The 2026 Wearable Market Is Splitting in Two, and the Health Data Is Going With the Premium Half

Three numbers from the same summer. Garmin’s fitness revenue grew 25% in the second quarter at a 64% gross margin [5]. In that same quarter, Counterpoint counted global smartwatch shipments down 4%, the first contraction in a year, and Omdia counted wrist-worn wearables down 2% overall, with basic bands down 9% and advanced smartwatches up 6% [19][20]. And two out of every three wearables shipped in the first quarter were earbuds [1].

None of these is wrong and none of them contradicts the others. They describe a market that is no longer growing or shrinking as one thing. It is splitting: by price, by device category, by region, and by who the buyer is. And the part of it that produces the health data everyone is building products on is not the part that is growing.

Split one: price

Counterpoint’s full-year 2025 report is the clearest single statement of the divide. Global smartwatch shipments grew 4%, the first growth since 2023. Underneath that, shipments of watches priced below $200 fell 9% and shipments in the $200 to $400 tier rose 48%. The average selling price rose 5% [2]. In the first quarter of 2026 the pattern continued: units up 4%, price up 6%, with Apple growing 21% and taking 23% of the market [3].

The second quarter, reported in August, sharpened it. Counterpoint has global smartwatch shipments down 4%, the first year-over-year contraction in a year, and attributes it to premium owners holding their watches longer while the basic segment keeps shrinking; Apple still grew 14%, and China took a record 38% of the market [20]. Omdia, which splits the wrist into advanced smartwatches, basic watches and basic bands, has the whole wrist down 2% but advanced smartwatches up 6%, basic watches down 3% and basic bands down 9%. Garmin’s share of the advanced segment rose from 11% to 15% in a year, the largest gain of any vendor outside Apple, and screenless trackers passed 15% of basic band shipments [19]. Omdia’s own reading is that demand has moved to two poles, minimalist screenless trackers and feature-rich sports watches, with the mid-tier left behind [19]. That is the split in a single quarter’s data, and it is the same split the full-year price tiers show.

The company results say the same thing from the inside. Garmin, whose business is the premium tier, reported fitness revenue up 25% on demand for its advanced wearables, at a 64% gross margin and a 37% operating margin for the segment [5]. Oura, which sells a ring at $399 to $499 and a membership on top, generated $311 of revenue per ring sold and grew paid members 100% year over year to 5.0 million [6][7].

The bottom of the market is where the decline is. India is the cleanest case because it is the largest market where the entry tier dominated. IDC counted 114.2 million wearables shipped there in 2025, down 4.0% and the second consecutive annual fall. Smartwatches fell 17.6% to 28.9 million units. The average wearable sold for $20.30 [4]. IDC expects Indian smartwatch shipments to decline again in 2026 by a mid-single-digit percentage as brands shift toward higher-value products [4]. The brands are not wrong to do that. It is just that the higher-value product is bought by a different person.

The premium brands’ own answer to the entry tier tells you where they think the money is. Garmin’s $199.99 CIRQA band and Google’s $99 Fitbit Air both launched this summer, both without a screen, and both feed a $6.99 or $9.99 subscription [17]. The cheap device has become the free sample for the insight layer, which we priced out in the 2026 price map.

Globally, IDC’s forecast has wristbands, the cheapest health-capable form factor, falling 6.8% in 2026 to 40.2 million units, squeezed from above by mid-tier smartwatches and from the side by rings [1].

One honest caveat about the headline numbers. IDC forecasts a 2.8% decline in smartwatch units for 2026, citing memory-related supply constraints that keep prices elevated and soften replacement demand in price-sensitive segments [1]. Counterpoint began the year expecting high-single-digit growth [2] and, after the second-quarter contraction, now expects about 1% [20]. Omdia’s 6% growth for smartwatches and Counterpoint’s 4% decline are not a contradiction either: Omdia counts only advanced watches as smartwatches and files the shrinking basic watches separately, which is also why Apple is 46% of Omdia’s smartwatch market and about a fifth of Counterpoint’s [19][20]. Move the category boundary and the headline moves with it. The one reading that survives all three trackers is the shape: entry tier weak, premium tier strong, average price rising. A market can grow or shrink a few percent in units and still be splitting.

