looked at us and he said, "What do you guys, what are you guys doing? You guys are changing like three, four, five variables at a time. You don't have a standard operating procedure," and he really gave us a framework for how to do this.

[00:14:44] He said, "Write down your operating procedure and only change one variable at a time because then you can isolate what exactly is going on and how that's directly impacting your product."

[00:14:56] KLee: Yeah, it's funny, we have so many stories of our experiences working with our food science and culinary advisors over the past year. KChan, can you walk us through what was the process or actually at what point did we decide like, hey, we need a food scientist and like why did we come to that conclusion?

[00:15:17] KChan: So we basically got to a point and got to a point in our formulation where we would add new ingredients and it doesn't necessarily change anything.

[00:15:29] So we would make something, we didn't like it. We would change a variable, make something. We didn't like it, but we couldn't tell the difference in terms of whether or not it was improving the product or making it worse.

[00:15:41] The differences at this point were so subtle because we were changing very, very, minute quantities of product, and at that point we felt like we had gotten to 80% of the formulation, completeness by formulation, completeness, I meant that we liked it enough, but we wanted it to be at 95% to 100% in terms of how much we liked this product. And we knew we had to bring in the big, the big guns, and that would involve a professional who could help us take, help us go all the way.

[00:16:13] KLee: So from there, I guess we'll talk a little bit about how we even found our food scientist advisor to begin with because it can be a very cumbersome process.

[00:16:23] And the one thing that KChan and I learned through this process is that you can't just assume that a food scientist knows how to work on the type of product that you're specializing in.

[00:16:36] It's not the same as in the tech industry where you might have like a front end engineer or a backend engineer and they might know, you know, C++ or Java in the, in the food industry. Actually, I guess that's a little bit similar in the food industry. Each food scientist does have a specialty, and if you contact us a food scientist who knows baked goods or who knows chips, there's not a guarantee that they're going to know how to make noodles.

[00:17:02] So in the beginning, we, you know, even if we tried to just cold outreach to people on LinkedIn, it's not like that food scientist. There's no guarantee that food scientists is going to know exactly what you're doing. And we kind of got lucky in our, in our search.

[00:17:15] For us. One of our friends was working at a food company, one that was growing really, really quickly.

[00:17:23] And we actually reached out to him because we had hit this local maximum kitchen and KChan talked about this before. We just didn't know how to proceed. We had hit the theoretical limits of our own knowledge. And so I had reached out to this buddy and I said, Hey man, can you do us a favor? I know you guys have a head chef.

[00:17:41] They are also a CPG company. And we said, can we talk to your head chef and just get some advice from him? And my friend was kind enough to introduce us to that head chef. And once the head chef found out what we were doing, he actually ended up looping in their food science PhD.

[00:17:57] So they had two people on their team working on the, on their product. And both of them started to just do regular calls with us because they were actually personally just interested in the product we were making. KChan. What, you know, what did we do from there?

[00:18:10] KChan: So the first thing we actually did was hop on a call with them. And to be, honest that that call actually went pretty terribly because we had never had a conversation with a food scientist before. We didn't actually know what was a productive conversation. So we told them that we wanted to make noodles.

[00:18:29] KLee: We want a specific type

[00:18:30] KChan: of noodle, that specific type of noodle, this is ingredient that we want to use at the time. It was actually still shirataki and we said, Hey, like can you make it for us?

[00:18:42] Typical

[00:18:43] KLee: like, you know, business person talking to like an engineer type conversation. Like it's like exactly the type of people that like engineers hate the most. In this case, it was a food scientist.

[00:18:53] KChan: And as KLee had mentioned. You know, different food scientists have different specialties. Their specialty was not so much noodles, so it was a sort of a roundabout conversation where they said, yes, it's possible, and we were waiting for them to provide us with instructions on how to do it.

[00:19:17] They ended the call. Then we emailed them, hopped on another call, asked them the same question, can you make this? What is the instructions to make this? They said it's possible. And then they ended the call and it kinda went like this for a while until KLee said, screw it. Like, let's figure out how to move the ball forward.

[00:19:36] And so what did, what did we do from there KLee?

[00:19:38] KLee: Well, we basically told them, hey, we want to go visit you guys in person and we're going to come to the table with basically the offer KChan and I talked and we realized that we really needed to bring on someone more formally,  as an advisor to the company, or at least someone who owned some equity stake.

[00:19:57] And, you know, have an incentive to help out. So we had told these two folks, Hey, we're going to go visit you. Be prepared to have a conversation where we're probably going to negotiate some equity terms. And by then, obviously we had been talking I think over a month, so there was some rapport.

[00:20:15] It wasn't just like we came in there like guns blazing type conversation. It was obviously a lot friendlier than that, but we honestly had no idea what to expect because we had never met these guys in person.

[00:20:25] KChan: We were basically convinced that the reason why they never gave us instructions on how to make the noodle was because they wanted ownership in the company. So our conclusion was to give them a bit of ownership to finally unlock that recipe.

[00:20:38] KLee: Yeah. We just thought it was in their heads and they like, they were like, Oh, well, we're not going to give it to you unless, yeah. Like KChan said so far from the truth. But, so this is what happened.

[00:20:47] We ended up flying to their hometown, met with both of them. And honestly, it was actually, it was a very surprising experience, I've got to say, because again, we, this is our first time interacting with food scientists.

[00:21:03] It was a food scientist, PhD and a chef, and it was our first time meeting them in person and they were a lot cooler than we expected, I mean, they were really fun actually.

