Andrew Connor: We’re starting to have meaningful sample sizes for a lot of foods. It’s interesting for us to see where our, I guess, is our collective zeitgeist just completely mistaken. I’m excited to think longterm. What will it look like when a lot of people are equipped with CGMs and metabolic feedback loops to bring more awareness to a lot of these foods? Because if you were to put a lineup in front of me of regular milk, almond milk, cashew milk, oat milk, and pretty much be defenseless and not really know that one of those is metabolic jet fuel.
Ben Grynol: I’m Ben Grynol, part of the early startup team here at Levels. We’re building tech that helps people to understand their metabolic health. This is your front row seat to everything we do. This is A Whole New Level. When watching a product get built from the outside, it’s pretty easy to see everything that appears on the front end. That being the app, that being the data you see in realtime. But it’s not always easy to see things like the infrastructure. The engineering time that actually goes into building the product. Creating one that’s stable, one that can process data, and one that’s scalable to help as many people as possible. For Andrew Connor, Co-Founder and Head of Engineering at Levels, well, he has a pretty different outlook on the way that he sees a product.
Ben Grynol: He and the engineering team, although small and scrappy at the stage we’re at, they have a different outlook on what needs to be done to ensure that the product works efficiently for members. When thinking about scale, Andrew considers things like, how will the product be able to process one billion data points per month? A number that seems quite significant, but one that’s not too far off in the distant future of Levels. On a more granular level, he thinks about the micro data points. Those being the specific foods or inputs that can be modeled to help inflict positive behavior change and education for the greater population and world.
Ben Grynol: One of the interesting things is, around the end of February and the start of March, we started to see some interesting infrastructure challenges. We’re still relatively small in data. We have roughly 1000 to 1500 monthly active users that are members, that are actually using the product day to day. We started to see some legs in product. I think that was just a byproduct of the infrastructure, the scrappy rails that have been laid down. But really it comes down to the data points. We are pulling in a lot of data that needs to be processed.
Andrew Connor: Yeah, it’s interesting because a lot of consumer startups, something like 1500 customers producing data in a month is nothing. At the same time, the devices we’re using are emitting a lot of data continually. Right now, this is April 2021. We have roughly 10 million data points per month that we’re taking in. This is glucose, heart rate measurements, sleep measurements, things like that. We are preparing. In many ways, we want to focus on product quality. Knowing the endpoint is that we’re going to be scaling up, but a lot of our engineering resources are focused on, how can we affect behavioral change in the product as effectively as possible?
Andrew Connor: The endpoint we’re going to is when we launch, we want to be prepared for two orders of magnitude more. So, a billion data points a month. That is meaningful for a small company both to keep costs down, but also this data is not just data that’s sitting in a database somewhere. We’re processing it. We’ve been focusing on building efficient data pipelines where data can come in from a lot of different sources. This is obviously the glucose monitors themselves. This is activity trackers, sleep trackers, things like that. We want to be able to process the data to drive insights. Some of these are quite intensive to calculate. Some of them are pretty easy to calculate. That’s been where a lot of our engineering resources have gone to, is thinking short term, how do we deliver the best product experience? Because if we don’t do that, then we don’t succeed as a company. But longterm knowing that if we’re successful, there’s also going to be a lot of associative data.
Ben Grynol: Yeah. I mean, that’s the funny thing, right? When the startup is in the crappy building phase, you never want to build things too perfect. But you also want to make sure that you’re building them so that they’ve got enough resilience to scale in a loose way. It’s like if you know that you’re going to a point where there are a billion data points, and that’s not too far off in the future, you’re like, do we build for this now? Or what’s the opportunity cost of not building for that now because you want to focus on other engineering problems? The interesting thing is our engineering team, given the number of data points, given the amount of data alone, is small, is quite small.
Andrew Connor: Yes, yeah.
Ben Grynol: I think we have one person who is our “data scientist,” [Jin Lu 00:05:48].
Andrew Connor: And part-time, right? She’s doing a lot of other stuff as well, so yeah. It’s been an intentional strategy of ours. Level is, I’m sure you’re familiar with red ocean, blue ocean startups. In red ocean, a lot of it is about execution. It’s about meeting customer demand. It’s very well-defined. Blue ocean is more of an exploration. Levels is very much that. We know there’s something here. We see it in the customers’ stories. We hear the people that have changed their lives. Finally they’ve been able to build feedback loops in their habits that allows them to lose weight effectively, or have steady energy, or increase athletic performance, or anything like that.
