Edacious founder Eric Smith explains why he built a food nutrition testing lab for growers based on the often-competing philosophies of environmental stewardship and consumer self-interest
Acres U.S.A. Could you tell us some of your background — how you came to start and manage a company that tests farmers’ products for nutrient density?
Eric Smith. Sure. I had been working in venture capital and philanthropy and had a long career at the intersection of ecology and economy, trying to find market-based incentives and ways to address climate change, reward land stewardship, and incentivize better production practices through market-based roles and incentives. I experimented with a lot of different models, and I spent a lot of time over the course of my career with payment for ecosystem services, both in developing countries and in the U.S. — how to get landowners paid for improving ecosystem services like carbon, water, biodiversity, etc. I worked in certification and third-party auditing to provide a stamp or certification that a practice was being implemented and to try to get the consumer to pay for something.
But ultimately, after 15 years of that work, I realized that unless it’s hidden in the cost of doing business and embedded in doing business a better way, ecosystem services were always going to be seen as a cost upon the system rather than a better way of doing something.
In finance, we talk about pricing externalities, but really it was that the market didn’t like the added cost of these things and wasn’t willing to pay for it. And that was especially true when it showed up for consumers. Consumers were less willing to pay for environmental attributes than many in the environmental movement had hoped.
We were in a tough spot — how can we get climate-friendly regenerative agriculture and good land management practices actually paid for, so that the producer and the land manager have incentive to adopt practices for healthier soils? After years of looking for solutions, we concluded that if we could find a way to connect the health and wellness movement to the environmental movement, we’d have a direct incentive paid for by people’s own self-interest in health and wellness. And that really amounts to higher food quality.
So we began to look at research and organizations that were focused on this space — seeking to understand the relationship between soil health and the nutritional quality of food. John Kempf was one of the people I spent time with early on in that discovery. There were a few other organizations, including Dan Kittredge and the Bionutrient Food Association. There were a number of researchers looking to understand this connection, including the Rodale Institute and others. The idea that healthier soils produce more nutritionally rich crops was very compelling, and if we could find a way to aggregate data to show that there was a difference between these food types based on the production system, we could create a market for paying based on differentiated quality.
The way you measure nutrition in food is through analytical chemistry. So we had to build a large lab and figure out all kinds of analytical processes and build data and infrastructure to make all the data comparable and sound. It’s been exciting, but the root of it was very environmentally focused in terms of encouraging better practices for land stewardship — while realizing that the biggest market driver was, “What’s in it for me? Why do I care?” And the answer to that is that people want to be connected to healthier, more nutritious, safer food.
Acres U.S.A. That makes a lot of sense. Can you talk about your process for testing? I know that a number of years ago, Joel Salatin had his eggs tested and was able to show that they had much higher nutrient levels than supermarket eggs. But it was really expensive. What have you developed to make this process easier, better, and less expensive?
Smith. In the past, you would have had to identify a lab, figure out what you should actually be testing for, and then spend tens of thousands of dollars to try to produce a comparable analysis. It was just cost prohibitive — it was inaccessible. Nobody had the data, and you could send results off to two different labs and get two different results. So it was also very confusing.
A lot of this inquiry and research originated in academic settings where researchers wanted to compare grassfed, pasture-raised, and conventional beef, for example, to see what the results were. There are plenty of good papers that show significant differences in fat quality, protein quality, vitamin quality, and minerals between those production systems. The evidence is there in the literature, but turning that into a commercial outcome was almost impossible.
The first barrier is the cost. To generate a single nutrient-density analysis of all the macros and micros that constitute a food, you’re talking about $3,000 to $4,000 per sample.
Acres U.S.A. Back in the day, or today?
Smith. In the market today — not at our lab — but if you go to any normal lab and say, “I want to know about all the fatty acids, proteins, amino acids, fiber, vitamins, minerals, and heavy metals in my product,” you’re talking about $4,000 per sample at any traditional commercial lab. Then you get a report back in the form of a certificate of analysis that gives you no context — it tells you nothing about whether this is a good level, a high level, or a low level. It’s unactionable.
We decided to look at the business model and see if we could innovate. So we set up our lab a little differently so we can run batch testing, which keeps costs low, and we were able to drop the cost of testing by 10x. Instead of $4,000 per sample, it costs about $400 per sample. And then we bear the cost of generating all these comparison data sets. We put that in our software, and people can see what that looks like in real time from a nutritional standpoint. The benchmarking is included in the cost of the service, so growers know how they’re performing.
Those are the reasons this has remained prohibitive — the cost, what the results mean, and how they compare. We’ve sought to solve all three of those problems.
Acres U.S.A. Can you speak about what kind of testing equipment you’re using — GC mass spec, ICP mass spec, etc.? And what’s the workflow when you receive a sample? Are you asking people to send it on dry ice, depending on the product?
Smith. First, there’s a call with our team to understand what your goals are. Are you doing controlled trials? Are you trying to experiment with different practices? Are you trying to tell a better story for your consumers? Then we figure out how many samples we should test so you can trust your results and your data.
