Most people following a ketogenic diet count net carbs, which means subtracting fiber and sometimes sugar alcohols from total carbohydrates. That approach works well for whole foods, where the fiber genuinely passes through your body without raising blood sugar. But the net-carb calculation gets unreliable once you start dealing with processed keto bars, sugar alcohols that vary wildly in how much they’re actually absorbed, and fiber additives that may not behave like the fiber in vegetables. The real answer is that both numbers matter, and knowing when to trust the subtraction and when to be skeptical can make or break your results.
What Net Carbs Actually Means
Total carbohydrates on a nutrition label include everything classified as a carbohydrate: sugars, starches, fiber, and sugar alcohols. Net carbs attempt to isolate the portion that meaningfully affects your blood sugar and insulin. The standard formula is simple: total carbs minus fiber, and sometimes minus sugar alcohols. The idea is that fiber resists digestion in the small intestine and sugar alcohols are only partially absorbed, so neither should “count” the same way a spoonful of sugar does.
In the United States, nutrition labels list fiber as a subcategory of total carbohydrates. In many other countries, including those using European labeling standards, the carbohydrate line on the label already excludes fiber. If you’re reading a label from the UK, Australia, or the EU and subtracting fiber again, you’re double-counting. Before you do any math, check where the product was made and what labeling convention it follows.
Why Fiber Gets Subtracted
The core logic is sound: dietary fiber is not broken down by human digestive enzymes in the small intestine. It arrives in the large intestine mostly intact. That’s where things get a bit more complicated, though, because not all fiber stays inert once it reaches the colon.
Insoluble fiber, the kind found in leafy greens, nuts, and the skins of vegetables, passes through largely unchanged. It adds bulk and keeps things moving. Soluble fiber, found in foods like avocados, flaxseeds, and certain legumes, dissolves in water and gets fermented by gut bacteria in the colon. That fermentation produces short-chain fatty acids, which your body does absorb and use for energy. Research in animal models has shown that soluble fiber fermentation can contribute a meaningful amount of energy through these fatty acids, to the point where it partially offsets the caloric “free pass” people assume fiber provides.
In practical terms for ketosis, though, the caloric contribution from short-chain fatty acids is modest. Studies estimating daily production in humans suggest the energy yield from colonic fermentation of fiber is relatively small compared to your total intake.
The bottom line for whole foods: subtracting fiber from total carbs is a reasonable approach. A cup of broccoli with 6 grams of total carbs and 2.4 grams of fiber isn’t going to spike your blood sugar from those 2.4 grams. But the calculation becomes less reliable when you move away from whole foods.
The Sugar Alcohol Problem
Sugar alcohols are where the net-carb formula starts to break down for a lot of people. These sweeteners, commonly found in keto candy, protein bars, and sugar-free chocolates, are partially absorbed in the small intestine. The key word is “partially,” and the percentage varies enormously depending on which sugar alcohol you’re eating.
Erythritol is the gold standard for keto dieters. About 90% of a dose is absorbed into the bloodstream but then excreted unchanged in urine, meaning it delivers close to zero usable calories and has minimal effect on blood sugar.
The larger sugar alcohols tell a different story. Maltitol, which shows up in many “sugar-free” chocolates and candy, has a glycemic index around 35 to 52 depending on the form, compared to about 65 for table sugar. That’s lower, but it’s far from zero. If you fully subtract maltitol from your carb count, you’re kidding yourself. Xylitol and sorbitol fall somewhere in between, with moderate absorption and moderate blood sugar effects.
Then there’s isomaltooligosaccharide, often listed as IMO or “tapioca fiber” on labels. For years, many keto product manufacturers used IMO and labeled it as fiber, allowing them to claim impressively low net carb counts. Research has shown this is misleading. A study comparing IMO to a true fiber (soluble corn fiber) found that IMO raised blood glucose and insulin significantly at 30 minutes, behaving more like a slow-digesting carbohydrate than a fiber. The researchers concluded that IMO should not be classified as a dietary fiber at all.