Andrew Connor: We know that we’re resonating with people, but our end goal is to affect behavioral change for a very wide population. It’s an undefined problem, as in it is not defined precisely what sort of product does that. You can imagine there could be a lot of different variants of levels that did different things. Back to what you said about a small engineering team, is we wanted to look for generalists. So, people that have a breadth of experience and have worked on a lot of different types of projects. Can move between project to project very efficiently and effectively. That’s something they enjoy. And also keep just a nimble team. Let’s cut out the bureaucracy and make it as lean as possible to be able to iterate very quickly. See what works, throw out things that don’t. It’s very much been an intentional strategy of ours from the beginning.
Ben Grynol: Yeah. I mean, it’s incredible how much the engineering team ships. Constantly, constantly shipping. It’s hard too because you’ve got engineers, as you’ve mentioned, bouncing between working on backend. Now it’s getting a little bit clear as far as the path where some team members can just focus on mobile, right?
Andrew Connor: Yeah.
Ben Grynol: That’s the way. When it started out, it was all over the map. It was exactly what you said. You’re a generalist, so you’re working on every part of the stack because it was necessary. The interesting thing is, so why is data science such an important part of what we’re building and such a hard problem to solve for as it pertains to metabolic health? Because you alluded to it with, you’ve got these inputs, right? We’ve got sleep. We’ve got certain inputs that we know affect metabolic health and metabolic function. Sleep, stress, diet, hormonal differences. That plays into it, a very big part of it.
Ben Grynol: Then even there’s one thing that we haven’t considered, but I started thinking about it more based on a blog post that [Haney 00:08:30] just jammed out on brown fat, cold conditions. When I started thinking about geography I’m like, man, I wonder if people in cold climates, self-serving because I’m from a cold climate, but if people from Northern parts of let’s say Scandinavia. If they’ve got different metabolic function than somebody from a warmer climate. I don’t know. Anyway, there’s a whole bunch of data points that we have to consider, so it’s interesting to hear your perspective on that.
Andrew Connor: Yeah, real quick tangent. Speaking of metabolic health and cold, I saw an interesting paper about obesity and elevation. They tried to control for all the obvious things. Population density and stuff like that. There’s this inverse correlation. Basically the higher elevation you are, the lower the obesity is. It’s fascinating. It’s not entirely clear why. In fact, I found a followup paper that was basically saying the obvious explanations of why are probably not true either. It’s not fully understood. As you mentioned, there are a lot of variables here. I guess I’ll zoom out a little bit, and then I’ll directly answer your question.
Andrew Connor: In my mind, for Levels to be successful, we have two theses we’re building the company on. The first one is that people have no feedback mechanism for metabolic health. It’s this nebulous thing. Chronic metabolic disease pops up over decades. It’s something that just seems to appear out of nowhere, but we know that metabolic health is not binary. It’s not like you’re metabolically healthy or you’re metabolically sick. It’s a spectrum, just like physical fitness or any other kind of fitness. That’s the first one, is that there’s no feedback mechanism.
Andrew Connor: Particularly, continuous glucose monitors provide an insight into that. Metabolic health is not health. It’s a subset of it. Continuous glucose monitors are not fully metabolic health. But now all of a sudden you can get an insight into what is happening realtime, so how to build that feedback loop. The second theses is that the raw data is not enough. We experiment with this early on. Helping people to get access to CGMs and helping them understand their data. Many people are just lost. It was overwhelming. Their glucose goes up, their glucose goes down. What’s normal? What’s not normal? It wasn’t clear to anyone. Besides the most extreme bio hackers, we found that people are really seeking for personalized insights. They want to know how their existing habits are interacting and how they’re able to improve and reach their goals. Whether it’s lose weight, optimize athletic performance, anything like that.
Andrew Connor: It became very clear to us that these personalized insights and the data that we’re collecting will be core to the business. This is the true value of Levels. Levels is not strictly about hardware. It’s sitting on top of a platform that didn’t exist previously. It’s being able to actually distill these kind of responses that people have metabolically to things that make sense. So, education, but also showing correlations between things. I think you called out several things like sleep, stress, hormones. There’s also gut microbiome. There’s a lot of variables here. One of the early reasons that data science for metabolic health is hard is there’s a lot of variability between people. What we would love, the ideal product, is you eat a slice of pizza. Then an hour later, give you a report of exactly what that piece of pizza did for you generically.