Traditionally, most labs work on a sample-per-analyte basis, which means that if you want to measure vitamin B6, one sample costs $350. Growers would ask, “Why should I measure B6? What does that mean?” But we do anywhere from three to six samples at a time for our customers, testing a variety of factors. Once they’ve onboarded, they can do one sample at a time, but at the outset, for the nutrient-density analysis, we want to give them a picture of the complete nutritional profile of their crop. That sets up the baseline.
Those samples come to our lab, and depending on the food type, they may be shipped overnight on dry ice. We try to make sure it’s harvested right at the peak moment and gets to our lab within a day or two. Then the first step is sample prep. This is where there’s an analogy to blood testing for human health — an industry that has exploded — but people don’t realize that blood is very easy to work with because it’s a liquid and largely roboticized. The problem with food is that it all needs to be prepped. We have to chemically and mechanically separate the sample into different types to go off to different machines to be analyzed.
The three main instruments in our lab are gas chromatography mass spectrometry, liquid chromatography mass spectrometry, and inductively coupled plasma mass spectrometry. Those are the three workhorses of any analytical chemistry lab. Our methods are standardized to AOAC-recognized best methods. We don’t do anything proprietary except process innovation and business-model innovation — how we think about serving our customers, rather than acting like a traditional lab with per-sample, per-analyte pricing.
Acres U.S.A. Can you talk about the marketing piece — how are farmers starting to use this?
Smith. Traditionally, when someone’s marketing their food, they might have a spec sheet that they share, and that spec sheet will tell the consumer a bit about their farm or their process or their product. If they include nutrition, but it’s basically a copy and paste from Google or a database — they’re never including their actual nutrition data because that testing has been prohibitive.
We provide nutritional spec sheets that allow you to market on the basis of nutrition — the relative quality of your product. You can say, “My wheat has this much more minerals, vitamins, protein, etc.,” and you can show that based on real lab testing. We like to say it’s the cheapest marketing investment you can make — using us as a verified partner to put together a nutritional spec sheet. You can then take that information and create a public page and share that and put links on your website to show your product has been tested.
Acres U.S.A. Are the comparisons to a USDA standard or to something you have tested yourselves — a conventional product off the shelf?
Smith. Both. Just to be clear, right now, we provide growers that comparison data, but it’s private; starting next year, we will turn those comparisons into a public view for them to share. We’re still figuring out how to turn that into a public page. Right now, they can screenshot that and copy and put it on a spec sheet and can share how they’re performing. But at the end of the day, it’s about the absolute value of that nutrition and then the relative value of that nutrition compared to foods of the same type. This gives them a complete nutrient-density specification — how nutritious their food is relative to other crops.
Our customers range from small family farms and growers all the way to very large brands. We work with the first half of the food system — everything from the farmer and the farm gate all the way to the CPG and the packer-distributor that wants to communicate downstream with the retailer or the consumer.
Acres U.S.A. Have there been many surprises yet? Results that were either much better than expected or not as good as hoped for?
Smith. Especially in animal products, we’re seeing the two extremes — some samples are far better than we thought they would be, and some are far worse than we thought they would be. We do both nutrition and contaminant testing, and we’re seeing a lot of contaminants show up in conventional products — and in organic, in some places — that are unintended but are showing up. We know there’s a lot of unintended fraud because of deceptive practices, largely due to grain imports that are being certified organic that aren’t. So a lot of the surprises are coming on the toxicity and contaminant side.
Another big surprise has been the variability. In a single fruit, vegetable, or grain, you might get 10 percent of your recommended daily value or 200 percent in what appears to be the same piece of food. We’re seeing that some crops, under certain practices and depending on varietal, can deliver incredible amounts of nutrition, and some varieties are essentially depleted of that same level of nutrition. That’s what we’re really trying to unlock — the ability to see that information so people can make better decisions.
Acres U.S.A. Are there any correlations that you can talk about yet for the farmer? For example, are there any soil health practices that we’ve all thought were great that aren’t really doing much nutritionally?
Smith. I don’t want to overgeneralize, and we’re still in the early days, but we do see that varietal selection plays an incredibly important role. We have lots of customers doing controlled trials and asking how different applications change results. The problem is that they’re in year one of that practice change. The nutritional difference may be that they’re shocking the system with a new practice, and that crop or tree might not be used to that practice. Nutrition can actually go down in year one because you’re introducing a whole new system.
We see this holding true on the contaminant side. There’s a three-year process of rebuilding and then coming out of it, and we’re seeing that with higher levels of soil organic matter, with the right genetics, there’s some pretty special stuff happening. We can see that in the data.
We’re the technology platform to help people discover this information, but we’re not able to ask all these questions at all times across all systems. Our customers who are doing controlled trials are trying to determine, “Was that a varietal or a practice that was driving that outcome?” Then they can start to make decisions about how to shift. But that’s going to be a long journey.
Acres U.S.A. Can you talk a little more about the varietal differences? You’re saying that whether I’m growing, as an example, Amish Paste vs. Speckled Roman vs. Plum Regal tomatoes could have a huge difference in nutritional content, even if they’re all grown in the same way?