A practical rule of thumb for sugar alcohols on keto:
- Erythritol: subtract fully, as it has negligible impact on blood sugar or insulin
- Xylitol and sorbitol: subtract roughly half, since they’re partially absorbed and do contribute some glycemic effect
- Maltitol: subtract very little or not at all, as its glycemic impact is substantial
- IMO: do not subtract, as it behaves like a digestible carbohydrate despite sometimes being labeled as fiber
Allulose and Other Newer Sweeteners
Allulose has become increasingly popular in keto products over the past few years. It’s a rare sugar that exists naturally in small quantities in foods like figs and raisins. Unlike sugar alcohols, allulose is technically a sugar, but your body handles it very differently from glucose or fructose. It has close to zero calories (less than 0.4 kcal per gram) and a systematic review and meta-analysis found that both 5-gram and 10-gram doses significantly reduced the area under the curve for post-meal blood glucose in healthy people.
In the United States, the FDA ruled in 2019 that allulose does not need to be counted as a sugar or as an added sugar on nutrition labels, though it still appears under total carbohydrates. For keto purposes, allulose can be fully subtracted. If you’re buying a product that lists allulose separately, treat it like erythritol in your net carb calculation. The catch is that not all countries have made the same labeling decision, so imported products may count allulose differently.
Why Processed Keto Foods Deserve Extra Scrutiny
The explosion of keto-branded products has created a category of food where the net carb number on the front of the package can be wildly misleading. The issue isn’t just IMO masquerading as fiber. It’s that many of these products combine multiple ingredients that each have partial absorption, and the net carb formula assumes none of them count.
A keto bar might contain soluble corn fiber, maltitol, and some IMO, with a label claiming 3 grams of net carbs. In reality, the maltitol is partially raising your blood sugar, the IMO is acting more like starch than fiber, and the soluble corn fiber is being fermented into short-chain fatty acids that your body absorbs. The true glycemic impact could be meaningfully higher than 3 grams of actual sugar would produce.
Research testing proprietary low-carb, high-protein foods has confirmed that these products do produce significantly lower blood sugar responses than pure glucose. But “lower than glucose” is a very low bar. A food can produce a noticeable glycemic response and still technically beat glucose in a head-to-head comparison. The question for someone in ketosis isn’t whether a keto bar is better than a candy bar. It’s whether the bar’s real carb impact is low enough to keep you in the metabolic state you’re aiming for.
If you find yourself eating multiple processed keto products per day and struggling to stay in ketosis, the net carb numbers on those packages are the first thing to question. Switching to whole-food carb sources where the fiber subtraction is straightforward, like vegetables, nuts, and seeds, often resolves the problem.
Individual Variation Is Larger Than Most People Realize
Even if you calculate net carbs perfectly, the same food can produce very different blood sugar responses in different people. A 2025 study published in Nature Medicine tracked glycemic responses to various carbohydrate sources and found striking individual differences. People who were more insulin resistant showed post-meal glucose responses to potatoes and pasta that were roughly 150% to 180% higher than those in insulin-sensitive individuals.
The researchers also identified distinct “spiker” profiles. Some people had their highest glucose spikes after eating potatoes, and these individuals tended to be more insulin resistant with lower beta cell function. Others spiked most after grapes and tended to be more insulin sensitive overall.
What this means for keto is that your personal carb threshold for staying in ketosis may be different from the standard 20- or 50-gram target that gets repeated in popular advice. Some people can eat 40 grams of net carbs and maintain ketone production. Others get knocked out of ketosis at 25 grams. Your metabolic health, insulin sensitivity, activity level, and even the specific foods you eat all play a role. This is one reason why people who test their ketone levels, rather than relying purely on carb counting, often get better results. Blood ketone meters have been shown to have excellent reliability for diagnosing nutritional ketosis, making them a practical tool for figuring out your personal threshold.
The Gut Microbiome Trade-Off
One underappreciated consequence of counting only net carbs and driving that number as low as possible is what happens to your gut bacteria. On a standard ketogenic diet, fiber intake tends to drop significantly, simply because most high-fiber foods also contain carbohydrates. A systematic review of the ketogenic diet’s effects on gut microbiota found that keto has a generally “anti-microbial” effect, reducing overall bacterial diversity. All four studies in the review that tracked fiber intake confirmed it was significantly reduced on a ketogenic diet, and the reduction in carbohydrates reaching the colon for bacterial fermentation was identified as the primary cause.
This matters because gut bacterial diversity is linked to immune function, metabolic health, and even mental health. Research on ketogenic diet formulas found that dietary fiber, more than fat ratio or fat source, drove the largest shifts in gut microbial community structure. Adding fiber to a fiber-deficient ketogenic formula restored microbial diversity in ways that simply adjusting fat levels did not.
