The Noom Diet: How It Works, What It Costs, and What the Evidence Shows

Noom is a subscription smartphone app, not a meal plan or a food-delivery service. Its core mechanic is a red/yellow/green food classification system based on caloric density — calories per gram, not calories per serving — so a food's color reflects how much energy it packs into a given weight rather than a moral judgment about the food itself. Leafy greens, most vegetables, and lean proteins fall into green; starches, whole grains, and lean meats fall into yellow; oils, sweets, fried foods, and calorie-dense snacks fall into red. Critically, nothing is banned: the stated target is roughly 30-45-25 percent green-yellow-red by calories per day, which functionally allows pizza or ice cream in moderation as long as the day's overall density balances out. Layered on top of the color system is a daily lesson feed — short, quiz-style modules covering topics like emotional eating triggers, all-or-nothing thinking, portion distortion, and habit-stacking, most running five to ten minutes. Users are also assigned a human coach (in practice, often shared across a large group of users with limited one-on-one bandwidth) who sends check-ins and can escalate to more frequent contact for an added fee. The app also tracks steps, weight trends, water intake, and sleep, and logs meals through a barcode scanner, photo-based recognition, and a large branded food database.

Noom's pricing has shifted multiple times since 2020 and continues to change, so treat any specific figure as a snapshot rather than a fixed price — always verify current pricing directly on Noom's site before signing up. As of Noom's own 2026 pricing pages, the core Noom Weight plan runs roughly $17 to $18 per month when paid annually in advance, up to around $70 for a month-to-month plan with no annual commitment — a meaningful spread depending on how the subscription is billed. Noom typically offers a low-cost trial period (often quoted around $0.50 to a few dollars for one to two weeks) before the recurring subscription begins, which is standard industry practice but has also been the specific point of friction behind most cancellation complaints, discussed below. Noom also sells Noom Med, a separate telehealth product that connects users with prescribers for GLP-1 medications like semaglutide or tirzepatide; that program is priced and billed entirely separately from Noom Weight, with reported costs ranging from roughly $70 for a coaching-only telehealth tier up to several hundred dollars per month once name-brand medication is included, largely dependent on insurance coverage. Because these numbers move with marketing cycles and promotions, the practical takeaway is to budget for the higher end of the range, confirm whether you are being enrolled in an annual or monthly plan before entering payment details, and check the specific auto-renewal terms — not the marketed trial price.

Most of the published research specifically on Noom's program comes from Noom's own in-house research team, and that needs to be stated plainly rather than buried in a footnote. A 2020 study in JMIR Diabetes by DeLuca, Toro-Ramos, Michaelides, and colleagues — several of whom are Noom or Noom-affiliated researchers — analyzed 14,767 adult users and found average weight loss of roughly 5-8% of body weight over 16 weeks, with older users losing weight more consistently than younger ones. An earlier 2016 study in BMJ Open Diabetes Research & Care, also led by Noom researcher Andreas Michaelides, followed 43 prediabetic adults through a 24-week Noom-delivered Diabetes Prevention Program curriculum and reported clinically meaningful weight loss tied closely to lesson engagement and coach message frequency. The most rigorous evidence to date is a 2026 randomized controlled trial published in Obesity Science & Practice, which followed 427 adults with overweight or obesity for 16 weeks of active coaching plus a year of follow-up; participants using Noom lost an average of 4.1% of body weight versus a 1.5% gain in a control group given standard U.S. Dietary Guidelines materials, with roughly twice as many Noom users hitting a clinically significant 5%+ loss threshold. That RCT design — a randomized control arm rather than a before-after comparison — is a genuine strength, but it's worth noting whether Noom funded or co-authored it before treating it as fully independent evidence, and to date no large-scale, fully independent (non-Noom-funded) randomized trial of Noom specifically has been published. Separately, real independent research on CBT-based weight interventions in general — not Noom by name — including a 2023 meta-analysis in Scientific Reports and prior reviews, consistently shows CBT techniques produce modest but real weight loss and durable improvements in eating-related psychological measures, which lends outside plausibility to Noom's underlying method even where Noom-specific data is company-funded.

Noom tends to suit people who have already tried strict, rule-based diets and burned out on them — the ones who can recite calorie counts for their favorite foods but still struggle with the underlying behavior patterns around stress eating, all-or-nothing thinking after a single dietary slip, or emotional eating triggers. Because the color system never technically forbids a food, people with a history of restrictive dieting or disordered eating patterns sometimes find the psychological framing less triggering than a hard-rule plan, though Noom is not a clinical eating-disorder treatment and should not substitute for one in anyone with a diagnosed eating disorder. It also tends to work better for people who will actually engage with the daily lessons; the JMIR and BMJ Open Diabetes studies both found engagement — lesson completion, logging frequency, coach message response — was the single strongest predictor of weight loss, more so than any individual food choice. Conversely, people who want firm, unambiguous food rules (a defined food list, a fixed macro target, a banned-ingredients list) often find the color-coded flexibility frustrating rather than freeing, since it requires ongoing self-regulation rather than removing decisions entirely. Noom is also a poor fit for anyone on a tight budget who only needs a calorie database without the coaching and lesson layer, and for anyone who wants meal plans or recipes built directly into the tracking flow — Noom's food guidance is deliberately general rather than prescriptive.