Andrew Connor: Pizza is good for you or whatever. Even crazier would be like, oh, you’re sensitive to wheat because flower, and that made the pizza. It turns out it’s just far more complicated than that. I had a local baker near me. Whatever flour he was using had very little metabolic impact on me. But I’ve also had bread from the regular grocery store that has a large metabolic impact to me. That’s really hard. I can’t overly simplify things into bread is good for me, bread is bad for me. It ends up being a lot more complicated. The levels of activity that you have undergone or the amount of fiber you’ve eaten in the past 24 hours impact the response you have to foods.
Andrew Connor: One of our early questions was, is it all noise? Is there signal here? It turns out there is. There’s very interesting things that we can get from these glucose monitors, especially in conjunction with other sensors. We’re starting to do interesting stuff with sleep and step counter, but it’s not incredibly trivial. We have to be careful. We have to be careful understanding the modes of error for these devices, and the kind of variability that we should expect. We have to be careful about the context someone is in. If they had a very poor night of sleep, they are most likely going to be more insulin resistant. They’re going to see much higher glucose spikes the next day. That’s something that we have to take into account in our product if we’re going to be able to come up with these very meaningful, personalized insights.
Ben Grynol: Yeah. The idea is that you want to give people nudges, positive nudges, in the right direction. I think one thing that everybody is pretty anchored on in a good way is that we never want to be prescriptive.
Andrew Connor: Yes.
Ben Grynol: We want to be objective, but we don’t want to be prescriptive. It’s like, don’t eat the pizza. That’s not our job. Our job is to say, “This is what that does to you from a personal perspective.” Because that might not do that to Andrea or Ben or whomever because of all the other factors. That being, how much fiber did you have in the last 24 hours? Did you get three hours or did you get nine hours of sleep? All of these other things. Then where we can be objective is, objectively if you get less sleep, you are going to have a worse response or a poorer response metabolically to eating that pizza. That’s where it gets really tough.
Andrew Connor: Yeah. I think the personalized-ness of this is where we are a little bit at the forefront of a lot of wellness. I’m sure we’ve all been very familiar with glycemic index and glycemic load. It’s actually fascinating too to deep dive into how these were calculated. What is the history of these things? They are population averages. They are from decades ago. The study designs are somewhat dubious. They don’t really demonstrate the breadth of response, the standard deviation of response to things.
Andrew Connor: They act like an apple is an apple for everyone. There’s a really good book that I read at the beginning of Levels called The End of Average by Todd Rose. I’m sure others have talked about this book. But he basically talks about how no one is average on most statistical measures. Whether it’s intelligence or even body size, you’d think that there would be an average body. It turns out that no one is that. It’s kind of subtle, and it’s because there are many different variables at play here. Your arm length, your leg length, your torso versus your legs. All these different things. One of I guess the motivating things for us early on was a recognition that people’s response to the same foods can be really different.
Andrew Connor: I remember when I first got a CGM with my wife, we would almost make a game of it and play with different foods. It’s fascinating. My response to rice consistently, the area under the curve for my response to rice is about twice my wife’s. Her response to potatoes is about twice mine. We’ve done this again and again. I can eat a large baked potato and stay under 100 milligrams per deciliter. She’s having a huge glucose response. This is not something that you can intuit. Not by family history, not by the way you feel, unless you’ve really taken it to the extreme. It’s something that I’m hopeful that we can actually push medical science to actually also embrace the individuality of healthcare.
Ben Grynol: Yeah, because that’s the thing. Getting back to being prescriptive, if we said “Don’t eat baked potatoes,” there’s your example right there. We never want to do that, and we don’t do that with diet. A lot of people say, “Well, what should I eat?” It’s like, eat what is right for you. Nudging people in the right direction so that they can get the insight that’s not just a matter of, you had a large response to this. Let’s say it was a baked potato. You had a large response. It’s, what does that mean? People, when they start to get the insight behind it they go, “Oh, that’s why I feel bad in the afternoon, I feel like I’m falling asleep, is because I had french fries and a burger for lunch.”
