Published 30th July 2026
ZOE’s Ultra-Processed Food Scanner: New microbiome research
Share this article
The ZOE Ultra-Processed Food (ZUPF) Scanner is one of the most useful features of the ZOE app.
It helps members understand the health impact of specific ultra-processed food (UPF) products and explains that not all UPFs are created equal.
ZOE recently conducted a study to put the ZUPF Scanner to the test, comparing it against the industry standard UPF definition: The NOVA classification system.
The results suggest that the ZUPF Scanner may be better at identifying foods that contribute to unfavorable gut microbial profiles than NOVA.
Currently, the most common way to classify processed foods is the NOVA classification system.
NOVA was designed to help researchers, and it works well for that purpose, but it was never meant to tell us anything about how a product affects health.
This is important because the latest research shows that some UPFs have a neutral effect on health, while others can even be beneficial (such as some high-fiber breakfast cereals).
Meanwhile, others (such as some sodas and processed meats) are strongly associated with poorer outcomes.
So, NOVA isn’t much use for people navigating the confusing modern food environment. This is where the ZUPF Scanner comes in.
We don't buy the hype — and neither should you
Our new app reveals what the food labels won't, using data from the world's largest nutrition study run by ZOE.
How does the ZUPF Scanner work?
The ZUPF Scanner is built into the ZOE app. To use it, all you need to do is snap a photo of a food product, and the results are generated in seconds.
While NOVA only considers the nature and extent of processing, the ZUPF Scanner takes into account factors that really matter for health, such as:
Energy intake rate: If the natural food structure (matrix) is destroyed, it’s easier to consume more energy in a shorter time. Foods with a high energy intake rate are typically soft-textured.
Hyper-palatability: Products that are specifically designed to be extremely tasty and difficult to stop eating.
The number and types of additives: Some additives, like vitamin C, aren’t a concern for health, but others, like certain emulsifiers, might be.
The ZUPF Scanner categorizes processed foods into four groups:
No-Risk: Minimal processing, which poses no risk to health.
Low-Risk: The processing has no or very low impact on health.
Medium-Risk: The processing may make it less healthy and could have a moderate impact on health.
High-Risk: The processing of this food makes it less healthy and could have a negative impact on your health.
Join our mailing list
Opt in to receive ongoing science and nutrition emails, news and offers from ZOE. You can unsubscribe at any time.
In a previous study, which you can read about here, we found that a higher intake of High-Risk foods was associated with poorer health markers, such as higher BMI and markers of inflammation.
We also found that consuming more No-Risk foods was associated with markers of better health, even though some of these foods would be considered UPFs under the NOVA classification.
In the latest study, ZOE’s scientists examined links between the ZUPF classifications and gut microbiome health. We recently presented our results at the Nutrition Society Congress at Newcastle University.
The latest ZUPF study
In this study, ZOE’s scientists explored processed foods, as scored by the ZUPF Scanner and NOVA classification, and their associations with the gut microbiome.
To do this, they used data from PREDICT 2 (549 participants) and PREDICT 3 (1,124 participants).
To assess gut microbiome health, we used our unique ZOE Microbiome Ranking system, which you can read about in more detail here.
In a nutshell, this system ranks species according to their association with markers of cardiometabolic health — both positive and negative. In other words, it ranks gut bugs from “good” to “bad.”
In our latest study, ZOE scientists investigated whether higher consumption of High-Risk and Medium-Risk foods was associated with a poorer profile of gut microbes (in other words, more bad bugs, fewer good bugs).
We also wanted to know whether the reverse was true: Is consuming more Low-Risk and No-Risk foods associated with more good, and fewer bad microbes?
Our scientists also looked at associations between good/bad gut bugs and UPFs as defined by NOVA. This was to see whether they were similar to or different from associations with High-Risk foods.
What did we find?
First, we examined participants’ consumption levels of UPFs (NOVA definition) and of ZUPF groups from No- to High-Risk. We found that UPFs accounted for:
41.5% of calories in PREDICT 2
29.2% of calories in PREDICT 3
However, ZUPF High-Risk foods accounted for just:
28.1% of calories in PREDICT 2
17.2% of calories in PREDICT 3
This shows that our ZUPF classification categorizes fewer foods overall as “highly processed” compared to NOVA.
We believe this is because the ZUPF Scanner uses a more precise classification system. So, rather than lumping all highly processed foods together, it better identifies which products are least healthy, while not penalizing those that are neutral or even beneficial for health.
But what does this mean for the microbiome?
ZUPF and the gut microbiome
Our analysis showed that intakes of High-Risk products were associated with 80 species.
When we investigated further, we found that individuals who consumed more High-Risk foods had higher levels of “bad” bacteria and lower levels of “good” bacteria, compared with those with lower intakes of High-Risk products.
We saw similar patterns among those with higher intakes of Medium-Risk foods, which were associated with 18 species (more bad species and fewer good species).
In contrast, higher intakes of No-Risk foods were generally associated with higher levels of “good” bacteria and lower levels of bad bacteria.
When we looked at bacterial species linked to higher UPF consumption, we found results very similar to High-Risk products: Higher UPF intake was linked to lower levels of good bugs and higher levels of bad bugs.
Of the 80 species linked to High-Risk foods, 64 were the same as those linked to UPFs.
Our new app gives you the power to see beyond the marketing
Make smarter, science-backed food choices in seconds. Scan. Score. Reveal the truth.
Why does this matter?
Our results show that not all UPFs are created equal. Some fall into the No- or Low-Risk groups and higher intakes of these foods were actually associated with greater numbers of good species.
However, some UPFs are linked to a poorer gut profile, as shown by the overlap between the High-Risk and UPF associations.
Importantly, despite the fact that High-Risk foods accounted for far fewer calories than NOVA UPFs, they still generated similar associations with participants' “good” and “bad” species.
Our results show that the ZUPF classification system could offer a more precise and specific way to identify the foods that may be driving the unfavorable associations between processed foods and our balance of “good” and “bad” gut microbes.
As we mentioned, NOVA is all about the type and amount of processing, while ZUPF takes into account other factors that influence a food’s likely impact on health.
This was one of the main reasons to design the ZUPF Scanner: We wanted to make sure we weren’t lumping perfectly healthy products into the UPF bucket.
In a nutshell, the ZUPF Scanner may more specifically identify foods that contribute to unfavorable gut microbial profiles. If you'd like to try the ZUPF Scanner, get started by completing this quiz now.