Split two: category

IDC’s 2026 forecast, category by category [1]:

Add the growth rows together and the picture is stark. The two categories growing fastest, glasses and hearables, are about two-thirds of all units and carry almost no health data. The two categories carrying the full health stack, smartwatches and rings, are about 26% of units, and the larger of them is forecast to shrink. In IDC’s forecast, hearables add more incremental units than every other category combined, while smartwatches and wristbands subtract units this year [1].

The glasses number deserves a closer look, because it is where the industry’s attention is going. IDC counted 2.25 million display-less smart glasses in Q1 2026, up 167% year over year, with Meta at 69.2% share, and forecasts 27.3 million a year by 2030 [9]. That is a real consumer category being born. It just is not a health category. IDC’s own smart glasses analysis contains no mention of health sensing at all [9].

The hearables number is subtler. AirPods Pro 3, launched September 2025, measure heart rate during workouts with an in-ear optical sensor and write it to HealthKit [10]. That is genuine health data from the largest wearable category on earth. But it is heart rate during exercise, not continuous, and developers querying HealthKit report no HRV samples from the earbuds [10]. An earbud is worn for the commute, not the night, and the night is where sleep, resting heart rate, HRV and temperature live. As a health sensor, two-thirds of the wearable market is a workout heart rate monitor at best.

Rings are the one category that is both growing and rich in data, and the base is tiny. IDC’s 4.9 million units in 2026 is about 3% of smartwatch volume. Oura, which held roughly 74% of ring shipments in the first half of 2025 per Omdia [11], sold 3.6 million rings in the twelve months to June 2026 and describes that as “approximately 2% penetration” of a 212-million-unit market [6]. Note that Oura’s 212 million excludes earbuds and glasses. It is roughly IDC’s smartwatches plus wristbands plus rings, which is the honest way to size the market for health data, and it is a third of the headline “wearables” figure.

Split three: geography

The regional split is the mirror image of the price split, because the entry tier and the premium tier live in different countries.

China is the growth. Smartwatch shipments there rose 15% in Q1 2026, with Huawei holding about 40% of the market, helped by a national electronics subsidy scheme and Huawei’s push into sleep, emotional wellbeing and arrhythmia features [3][16]. In Q2, with the global market shrinking, China grew another 7% and took a record 38% of worldwide smartwatch shipments, and Huawei’s 22% global share put it ahead of Apple for the quarter [20]. IDC’s China data has wrist-worn shipments at 18.14 million in the quarter, with adult smartwatches up 15.3% and children’s watches up 22.4% [16]. Counterpoint credits China with leading the global market’s return to growth in 2025 [2].

India is the decline, for the reasons above: a saturated entry tier, a $20 average price, and two consecutive annual falls [4].

The United States is the ceiling. 46% of adults own a wearable, and it is the only large market where a national survey still shows steady growth [8]. Britain, measured by YouGov, has been flat at 35% to 37% since early 2024. We compiled the national survey figures for eight markets in smartphone vs wearable adoption by region; the gap between smartphone reach and wearable reach is at least 45 percentage points in every market with reliable data, and the US is the narrowest gap on earth, not the typical one.

The regional split has a practical consequence for anyone with a global user base: the assumption “our users have wearables” is a US assumption, and even in the US it is a minority assumption.

Split four: the buyer

The survey and filing data describe who is on each side.

Rock Health’s December 2025 survey of 8,000 Census-matched US adults found 46% owning a wearable and 57% owning at least one wearable or connected device such as a smart scale, blood pressure cuff or CGM [8]. Its description of owners versus non-owners is the finding that matters: device owners skew younger, wealthier, more urban, healthier and more likely to be commercially insured, while adoption is consistently lower among adults over 55, lower-income households and rural residents [8]. About 22% of owners have more than one device. Among owners, 59% have discussed their data with a clinician [8].

Oura’s S-1 is the first time a premium wearable company has published its own member demographics, and they refine the picture rather than overturn it. About 72% of members are women. About 31% are 29 or younger, 42% are 30 to 45, 27% are 46 or older. About 37% report household income under $100,000, which means roughly 63% are above it, and therefore above the national median household income [18]. More than half report at least one chronic condition. Fifteen percent of US members work in healthcare [6].