Andrew Connor: The thing that actually gets me really excited is that if we are successful, we’re able to successfully and effectively show people the levers that are available to them. Some levers are more exciting for some people and more exciting for others. For example, an experiment we’re doing internally in the company right now is drinking a prescribed amount of glucose and then doing different activities or sitting still. Early data is fascinating. It turns out that a light walk, and this is not like a Peloton hard bike ride or a run or anything, but a light walk has a pretty significant impact to the glucose variability.
Andrew Connor: We know that glucose variability is important because it can increase insulin resistance. Knowing that in your back pocket is very enabling because being told you can’t eat something is difficult. It’s difficult for me. I think it’s difficult for a lot of people. We know this from looking at diet compliance studies that restrictive diets are difficult. But I think as we equip people with tools to be able to mitigate certain decisions, or just tools that they can assimilate into their habits that are easy for them. Whether it’s incorporating intermittent fasting, whether it’s more fiber or more mixed meals in their diets. Replacing certain foods, so I eat less rice now.
Andrew Connor: That was an easy change. I don’t eat zero rice. But it’s something that being aware of is pretty powerful. For me, the really exciting thing is seeing people understand, with the feedback loop, understand what levers they can tug on to reach whatever goals they have without things being extreme, all the way one way or all the way another. That’s the answer why we don’t as a company advocate for any specific diet. I think some people in the company really like keto diets. Kasey, our Chief Medical Officer, loves plant based diets. I think all of them can work, especially with these feedback loops.
Ben Grynol: Let’s push on the one factor of wanting to influence positive behavior change, but knowing we’ve got this dichotomy that we have to deal with. It’s positive behavior change versus creating negative habits, or influencing these new habits to surface where we heard sometimes that people will start to change their habits so drastically. This is coming from your point about being extreme, right?
Andrew Connor: Yeah.
Ben Grynol: We don’t want things to be extreme, but it’s natural when things feel… you’ve got feedback. They feel like a bit of a game almost, they’re gamified. We do hear from people that they’re like, “Oh man. Now I pay attention to every single input I have. I won’t have a grain of rice.” It’s like, “Oh, maybe that’s okay if the grain of rice, one grain of rice, is really not helping you be metabolically healthy.” But the point isn’t depravation. The point is health. That’s a challenge that we have to figure out from a company standpoint, from a messaging, from a product standpoint, from a data science standpoint. It’s really hard.
Andrew Connor: Yeah, and it’s something that we’re thinking deeply about on our product design because it’s easy for us to be overly critical. Especially because everyone is starting out in a different place. Some people start Levels and they pretty much look amazing. They had good habits to begin with. They are still able to learn a lot, but they’re not working from inside of a… they don’t have to recover from anything. Whereas a lot of people try levels because they’re interested in feeling better or losing weight or something. They learn oh, I really need to change my life habits. It can feel really critical, like a mountain that is really hard to climb or something. It’s a pretty significant product challenge. Something that I was thinking earlier is Peloton and Strata have leader boards. If you look at it, what does it mean to be at one of the top of those leader boards? It means that you’re a pretty incredible athlete.
Andrew Connor: If we were just to rank people by glucose variability or average glucose or something like that and make a leaderboard, what does that look like? It can promote some very unhealthy behaviors because the effective metabolic health is also about balance. It’s about maintainability and things like that. That’s why there is no leaderboard in the Levels app, because we have to be very careful about the incentives we’re creating and how we frame things for people so that they can achieve whatever their goals are in whatever position they already are in.
Ben Grynol: Yeah, I’ll use myself as an example here. I know that I’m not a big consumer of sugar, nor have I been previously. I just don’t like sugar, but I do like things like blueberries or raspberries, fruit. I know that even when I eat it in the morning, even if the order is changed or it’s with high fat, high fiber things, I still get a response to it. I have to tell myself, I’m like, “Eat the fruit. It’s good for you. It’s healthy to have blueberries. Don’t eat a bucket of them, but have a proper portion.” Knowing there’s a psychological hurdle to get over, we don’t want people to go, “Oh, I’m avoiding eating that thing altogether, something that’s specifically healthy.” I can’t remember who I talked to about this before, but it’s like our goal isn’t to be like, “Drink a bottle of olive oil. That’s where you get your calories because you don’t get any response to it.” That’s not healthy.