Smith. Yeah, it really could. The difference is going to be driven by how that varietal was conditioned for that soil. The organic seed industry is struggling because many of those varietals have not been bred for organic systems, or growers are using conventional seeds in organic systems, and they’re not optimized for those systems. This is a universal issue — we’re using the wrong tools in the wrong places, and we all know that. If you have an incredibly high-input market garden and you’re using conventional seeds, it’s going to perform differently than seeds that have been in 15 percent organic matter for years.
Varietal plays a very important role in driving the nutritional productivity of the plant, and that’s what drives taste. That’s what drives nutritional outcomes. Breeding is a big part of this story, but the soil also plays a huge role. It’s a tough thing to balance.
Acres U.S.A. Do you have any results yet on genetically modified corn versus a conventional hybrid versus an open-pollinated heirloom?
Smith. We’ve done a lot of heirloom corn versus GMO traditional corn, and you’re changing the macronutrient structure. The heirloom varieties tend to have higher levels of oil, lower levels of protein, far more micronutrients per serving, and more phenolics and flavonoids. Modern varietals have been bred for carbohydrate production — they’re putting on sugars and starches at a much higher rate, which means the seed is more a combination of water and structural starch and carbohydrates rather than nutrition. And that makes sense — we’ve been breeding for that outcome and for yield for a long time. So of course that’s the outcome we’re seeing.
We just have to ask ourselves, “Is that the right outcome, and why are we doing that?” There will be trade-offs, too, between yield and nutrition in some cases. If you’re optimizing for a really high-nutrition variety, you’re going to see more nutrition per acre, but you’re going to decrease yield because the plant is putting more work into complex metabolic processes instead of just growing bigger.
Acres U.S.A. What do you think this will look like 10 years from now — will every grower be testing their products? And will consumers? And, of course, the inevitable question is whether there will ever be an in-field or handheld device to measure nutrient density.
Smith. For us right now, we’re trying to grow to provide a verified outcome — a third party between the producer and the consumer. If somebody were just reporting themselves from a quality standpoint, the consumer would have to discount that. We think handheld — what I’d call in-field sensing and in-field measurement technologies — will be a part of assessing quality in the supply chain, but it’s a different business than ours because those devices are never going to be able to measure at the depth we can.
We do a lot of photo spectroscopy in our lab — a combination of XRF and UV-Vis and infrared — to try to drive down cost. We just demonstrated that we can measure an omega-6-to-omega-3 ratio in dairy products at $3 a sample rather than $300 a sample. That’s exciting, but doing omega-6-to-omega-3 verification — which is essentially grassfed verification — for a $3 test is not really needed. I can pull samples off the shelf and tell you whether something is grassfed. But at the end of the day, that’s a different business model and proposition. Low-cost testing becomes important for chain of custody and for verifying that you’re getting a certain quality.
Where this is all going, in my view, is really about helping farmers make better decisions and helping them get paid based on the nutritional outcomes of their product. I envision a world in which a farmer knows the nutrient density per acre of their farm and the nutrient density of the item they’re producing, and then we guide farms and supply chains to optimize for nutritional outcomes rather than just yield outcomes. There’s no doubt that we’re awash in calories and carbohydrates, and much of that is wasted in the supply chain.
We have an opportunity here. I would love to see animals back on the land. I would love to see more fruits, vegetables, and market gardens dominating a bigger piece of our food system. I think we can design higher-quality grains that are then less milled. They become stone-ground and are treated with the right consortia of bacteria and turned into really nutritious grain-based products.
I think there’s a way out of this, where nutrition becomes the benchmark for how we make decisions.
Acres U.S.A. What’s your philosophy on helping the consumer understand this? Are you trying to develop a single metric that would somehow be computed — a nutrient-density score? And do you think something like that should be standardized by law?
Smith. We do envision some type of roll-up score that conveys relative nutritional quality. We’re a lot more focused right now on apples-to-apples, apples-to-oranges, and apples-to-fruit comparisons than comparing apples to beef. There’s so much complexity when you try to create a universal scoring system that compares an apple to a steak, to a box of pasta, to a candy bar. They all have such different nutritional profiles that you’re trying to account for everything, and then you have to start accounting for what’s bad about it.
It is possible, and we do have a nutritional profiling system in place, but we’re somewhere between crawling and walking on that because we don’t think it’s really necessary right now to move this market along. With the consumer, we need to split this into two markets — the processed-food center of the store versus the perimeter of the store. Right now, we’re just trying to bring innovation to the exterior of the store — all the fresh food, produce, grains, staples, etc.
When you get into judging the nutrient density of a processed food item, we philosophically disagree that you should apply nutrient density to those things because, inherently, in the way they’ve been processed, fortified, and added to, they are no longer nutrient-dense in the way nature intended.
We don’t know that we want to apply the same framework to judge the quality of whole foods that we would apply to processed foods, but we do have retailers engaged with us and the government on front-of-pack labeling based on nutrient density. We’re just trying to figure out how to navigate that in a way that improves the value of whole foods relative to processed foods.
