This quarter, we tested minimally processed foods like dried fruit and nuts and cold-pressed energy bars straight off grocery store shelves. We tested big brands alongside small niche companies.
The result? A market snapshot of 27 natural food brands. In this 20-minute webinar, Mary Galloway reviews the findings. Learn:
- moisture secrets of fresh-tasting, clean label dried fruit and nut snacks
- what some big brands do much better than small ones
- which brands have red flags for mold growth—and worse
- the surprisingly high cost of moisture inconsistency
Mary Galloway is the application scientist for METER Food and manages the Food Research & Development lab. She is a contributing author on several publications concerning water activity and its influence on physical properties. Mary uses her years of experience to help customers understand and solve their moisture-related product issues.
In the webinar today, we’re going to talk about an overview of the Market Snapshot, this study that Brad was just talking about, we’re going to talk about what the data shows. What kind of trends are we going to be able to find related to natural foods and in particular, our focus was dried fruits, nuts, and cold pressed bars. How do you get the moisture equation right? How can we use water activity, moisture migration and what we know about that to make a good, safe and reliable product? And what are the costs if you don’t get that right and some key takeaway points?
Study set up
We started with a lot of products. We started with 96 different products: 41 different kinds of dried fruit, 37 different kinds of nuts, and 18 cold pressed bars and we focused on measuring their water activity and moisture content, pH for the fruit, and then overfill in packaging based on their list weight for all of these products just to see how well a manufacturer is able to fill their packaging.
These products that we’ve purchased were both in stores and online. And they were from regular grocery stores and specialty stores that focused on natural and organic products; retail chains, large box stores, everything—we just tried to get a really wide cross section of where people could source these types of products. It was a lot of data to go through. For every dot that’s on each of these graphs, we had three samples of that product and we made three sub samples and tested each of those three times. That’s 27 samples, 27 data sets for each dot on this graph. You can see that’s a lot of data to sift through.
Plus we have a lot of variety. When we’ve done other snapshots, like beef jerky, or pet food, we can kind of take it as one big grouping, but we can’t really do that here. We need to really break it down individually to look at what the trends are, comparing apples to apples, as it were. So we have dried fruit and these are the ones that we tested here. Apricots, blueberries, cherries, you can read down the list there. We have some nuts. We also have cold pressed bars and you’ll see that the brands are listed in the legend there. We just coded them so you can’t see what brands are here, but these are all major brands for cold pressed bars. They are a mixture of raw ingredients, and then they’re formed and packaged. There’s no baking involved with these. I’m going to start going through them one by one, and I’m just going to take one type of fruit, one type of nut and and go through that as an example so we can pull out some of these trends. [Note: the full data set is available to attendees on request.]
Most dried fruits follow a trend. It’s hard to see in this particular graph, so we’re going to focus on the causes of the variability, the impacts of pH and water activity (including how those work together), yield, and then snack versus ingredient.
So our first one here, I’m going to look at our blueberries. When we look at a fruit we’re really looking for the variability that comes from variability in the sugar and fiber content. The sugar will bind the water and that will lower the water activity. With the fiber really we’re talking about the starch in the fruit.
Osmotic and matric effects
When we talk about sugar and the binding of water, that’s really an osmotic effect. If you hear that term anywhere in the science community, that’s what we’re talking about here. The fiber is related to the structure—that’s the matric effect. That’s what makes up the structure of that fruit. We can have variability in different varieties of the blueberries. And we can also look at drying methods—oven or dryer, versus maybe on the ground. There are some fruit producers that will just leave the fruit out and they’ll have a natural sun dried product that they will just leave and harvest later.
Isotherm trend line
For the blueberries, you can see that most of our data points are following really nice trend when we relate water activity and moisture content, which is really nice. There are a few outliers, nothing is quite perfect, but I wanted to call attention to one in particular. This one about point eight six water activity, was really an outlier. And interesting to know about that— it is just a sweetened blueberry. I thought maybe if it was unsweetened [that might explain it] because sugar, like I said, will drop water activity. Maybe if it was unsweetened, it would explain where it was because we don’t have the sugar to lower the water activity, but it actually was a sweetened blueberry. So that wasn’t the case for this. I don’t know what the cause could be. It could be related to variety of blueberry, it could a particular group, when it was processed, that actually didn’t get sweetened like it was supposed to. Most of the time, you can see as we follow along the isotherm, they follow a really nice predictable trend. But we are going to see more variability with the fruit and with every type of fruit we saw some differences.