Andrew Connor: Yeah, I mean, you’re absolutely right. We even see this with exercise. High intensity exercise can cause glucose spikes. All of the research we’ve ever found, these are not bad. It is basically glucose available for your muscles or whatever activity you’re doing. It’s another example on how naïve data processing is difficult. We automatically detect strenuous activities to mitigate scoring factors, but we actually want to do a lot more into that and actually look at how exercise intensity and glucose response connect to each other. And what can we tell about your metabolic health from how that happens? The answer is not all glucose variability is bad, or you need to be at rock bottom fast of glucose. It very much is about balance. It’s something that we want to be a slice to help people understand what feedback loops exist and what variables matter for them. But it’s not something to be taken to the extreme.
Ben Grynol: Moving forward, in the future, there will be a time that we start to look into other analytics. How does that change from an engineering perspective, from a product perspective? Right now, the billion data points that you’re talking about, that’s glucose related data. That’s not even related to, okay, now here’s one more analyte. What does it look like when we add a bunch of different analytes? How do we think about that?
Andrew Connor: Yeah, so certain analytes probably are fairly informative for metabolic health also. I very much defer to the medical experts there, but something like cortisol could be very interesting. I don’t know if you’ve had the experience of being in a stressful situation or taking a stressful meeting or something like that, and noticing your blood glucose is significantly higher than normal, and you’ve eaten nothing. That’s kind of a shocker. It’s like, oh my goodness. The cortisol actually affects me. You can imagine a continuous cortisol monitor gives us another dimension. In terms of the amount of data to process, it would not meaningfully change anything.
Andrew Connor: Longterm, we want Levels to be for everyone. We want the price point to be accessible to as many people as possible and Levels to be accessible to as many people as possible. So, we are planned for a lot of growth. I mentioned a billion. That’s kind of what we want to target for data points per month when we launch, but it’s going up from there. Bringing in new analytes or just having more growth looks roughly the same. There will also be analytes though that will just be different, like inflammation markers or something like that. We’ll give people another really interesting slice into their health in a different place, a very complementary place, than glucose or metabolic health. But the interaction between them won’t be as much. Something like ketones obviously interacts quite a lot. If someone is trying to be as fat adaptive as possible, being able to incorporate keto data points is really interesting in conjunction to glucose.
Ben Grynol: Yeah, especially with some of the research. Dom is doing, Dom [Dagostino 00:27:09], he is on our board. Not our board.
Andrew Connor: Our medical advisory board, yeah.
Ben Grynol: Dom is one of our medical advisors. That is the label for them now, our advisors. Dom is doing a lot of research around ketones and metabolic health. How does that relate to some of the things that he’s seeing in Levels? It’s really interesting because you start to realize that as your metabolism adapts to whatever diet that you choose, this becomes more empirical. This becomes objective data where you’re like oh, okay. If your body is in a state of ketosis, then you are going to react differently to the same input of the input of X to X. It’s much different than if you’re comparing it when you’re not in a state of ketosis. Even that alone, that’s a really interesting data point. It’s something that informally, we’ve had people who have been through the data already that are independently measuring their ketones. They’re comparing data.
Ben Grynol: There’s actually one person that I know that’s doing the data export and checking their own data from Levels, and comparing it to the data that they’re collecting independently. It’s interesting just to see what people are interested in when that’s a ton of friction to actually find out that data, but somebody… and granted, these are people who are very invested in data out of intrinsic interest or they’re invested in their health. But it’s interesting to see what people are doing on their own already.
Andrew Connor: Yeah, it’s fascinating. I’m also really interested in what other signal we can get from glucose outside of strictly just pairing food with glucose and looking at variability and things like that, that are under the curve. There’s a couple of interesting periods. As you sleep, you’re in a fasted state, so it’s pretty special. The holy grail of what we really want to understand is insulin resistance. Continuous insulin monitors do not exist. The molecule is quite difficult to detect in a way that the glucose molecule is not. In many ways, we are measuring glucose, but we want to understand insulin levels because insulin resistance is the core to a lot of metabolic diseases.
Andrew Connor: Periods such as more stable periods of just sleeping, or right as you’re waking up, seem to be pretty interesting. One of the things we have on the app right now and we’re considering to do some research into is the dawn effect. I don’t know if anyone else has talked about it, but the rise in glucose as you wake up in the morning or as your body prepares to wake up without any kind of mood stimulus. There’s literature that seems to connect this with insulin resistance. There’s a couple causal mechanisms that are plausible here. This is another really exciting area that us being able to collect a lot of data lets us kind of glean through it and find really interesting things that previously haven’t really been researched. A lot of metabolic research goes to study diabetes. While that’s incredibly important, most people do not have diagnosed diabetes. We want to help people prevent that as much as possible. Us being able to be on the forefront of some of this research is incredibly exciting to me.