The practical takeaway: if you’re following keto and counting net carbs, don’t use the fiber subtraction as a reason to avoid fiber-rich foods. It’s the opposite. The subtraction is there to encourage you to eat those foods without worrying about the carb count. Low-carb vegetables, flaxseeds, chia seeds, and avocados are your best tools for keeping gut bacteria healthy while staying in ketosis. People who eat only meat, cheese, and oil on keto and hit very low net carb targets may be winning the carb battle while losing the microbiome war.
Total Carbs vs. Net Carbs for Medical Ketogenic Diets
If you’re following a ketogenic diet for epilepsy or another medical condition rather than for weight loss, the rules are stricter. Clinical ketogenic diets, particularly the classic 4:1 ratio used for drug-resistant epilepsy in children, are typically calculated using total carbohydrates, not net carbs. The precision matters because even small deviations can affect seizure control.
Interestingly, emerging research suggests that the fiber component of these diets may actually support their therapeutic effects rather than undermining them. A 2025 study in Nature Communications found that adding fiber to a low-fiber ketogenic formula not only restored microbial diversity but also restored seizure resistance in a mouse model. Supplementing fiber-containing ketogenic formulas with additional fiber actually enhanced their anti-seizure effects beyond baseline levels. The researchers found that this benefit was not replicated by simply giving short-chain fatty acids orally, suggesting the mechanism involves the interaction between fiber and gut bacteria rather than just the metabolic byproducts of fermentation.
For anyone on a medically supervised ketogenic diet, the decision of whether to count total or net carbs should be made with your medical team. But the science increasingly suggests that finding ways to include fiber, rather than restricting it, may improve outcomes even in the most restrictive therapeutic protocols.
Digestive Side Effects of Sugar Alcohols
Beyond the question of whether sugar alcohols affect ketosis, there’s the very practical matter of what they do to your stomach. Sugar alcohols that aren’t fully absorbed in the small intestine travel to the colon, where they draw in water through osmosis and get fermented by bacteria. The result, especially for people who aren’t used to them, can be bloating, gas, cramping, and diarrhea.
The severity depends on the type and amount. Erythritol is the best tolerated because most of it is absorbed before reaching the colon. Larger sugar alcohols like sorbitol, mannitol, and maltitol are more likely to cause problems because a bigger fraction reaches the large intestine undigested. A scientific review of gastrointestinal disturbances from sugar alcohols noted that tolerance varies significantly between individuals, that the molecular size of the sugar alcohol matters, and that while the resulting osmotic diarrhea shows no actual intestinal damage, it can be quite uncomfortable.
Most people develop tolerance with gradual exposure, but “gradual” is key. Eating an entire bag of sugar-free gummy bears (a famously unpleasant internet story) is not how you build tolerance. If you’re new to keto and loading up on products sweetened with xylitol, sorbitol, or maltitol, start with small portions and increase slowly. Or just lean toward erythritol and allulose, which cause fewer digestive issues at typical serving sizes.
When Total Carbs Might Be the Better Number
For some people in specific situations, tracking total carbs rather than net carbs is the simpler and more reliable approach. If you’re just starting keto and want to establish ketosis quickly, keeping total carbs under 20 grams per day virtually guarantees you’ll get there, regardless of any calculation errors. You can always loosen up to net carbs once you’ve confirmed you’re in ketosis and understand how your body responds.
Total carbs also make more sense if your diet includes a lot of processed low-carb products with sugar alcohols and added fibers whose real-world absorption you can’t easily verify. When you don’t trust the label, counting everything eliminates the guesswork. And for people with significant insulin resistance, where glycemic responses to even modest carbohydrate loads can be amplified, the conservative approach of counting total carbs provides a larger safety margin.
On the other hand, if your carbs come mostly from vegetables, nuts, seeds, and avocados, counting net carbs makes sense and frees you to eat a more varied, nutrient-dense diet. Someone eating a large salad with 15 grams of total carbs and 8 grams of fiber is not in the same metabolic situation as someone eating 7 grams of sugar. Treating those as equivalent by counting total carbs would unnecessarily restrict healthy food choices and potentially hurt your gut microbiome in the process.