Let’s talk about pH and water activity. Fruit is acidic. And if you have something that’s acidic, you can allow for higher water activity values. The rule is if you have something that’s not heat treated, and then not packaged right away, like a dried fruit, if you have a pH lower than 4.2, you can have a water activity that’s as high as 0.92 or lower. Even though this one is at 0.86, which would be beyond microbial limits—not just affected by mold, but into the zone where we could have potentially hazardous bacterium growing in there—because of that pH, we’re able to still have a safe product at that high water activity. These blueberries have a pH of 2.9 and the average for the snapshot is 3.6. We are way below that 4.2 range, so that’s really nice.
Looking at the isotherm trend line, you can kind of pick whatever spot you want that is perfect for your fruit. If you would like a fruit that has a little more sugar or a little higher moisture content to be a little tastier to consumer then you can pick the spot you’re trying to go for. In essence, you can use the isotherm to balance the water activity and the moisture content.
Snack vs ingredient
The last thing I want to talk about is snack versus ingredient. These are two different things and they have different criteria. You might want a snack that has a nice water activity but focuses on a good moisture content. You want it to be sweet, and you want it to be nice and tender for snacking. But if it’s going to be ingredient sourced to someone else, they might want a drier fruit, they might want a lower moisture content, lower water activity [to match the water activity of their product], and they also want no added sugar because they don’t want to have that included in their product or their mix or whatever. With a fruit, you can lower the pH and have a safer product at higher water activity as a single ingredient snack. That’s not the case when you mix it with something else. It still has a high water activity environment, now you’re diluting that pH, and the pH is not there to protect the product. We have to account for that when we’re mixing things, especially when we’re talking about cold pressed bars.
Isotherms and trend lines
Another thing I mentioned a bit before, but I wanted to stress again, looking at this graph here on the left for the dried fruit—you can see patterns in the data—trend lines for the apricots and blueberries for example are really easy to see. But every one of them has an outlier. The blueberries, the one at point six was way out there. Every type of fruit had some kind of outlier. The worst of them were the mangoes actually.
[Mangoes] actually showed no trend between water activity and moisture content. And the reason for that is that the mangoes are very fibrous. If you think about mangoes, they are very fibrous and have varying amounts of sweetness, so the sugar content is variable. And when you have both of those things going on, you’re going to get a lot of variability. The takeaway from this is that you need to be prepared for that variability in your product, especially when you’re sourcing it from somewhere else. You can’t just assume that you’re always going to get a repeatable product every time. You’ve got to test it yourself to know where things lie because it can really impact your overall product. It could be pushing your water activity too high if it’s like that outlier with the blueberry or give you issues with moisture migration.
High variability in fruit as an ingredient
You need to know that there’s high variability and fruit or I should say there are cases where there is high variability and there tend to be outliers with the fruit. It’s difficult to predict that.
Now I want to move on to nuts. We’re going to talk about causes of variability, the impacts on yield and snack versus ingredient and what we need to understand about that. Let’s take almonds. There are 15 different types of almonds, all types. Raw versus smoked, salted, unsalted, flavored, all of those are in this data set. You can see that they trend very, very nicely and very closely together. They’re different and more reliable than the fruit.
This is pretty common for all of the nuts. They do have less variability. The variability in water activity and moisture content is almost entirely due to roasting. When we roast, we’re going to drive off more moisture. Salt can play into that too, because it acts like sugar, in that it’s a humectant that will bind with water. It can lower the water activity while maintaining that same moisture content, which is nice. Sugar will do the same, so your moisture doesn’t go anywhere, it’s just bound and not available for anything else [like mold or microbial growth]. The moisture is still there and it impacts the yield and texture as well.
If you have a higher water activity in fruit or nut, generally they’re easier to chew, maybe a little better for snacking. But if you’re sourcing your nut for another product or you will have a whole whole nut and you want to ship it, you want to make sure that doesn’t break: these can all have an impact on the ideal water activity for your product.