Ben Grynol: Yeah, and we know, we just know from the research that exists, that it is possible to change your metabolic health or your metabolic function. Somebody can go from being insulin resistant, and over time with the right behavioral changes, their body will adapt. That’s the goal. The goal is that people, the vast majority of people, are on the path of being insulin resistant. Let’s just use the U.S. alone. In the U.S., there is a significant number of people. It’s 88 million Americans are pre-diabetic. 84% of those 88 million don’t know that they are pre-diabetic. There’s a large group in the population, let’s just lump it in to saying it, close to one third of the U.S. population is in a state of insulin resistance and on a poor metabolic path. The detrimental outcome is that if you continue on this path of eating highly processed foods or things high in sugar, things that are just not good for you to begin with. It leads to such great downstream costs in healthcare, lifespan. Your health span and your lifespan, everything starts to decrease. If we can inflict some change on saying, “Hey.”
Andrew Connor: Oh yeah.
Ben Grynol: “You can reverse this. You don’t need to be insulin resistant. Here is some data. Here is some insight that will help you not go down that path.” I think it’s pretty meaningful.
Andrew Connor: Oh, very much. It shocked me when I learned how few people who are pre-diabetic know it. We encounter this in Levels. We are a wellness product, but a lot of people do come to Levels thinking I just want to lose some weight, or I’ve been feeling tired lately, or something like that. And they realize oh, I really need to change some of my life habits. We’ve seen it. We’ve seen people that was the kick in the pants that they needed in order to finally change their habits. The other interesting thing for me, looking at the data we have with glucose paired with food, is our food system has very little accountability. I’ll give you a very specific example. My wife and I love to go on hikes. I used to take a Cliff bar with me all the time.
Ben Grynol: The Cliff bar.
Andrew Connor: Yeah. The interesting thing is I didn’t consider Cliff bars healthy. I’ll be very clear about that. I was not deluded into thinking they were healthy, but I guess I kind of thought they were health-ish. It’s oats and seeds and stuff like that. I mean, yes, it’s sweet, but I just never really thought about it. It’s very much branded toward activity, towards people being healthy and making good decisions. I remember I was excited because it was one of the first foods I wanted to test. I already learned I was very sensitive to rice.
Andrew Connor: Well, in many of the Cliff bars, the sweetener is rice syrup. I’m also sensitive to oats. Actually, a ton of people are. That’s actually the most consistent shocker for a lot of people when they get on Levels, is they’ve been eating oatmeal every morning for years. It’s like oh, wait a second. I’ve been doing this to myself.
Ben Grynol: Because it’s “healthy.”
Andrew Connor: Yeah. I want to know who did oatmeal’s marketing, because I don’t know how. Gram for gram, oatmeal is like triple the metabolic impact of wheat. I don’t quite understand it. Anyway, I had a Cliff bar. My glucose is pretty rock solid usually. I went up to 160 milligrams per deciliter, which is the highest I had gone that month, from a single Cliff bar. I realized oh, not only is… okay, I’m sensitive to some of the ingredients in Cliff bars. But there was also no accountability. We have all of these parts. I take on Cliff bar, but Cliff bar is not unique. I think if you looked in a lot of Sporting Good’s foods, Gatorade and all these other things, it’s these products that are associated with doing healthy things or being active that there’s really no accountability.
Andrew Connor: This is another thing I’m really excited about, is we’re starting to have meaningful sample sizes for a lot of foods. It’s interesting for us to see, where I guess is our collective zeitgeist just completely mistaken? I’m excited to think longterm, what will it look like when a lot of people are equipped with CGMs and metabolic feedback loops to bring more awareness to a lot of these foods? Because if you were to put a lineup in front of me of regular milk, almond milk, cashew milk, oat milk, and pretty much be defenseless and not really know that one of those is metabolic jet fuel. Yeah, I’m really excited as accountability in our food system becomes a lot more popular.
Ben Grynol: Well, it’s a really wild thing to think about. There are a couple things I’ve been thinking about. One of them, and not to go on too much of a rant about this, but in the ’50s, ’60s, ’70s, I mean, it doesn’t matter the timeframe. But not that long ago, like 60 years ago, people didn’t even blink an eye at things like smoking. It was just something you did. In today’s age, people don’t blink an eye at eating sugar and highly processed foods.