Functionally, both Noom and a free tracker like MyFitnessPal or Cronometer accomplish the same core mechanical task: they make you aware of what and how much you're eating, which is the single most consistently evidence-backed behavior change in weight management research regardless of the specific diet philosophy layered on top. The difference is what surrounds that logging. A free tracker gives you a database and a calorie or macro target and stops there — it assumes you already know how to interpret cravings, plan around a bad day, or recover from a binge without spiraling into abandoning the plan entirely. Noom's daily lessons and human coach exist specifically to address that gap, walking users through the psychological mechanics of why diets fail rather than just what to eat. The tradeoff is cost and pacing: Noom's lesson-based structure means slower onboarding (some users report weeks before reaching color-coded food guidance in depth) and a materially higher price than a free or low-cost tracker. For someone who already understands portion control and just needs a log, Noom's premium is arguably paying for content they don't need. For someone whose past diet failures were driven by behavior and mindset rather than lack of nutrition knowledge, the coaching layer is the actual product being purchased, not the food database.

Noom settled a lawsuit in 2021 (Noom Inc. agreed to pay $56 million to resolve a class action) over allegations that its trial-to-subscription conversion and cancellation process made it difficult for users to cancel before being billed for a long-term plan — a pattern commonly described as a 'dark pattern' in consumer-protection circles. Independent of that settlement, subscription cancellation friction remains one of the most frequently repeated complaints in user reviews across app stores and consumer review sites as of 2026, alongside complaints about auto-enrollment into annual plans when users believed they were signing up for a short trial. Noom has also drawn scrutiny for marketing claims that lean on psychology-forward language ('this isn't a diet, it's a lifestyle change') that some public health researchers and dietitians argue can overstate what a coaching app can realistically deliver compared to in-person registered dietitian or licensed therapist care, particularly for anyone with a clinical eating disorder or a complex medical condition. None of this means the underlying CBT-based method lacks merit — the independent meta-analytic evidence for CBT weight approaches generally is real — but the criticisms are specifically about business practices and marketing tone, not about whether the psychological framework itself is sound.

Read the cancellation terms before entering payment information, not after — specifically note whether you're being enrolled in a monthly or annual plan, and set a calendar reminder several days before any trial period ends. Engage with the daily lessons rather than skipping straight to food logging; the JMIR and BMJ Open Diabetes research both tie weight outcomes to lesson completion, not just calorie counting, so treating Noom as a bare-bones tracker forfeits the part of the product with the strongest evidence behind it. Weigh yourself under consistent conditions (same time of day, similar clothing) rather than reacting to daily fluctuations, since Noom's own trend-line weight graph is designed to smooth out normal day-to-day water-weight variance. If assigned a human coach, use the messaging feature actively in the first two weeks — engagement with coaching correlates with outcomes in the available research, and coach responsiveness varies, so an unresponsive coach is worth flagging or requesting a change. Finally, treat the green/yellow/red system as a density guide, not a purity test: research consistently shows all-or-nothing thinking about 'bad' foods predicts worse long-term adherence than flexible moderation, which is the exact psychological pattern Noom's lesson content is built to interrupt.

The behavior-change principles underlying Noom — CBT-based reframing, habit tracking, addressing emotional eating triggers — are not proprietary, and several no-cost or low-cost paths apply the same ideas. Free CBT-based workbooks and self-help programs adapted from clinical obesity treatment (such as those based on the Beck Diet Solution framework) walk through similar cognitive-restructuring exercises without a subscription. Cronometer and MyFitnessPal's free tiers provide the same caloric-density awareness Noom's color system builds on, simply without the daily lesson content; pairing either with a free mindful-eating or habit-tracking app can approximate Noom's structure at a fraction of the cost. Many public health systems and insurers in the US, UK, Canada, and Australia offer free or subsidized access to the CDC-recognized Diabetes Prevention Program (DPP) — the same curriculum framework Noom itself licenses and adapts commercially — through community health centers, employer wellness benefits, or the NHS Digital Weight Management Programme in the UK. A registered dietitian consultation, even a single session, can also translate the same portion-awareness and behavior-pattern concepts into a plan tailored to an individual's specific triggers, medical history, and food preferences — something no app-based color system does.