Also, I wanted to talk about possible rancidity at low water activity. And the reason I want to talk about this is because it’s not intuitive. Most of the time, people think that rates increase together, when we increase the water activity, other reaction rates increase. That’s actually not the case for rancidity. I drew a little graphic on here to show you how the oxidation rate changes. At low water activity, it starts out very high. It comes down and is lowest around 0.32, 0.34, and then it goes back up again. This little graphic, it’s just the concept, it doesn’t have scale or units or anything like that. But I want to point out how it falls with our data set on almonds. If you look at this almond, that one, that’s very, very low there. It’s actually at a high rate of rancidity. And when you compare that to all the rest of them, it’s actually at higher risk of rancidity than pretty much any other almond we have here. It might surprise some people to know that if you have a low water activity, you could actually be have a much higher rate of rancidity for your nut.
For yield, just like we talked about with dried fruit, you can kind of pick where you want to be on the isotherm, depending on what properties you want, what kind of texture, what kind of flavor, that kind of thing. You can pick your sweet spot and then use water activity to hit it every time.
Snack vs ingredient
Snack nuts vs. nuts that will be included as an ingredient in another product—they have different criteria. If you have a snack almond, like I mentioned, what you want for that type of product might be different than what you want if you’re sourcing it for something else. You need to be aware of the differences. And depending on what you’re mixing it with can change what water activity and moisture content are ideal for your product.
Cold pressed bars
With cold press bars, you’ll notice that they are much higher in water activity range. Some are actually past microbial limits, which we talked about before: 0.70 for mold point, 0.85 for potentially hazardous pathogens. For cold pressed bars, you need to be aware of ingredient readings. If you’re going to be mixing something like that blueberry with the high water activity, it works to have a low pH with the blueberries by themselves. But now if you mix it into a bar, that pH gets diluted, it’s not keeping anything safe anymore. But you still have the high water activity of the fruit. What happens when you mix that with something else? And now do you have an issue or not?
Measuring incoming ingredients
Knowing what your ingredients are, and setting the ingredient specifications properly will keep your product from having variability that you don’t want. The other thing you need to know is there’s going to be moisture migration. If you have two or more ingredients that you’re mixing together with different water activities, and it’s likely that you will, there’s going to be moisture migration. You need to know what that is and how to control it. I’ll be talking about moisture migration in detail in just a little bit.
Bars in the danger zone
Focusing on these four here, these are beyond that microbial limits. I had a big questions when I saw this knowing that these bars cold pressed, knowing that these are raw ingredients, and they are sitting on the shelf at this high of water activity. I wondered how they could possibly be able to do that. How could they possibly be able to have a water activity so high and not have any issues?
You’ll notice that there are two in particular, they’re in this extra dangerous zone. Those, interestingly enough, are cookies and cream cookie dough. The other two that are a little bit lower than 0.8, those were peanuts. All of these ingredients have historically and not that long ago actually, been connected with salmonella. This contamination can happen with low moisture foods. A peanut, when it’s in its natural state is fine. It’s at a low water activity, 0.3, 0.4. And it’s fine down there, even if it was contaminated, the pathogens can’t grow. But the problem is, you take the peanuts, make peanut butter, or whatever. And then you put them into something else that has a higher water activity. And now you can have a problem because in the higher environment, that pathogen can start growing. It becomes a big safety issue.
So how do these bar manufacturers get around it? Well, they pasteurize their ingredients. And if they pasteurize their ingredients, they kill off any of these pathogens before they mix them together. [Note: the Q&A has further discussion of this issue.]
Another thing I want to talk about is temperature abuse. Because if you make a product like this, and then store it or ship it to a place where it can be exposed to extreme conditions, you need to know what’s going to happen to the water activity of your product. In general, a higher temperature means a higher water activity. You need to answer some questions: What happens when it’s at a higher water activity? Do we have more issues? Is it still safe?
Also, we can have ingredient moisture migration. So how do you prevent or at least control moisture migration? The short quick answer is you have to keep the water activities the same. Here’s the hard truth about moisture migration. It’s actually driven by water activity and not moisture content. If moisture migration was driven by moisture content, that would be great in a way, because I think it would make it easier for people to understand. More people are familiar with moisture content. Water activity is a little more difficult concept for people to get.
But the differences allow production of some different types of food. If moisture content did control moisture migration, we couldn’t have moist/crunchy combinations like raisin bran. We’d have hard raisins and soggy flakes. And we’d have no cheese filled crackers or cream filled snack cakes. We couldn’t have any of those things because those moisture contents of those individual components are completely different from one another. You would have this homogeneous product, but it would be kind of a nasty mess.