[00:21:13] It was, it was pretty awesome. And I can't guarantee every conversation or person you work with is going to be like this, but it ended up being just a really incredible first experience where we got there. We busted out this luggage suitcase full of our own ingredients, and we basically showed them the process by which we had been creating our noodles.

[00:21:33] And then they showed us a little bit around how they might approach it. And they actually already had met together and tried to come up with their own formulation. Fortunately, ours was actually much better, so they knew we weren't joking around.

[00:21:47] And I think that's important, right? Like when you show up, you definitely want to, ideally, you want to see if you can come up with your own formulation and come to the table with at least a base formulation set in place.

[00:21:58] KChan: Yeah. Yeah. So, so ultimately we ended up signing them up as advisors, as well as investors into the business because they were incredible people and super helpful from a processes standpoint, helping us understand what it takes to commercialize recipe, but they weren't necessarily noodle experts.

[00:22:17] And. All of this is to say that the way KLee and I had approached the food science hiring process was not the most ideal process. We kinda got lucky and it worked out because we were able to bring on board some very awesome people. But if we were to do this over again, of course we'd still bring them on board, but do your diligence.

[00:22:36] It's so important to find someone who specializes in the specific type of product that you're looking to make. It wouldn't make sense to hire a beverage person to come formulate a pasta for you. Or a sauce person to formulate a cracker for you.

[00:22:51] It, it requires a high level of specialization and there's so much nuance in between this that it's really important to figure out, Hey, has this person worked on this product before? Because that's going to give you the most bang for your buck when you're looking to solve for a specific problem.

[00:23:07] KLee: Definitely. And by the way, these, these two people that we mentioned, we're really good friends with them and they've helped so much in this process, even if they weren't specialists in exactly what we were doing.

[00:23:19] They did help a lot with our process. And as KChan mentioned before, they professionalized us. So that first time we met and they watched us produce our noodle, we realized later too, because the food science PhD actually told us, he said, "look, I wouldn't have worked with you guys unless I watched you guys work and I saw how you guys are problem solving. I saw that you came to the table with this spreadsheet prepared, how you iterated through formulations as I gave advice."

[00:23:43] And I think, he was really looking to see if we were serious and along the process he, that's where he gave us advice like, "Hey, why are you changing more than one variable at a time? Or why are you changing this process midway? Like it seems like this is working. Or he'd even say like, it looks like you have seven ingredients here. What is the functional benefit of that ingredient?"

[00:24:02] And when we couldn't explain it, you'd say, well, you need to take it out. And try the formulation without it and then try it again to see exactly what the functional benefit is. And if it's not necessary, just remove it. Because it's the tendency of people who aren't familiar with food is to try to stick everything into something and see what works. But that's not necessarily the best approach.

[00:24:18] KChan: Yeah. So while we tactically didn't get a food scientist who was going to support on making noodles, well, we did end up getting, was a great mentor along the way, which we definitely can't stress enough is so important because if you're moving into the food industry for the first time, there's so much nuance that is difficult to comprehend.

[00:24:38] And for us moving from tech, the question we get a lot is, have you guys done food before? And the answer is always no. And people are surprised at how much progress we've made.

[00:24:47] And a lot of that progress is just due to the professionalization of having these mentors and advisors help steer us in the right direction.

[00:24:56] KLee: I think a common follow up question here would be, well, how did you guys even structure this equity incentive? Obviously that's, you know, it's a, it's a conversation that we could spend another session on and probably, not appropriate here.

[00:25:10] But what we will say is we did get pretty creative, I think in the early days when we didn't, both parties were still Starting to work with each other. We even tried, an arrangement where it was, okay, let's do a flat equity, percentage now with the ability to earn an extra percentage when we hit certain milestones, for example, when we get the first formulation up or when we get a product to market.

[00:25:36] And so we considered basically trenching the equity percentages. These were done over a standard four year vesting one year cliff. So very similar to how you might earn equity in any other company.

[00:25:49] And this is obviously something you just need to negotiate with your food scientists. And it really just depends on what stage you are, how much you value the scientists or the chefs that you're working with, and how much time they're willing to spend with you.

[00:26:01] The two folks that we brought on board have spent a good amount of time with us, even though they've been busy with their own companies. You know, they try to be as helpful as possible. We'll try to schedule regular calls whenever we can. And so, we've been very happy giving them a, a large percentage of equity for that reason.

[00:26:22] KChan: So yeah, we ended up bringing them on board. Eventually they guided us towards finding a specialized food scientist who had worked in noodles.

[00:26:33] They helped us with the vetting process and we just took a standard approach afterwards. So while it was a bit roundabout for us to get to a food scientist, it was still a great experience for us that, that we definitely don't regret.

[00:26:47] KLee: So we leveled up, you could say.

[00:26:49] KChan: With that said, we had a team of food scientists supporting us. We had our ingredients and our formulation, at least a formulation that we were happy with, it wasn't necessarily a commercializable formulation, and we can go into this distinction between a kitchen formulation and a commercial formulation in a future episode.

[00:27:12] But the next step for us was to actually find a manufacturer who could help us take this recipe and commercialize it and make it ready for market.

[00:27:22] KLee: So in our next episode, we're going to talk a little bit about our thinking behind how we actually decided to commercialize this. Some people will tell you, make this at home and then bring it to a commercial kitchen, and then later on consider a co-packer.

[00:27:37] And we'll talk a little bit about our thought process of jumping straight to that co-packer search process.

[00:27:44] Thanks everyone for joining us for this episode of Feast and Fortune brought to you by immi.