Ben Grynol: Fast forward 60 years, now we’re like, “Oh, smoking is clearly not good for your health.” We look back on it and we’re like, “I can’t believe that we thought in the ’60s you just sort of did it.” I think we’ll look back, like in 50 years from now, we’ll look back on this period and be like, “Can you believe that we used to eat Twinkies and thought that was okay? Can you believe that we used to just hammer sugar, or hammer highly, highly processed carbohydrate products? Very, very highly processed breads and hamburger buns and things like that.” I think we’ll look back on this period and we’ll be like, “I can’t believe we thought that was okay.” I mean, hindsight is 20/20, but I bet you that’s something.
Andrew Connor: Yeah. When you think about it evolutionarily, we’re pretty much defenseless because in the span of humankind, we never had things that could light up our taste buds and our senses quite the same way to the extreme. We’re able to take food to 10, to make it as addictive to possible and as delicious as possible.
Ben Grynol: If that’s the word. If that’s the worse you choose, sure.
Andrew Connor: Yeah. I think our attention is limited, and so we saw this. As fat was vilified in the ’70s and ’80s, sugar was introduced to food because it’s not really solving… no one wants to eat bad tasting food and it’s easy to eat packaged stuff that’s shelf stable and all that kind of stuff. I think in many ways, we need to empower people to make really good independent decisions for themselves. Look at how their body responds to things, how they feel. How their glucose responds. But also just working with their primary health physician that cares deeply about holistic health. About being healthy as a person, not just the management. I’m seeing a really big shift of that. I think 10 years ago, it wasn’t as popular. 20 years ago, it certainly wasn’t. It’s something that I think a lot of people are waking up to.
Ben Grynol: Yeah, it has to do with convenience. We live in the age of Amazon. People want things quickly. They want things conveniently. That’s where a lot of these highly processed foods come from, or things like restaurants. Let’s talk about restaurants and on demand food delivery. It is a massive industry globally. A lot of restaurants are serving highly processed food. It doesn’t mean all of them are, but a good chunk of restaurants might be serving food that is not giving you a good metabolic response because there is sugar in the sauces, or there is just all these implications.
Andrew Connor: Well, it’s interesting. My wife is vegetarian and we eat a lot of vegetables at home. We just shifted a lot of our diet to discovering really good vegetable based meals. It’s always shocking when traveling how little vegetables are easily available. We’re originally from the southern U.S. Most restaurants, if they have a side of broccoli or something like that, it’s like four or five fleurettes, kind of just steamed. It doesn’t taste good. It’s like this afterthought for someone trying to be healthy or something like that. It’s a shame. It’s very difficult. Even if you were informed, it’s kind of hard to succeed.
Ben Grynol: Yeah. Do you want a side of celery with that? But that’s the vegetable. You’re like, “Okay.”
Andrew Connor: Yeah, or iceberg lettuce on your hamburger. Something like that. I would imagine the average entrée at a lot of sit down restaurants, in terms of grams of vegetables, is pretty minor. It’s difficult because I think people are not accustomed to looking for that. The demand is not there, so restaurants, they supply where the demand is. It’s delicious tasting food that I don’t have to think about.
Ben Grynol: Yeah. There’s even, when Kasey and I were jamming, there’s major changes to policy that need to be made because it stems so… it’s so deeply rooted from the time that people are children. In the school system, Kasey was saying that pizza is classified as a vegetable. I could not believe that when I heard it. She’s like, “No, it’s true.” It’s like how do you expect if the way that people are being educated, if the food that they’re being given isn’t giving them a fighting chance of being metabolically healthy from the ground up, we’ve got a lot of work to do.
Andrew Connor: It’s hard, yeah. The mission Levels is on is a multi-decade mission. We will not be able to do it alone, but we’re really hoping that we’ll be able to have a meaningful impact to reversing a lot of these trends.
Ben Grynol: It sounds like the name of an ’80s band, like Hario Speedway or Hario Speed.
Andrew Connor: H-A-R-I-O, Hario Switch. It’s a cross between a pour over and almost like an arrow cross or something.
Ben Grynol: What’s it called? A Hario Switch?
Andrew Connor: A Hario Switch. I had to buy it from Amazon Japan. Yeah, no. This is serious stuff.