Blueberry to almond example
So it’s actually wonderful that water activity is the driving force for moisture migration. It creates some great formulation opportunities when you understand it. So here I’ve got an example where we’re taking the blueberries that we were showing in the example before and the almonds that we also talked about here. What happens when we add them together? Well, we won’t know unless we know their water activities. So, in this example, we’ve got the blueberries that are really high, at 0.86 water activity. And our almonds are at 0.53. So which way is the moisture going to go?
It’s going to go from the high to the low until they are equal. Always. So water will move from the blueberries into the almonds. It’s simple, and that’s how it works. But where will the water activity end up?
Direction is easy, final water activity more complicated
That’s the complicated question because it’s not in the middle. It’s not like a weighted average of the two water activities. Actually you need to know the isotherm to know how each responds to moisture changes. It’s a little complicated. It’s not like an average in the middle although we do have tools that can help us figure it out, predict it. But that’s another webinar.
In this example here, you’re going to have moisture leaving the blueberries going into the almonds. And that could change the almond’s texture. It might get a little softer, squishy or not desirable. It could also increase the oxidation rate. It’s already at 0.53. And we know if we keep going up at this point, the oxidation rate is going to increase. So we could have rancid soggy almonds if we were going to be putting these two together.
Different incoming ingredients
Okay, what if we have the blueberries at 0.41 water activity? The almonds are still at 0.53, what’s going to happen here? Well, almonds have the higher water activity this time. And it’s going to move to the blueberries. And this might be okay. You would need to predict the final water activity in the blueberries and the almonds [you can do this with isotherms] and whether this is something that’s going to produce the right kind of texture for the blueberries and the almonds.
Equalized water activities
What if they’re at about the same water activity, is anything going to happen? Well, no, they’re about the same. And they’re just going to stay that way. You’ll be able to maintain the texture of your blueberries and the texture of your almonds, and you would have no troubles with these two.
Better way to measure moisture
All right, the other thing I wanted to talk about was water activity being a better way to measure moisture. I have here an example, with pecans, a real case study with a pecan grower who was having mold issues. They always had a spec of 4% moisture content for their product here. And you can see that in the orange data on the on the left graph here. By using the isotherm for pecans, you can see that it corresponds to a water activity below point seven, so it shouldn’t be molding, that shouldn’t be an issue.
Precision plays a role
Now if we zoom in on to the right graphic right here, this is just the same information zoomed in so we can see a little more detail. We In fact, do see at 4% moisture content, it is below about .685, let’s say, still well below point seven. Not really an issue. However, the measurement method [the customer was using] for moisture content was only accurate to a half a percent moisture content. If the measurement is 4.25, so an extra 0.25 increase in moisture content puts the product close to this point seven value. It’s not much of a difference— and it’s not something that you could see using typical moisture content measurement methods. However, this little increase here is really large on the scale of water activity. A 0.02 increase in water activity is huge. We can easily see that when we measure water activity where it’s very, very difficult if not impossible to be able to measure the same when we’re just looking at moisture content. And that’s because the scale is so much better with water activity, it’s actually six times more precise. So keep measuring moisture content, but if you’re measuring water activity, and you’re using the isotherm, you can easily see where your product is going to have issues. You can say, Oh, well, this is actually in in danger zone for having mold growth, we need to dry or roast a little more to drive off the water so we don’t have issues once it leaves.
You might say then, why not move the specification to be safer instead? The reason is yield, and we’re going to talk now about the impact of the precision in your measurement on yield.
We’re going to look at it very simply with some simple numbers here using dried prunes. A producer making 30.5 tons every year has a target moisture content about 30%. And let’s say they want to increase that target moisture content by a half a percent now. We can use the isotherm to see this won’t be an issue for mold or anything, we can just increase our moisture content because we have that kind of visibility. What will happen? So no mold issues, we’re can look at pure yield.
A big factor in being able to shift their spec is being able to measure more precisely. Initially you can see their variability is pretty wide, and the average moisture content is 30%. If we’re able to tighten up that process by improving the precision of the measurement, let’s say control the moisture content to half a percent. [Depending on your product, using the isotherm to make the moisture content measurement could easily give you that kind of precision.] Now we’ve shifted this graph here because we have a precise measurement, a tighter process control, you can adjust your moisture content to produce a nice, consistent, safe product every time.
Doing the math
If a more precise measurement does allow the producer to increase their moisture content by half a percent, that is another 1500 tons of prunes that they could produce. And if prunes are selling for $3,250 per ton, which is about right, the annual increase is about a half a million dollars. With better monitoring, they can increase their yield and their revenue. You can also look it out the other way. If they’re not controlling if they’re looking at it in this way, where they have this wide, wide variability, or if they’re consistently overdrying their product because they’re not measuring preciesly, they’re losing. If it was a whole percent, we’re talking about a million dollars. It’s huge. It’s a really big deal if you can control your process with a tighter measurement.
Big brands vs small
One of the questions we posed coming into this study was, do big brands do better than small? Are they better at controlling their process? It’s a difficult question. I looked at it a few different ways. One of the ways I decided to look at it was this overall water activity variation, because we can see water activity has this really great accuracy when you measure it properly so you can just really dial in and we can get a lot of detail in that reading. So what is the overall variation of water activity? And basically, to me this would be assessing their process, are they getting nice consistent water activity readings, or are they all over the place.
On this graph, the blue is the standard deviation. That’s just how varied the readings were. Then we have a water activity max and min. I normalized all this to the initial water activity, so you can compare data and just looking at variation from a standard. Y is a big company and K is a small company and they both deal with dried fruit. For nuts R is the big company and S is the small company. Cold press bars, B is the large company and V is the small company. It was surprising to see that there’s actually quite a lot of variation. It seems that the larger brands here do not do as well as some of the smaller brands. With cold pressed bars actually the larger brands do better, but for the nuts and the dried fruit, you actually see more variation with the large brands. You have to consider that we might not be looking at the same products. We’re not comparing apples to apples, and some of those products themselves can have a lot of variation. If the large producer here made mangoes and K did not, just because they had mangoes they might have a large variability in their readings, but the readings variation is quite high. What if we don’t look at overall variation? What if we just picked one single product, so apples to apples here, and now you can see that the scale is much smaller. We’re much tighter now when we’re looking at a single product. But we still are seeing, especially for the dried fruit and nuts, that the big producers are having more variation than the smaller. I don’t know anything about their processes, so I can’t really talk about what they need to do or change. But, I can see here that they are having quite significant variation for the same. Maybe the small companies were doing a little better because they’re working in smaller batches. They’re able to control and monitor their processes a little better. Cold press bars on the other hand are actually pretty well matched the small to the large.
There’s one last way to look at process control, and that’s by looking at the overfill. This is based on package weight, so that’s been normalized as well. We’re just comparing the percent overfill for some of these companies. Now you can see this is where those big companies do better for fruit and nuts in particular. I wanted to point out something that’s kind of unusual here with K and with V, you’ll notice that the gray, the minimum overpack, is actually on top They always were overfilling. It got put on top, just an anomaly of the way the data were presented.
Then I threw this one on here for fun. These are house brands with a co-packer. We have some ridiculous overpack. For this particular company for fruit they had, what was it, 27% over pack. That’s crazy. Over a third more than they the listing on their package. I did check all of the data I want to just assure everybody that it’s real. This is an average of a lot of data. It’s not just one little outlier that we found, but a compilation of a bunch of data here.
Let’s talk about these house brands with co-packers. So interestingly, this one is a large box store, and I know that they work with mid-sized co-packers, and they do impose a fairly tight control specifications. G is a large retail chain and you can see that they perform about the same as this first company here. This last one is a little smaller company, it’s a retail chain too, but it focuses on organic and natural foods and they do use small co-packers and they don’t have tight control and you can see that here. A couple other things I wanted to point out this is overfill by brand and everything that they make. It’s fruits and nuts and bars, it’s everything all combined into one. I also wanted to point out too, that some of the reasons that these could be under packed is maybe because of inadequate packaging. Because if they’re filled correctly but the packaging isn’t preserving the product like it should it could lose moisture and when it loses moisture, it loses weight. And that’s where we could be seeing some of this underfill. These are my theories, I don’t know of course. So the takeaway from this here, do big brands do better than small? In some areas, yes, but in a lot areas they don’t. Big brands may think that they are doing better than they really are. Everyone can improve.
Dried Fruit: Summary
Dried fruit has more variability and it’s really hard to predict those outliers. I would like scientific explanation, but I don’t have one. It’s hard to predict those outliers. Fruit also uses that pH to allow for a higher water activity. And you need to use caution when you use it as an ingredient because you have to account for that variability. So if you don’t know what’s coming in at you, moisture migration could be an issue when you create a high water activity environment and include dried fruit as an ingredient.
Nuts have less variability, because of the structure of the nut. Most of the variability comes from roasting and driving off water. They are more susceptible to moisture migration when used as an ingredient because they typically live in a lower water activity than dried fruit. You can have rancidity at lower water activities. You need to use caution when using as ingredient because potential contaminants can become an issue when they are used in products with higher water activities. We talked about salmonella with the peanuts, and multi-component cold pressed bars in higher water activity ranges, some beyond microbial limits.
Cold pressed bars: summary
Some cold pressed bars had high water activities. They get around microbial proliferation issues by pasteurizing their ingredients prior to processing, but they need to have a tighter process control and make sure they know what they’re putting together so that they can monitor and control any moisture migration that might be happening. Water activity drives moisture migration, and is a better way to measure the moisture content. You can use an isotherm to allow you to have tighter processing specs. You can also find a specific water activity-moisture content combination that is just the right place for your product. Then you can control your process to consistently hit that. If you reduce variation in processing, it can be worth a lot of money. Plus the added benefit of being able to deliver a consistent product with a spec that’s tied to quality and safety, because you know if there are going to be any issues, and you can mitigate those before they become a problem.
Question: What’s the general water activity guideline for or limit for for dealing with mycotoxins or other fungi?
Answer: If you’re talking about things that are going to make people really sick, so this is not talking about mold or things that are unappealing, but what actually makes people sick, it’s 0.85 aw. And one thing I want to kind of stress about those bars that are pushing that limit: they have to be careful They do have to pasteurize, and they have to make sure that they are careful about that process. So when I say they have to have a very tight process, they can’t fail. If they have their product at that high water activity, they can’t fail in any of their pasteurizing or anything that goes into that product, or they will definitely have an issue, especially using raw ingredients that are grown in the wild. Natural products come with contaminants.
Question: You mentioned pasteurizing, and there’s another question: if they pasteurize their products or ingredients, do they need to worry about water activity at all?
Answer: I would, because here’s the failsafe. You can pasteurize everything, you have to prove it, it can’t fail. You can’t fail your pasteurization. However, if you can drop your water activity below 0.85, it’s in an environment now where nothing can grow even if it was there, even if you had a failure in your pasteurization process. That’s just science. I would probably feel more comfortable eating a product that was a little lower.
Question: In your evaluation of all of these products, with dried fruit, how come you didn’t measure brix values as well as pH?
Answer: We started out measuring Brix, and we got unusual answers, which I didn’t expect. Because knowing about osmotic effect and how sugar is a humectant and binds with water, I would have expected if we had a higher Brix, which means a higher sugar content, that we would have lower water activities. But our data weren’t showing that. Logically and scientifically, that’s how it should work. So we need to dig into that more, it’s a future project. It could be as simple as our process for determining the Brix, which is hard on a dried fruit. It’s easier to do beforehand, but it’s hard to do after the fact and that’s where all our fruit came from [we were reconstituting commercially purchased dried fruit]. I wasn’t 100% confident in how we were evaluating it, enough to make any kind of determination about that or include it in our data set.
Question: Final question here. Thanks again, for all of you who have who have contributed to these questions. They’re asking about about glycerin and cold pressed bars. Is that something that can affect the water activity?
Answer: Yes, 100%. Glycerin is a humectant so it works like sugar and salt. Humectants will bind with the water in the product so it’s not available for any spoilage. That’s how the water activity drops. But that moisture is still there, the water didn’t go anywhere. So you can have something that maintains its moisture content and at a lower water activity, in a safer zone. Glycerin definitely can be — dried cranberries use glycerin, and it helps with being a moisture barrier as well. There are vegetable sources glycerin too. There’s quite a range of different humectants you can use. I’m not an expert on that. I can’t give recommendations beyond the basic but an ingredient company could certainly help you try to find something that would fit what you’re looking for.