So the premium tier is not the caricature of the Silicon Valley biohacker. It is disproportionately female, largely under 45, largely above-median income, and heavily populated by people who already manage a condition or work in the health system. That is a population that is already engaged with its own health. It is the population health programs find easiest to reach and least necessary to reach.

The people the data says are not buying are the ones every prevention argument is about: older, poorer, rural. And the device category that used to reach them on price, the sub-$200 watch and the wristband, is the one that is contracting.

Where this collides with policy

Three governments have spent the last eighteen months writing wearables into health policy on the assumption of broad reach.

In June 2025 the US Secretary of Health and Human Services said his vision was that “every American is wearing a wearable within four years” and promised one of the largest advertising campaigns in the department’s history [12]. As of this writing we can find no evidence that a national campaign has aired. The arithmetic is unforgiving regardless. Rock Health’s 46% leaves 54% of adults to reach before mid-2029. IDC’s forecast for the entire world’s smart ring production in 2026 is 4.9 million units, which is under two percentage points of the US adult population even if every ring went to an American [1][18]. The smartwatch category, the only one large enough to matter, is forecast flat to down. Reaching the remaining 54% with the current product mix would require the entry tier to reverse a two-year collapse, and the entry tier is being abandoned by its own manufacturers because the premium tier is more profitable.

CMS’s ACCESS model, which began paying for technology-enabled chronic care on 1 July 2026, is more careful than the coverage of it suggests. The payment document does not pay for wearables. It pays outcome-aligned amounts of $180 to $420 per beneficiary per year depending on the clinical track, half of it withheld until outcomes are met, and adds a $15 fixed payment for rural beneficiaries “to offset higher operational costs related to connected device distribution and support” [13]. Fifteen dollars is a considered estimate of the cost of getting a connected device to a rural Medicare patient. It is not a ring.

England’s ten-year plan is the one that engages with the split directly: it commits to wearables as a standard part of chronic and post-acute care by 2035 and to providing devices free in the areas of highest health need and deprivation [14]. That is the correct diagnosis. The people who most need passive monitoring are the people the market is least likely to sell a device to, so the state proposes to buy it. Whether a health service can distribute and support hardware at the scale the market is walking away from is an open question, but at least the question is the right one.

The same tension is what STAT identified in August in the insurer programs modelled on auto telematics: rewards for wearable-verified behaviour flow to the people who already own the wearable, who are already healthier, wealthier and better insured [15]. Whatever such a program measures, it measures it in the premium half.

What a split market means if you build on it

The temptation is to read the headline, “wearables up 4%,” and plan for a growing addressable base. The data says to plan for two bases that are moving in opposite directions.

1. Size the wearable-connected share of your users honestly, and measure it as a KPI. In the US it is at most 46% of adults, less among the populations most health products claim to serve, and it varies by income and region in ways Rock Health has documented [8]. In Britain it has not moved in two and a half years. In India it is a shrinking single-digit share of a $20 device base. Track the connected share in your own product by cohort, because it is the denominator of every wearable-dependent feature you ship.

2. Build for two data tiers, not one. The premium minority carries a ring or a $400-plus watch: rich sleep staging, HRV, temperature, ECG, and increasingly a licence agreement that governs what you may do with it. We have written about how those licences now treat AI; they are getting narrower, not wider. The majority carries no device. What it does carry is a phone, which records steps, movement, phone use, and on both platforms a usable sleep estimate, without any wearable at all. A product that only works for tier one has chosen to work for a minority that is not growing quickly.

3. Do not confuse the growth categories with health data. If a plan depends on “wearable adoption rising,” check which wearables. Earbuds and glasses are where the units are going, and they produce a workout heart rate at best and nothing at worst. The category with the richest data, rings, ships fewer units in a year than Apple ships watches in a quarter [1][3].

4. If you run an incentive program, the selection effect is the program. Rewards for wearable-verified activity reach the insured, the urban, the under-45 and the already-engaged, because that is who owns the verifier [8]. A program that wants to reach the other half has to be able to verify from the device that half already carries.

5. The premium half pays for insight, and will keep paying. Oura’s members renew at 85% after twelve months and 63% start on the annual plan [6][7]; the whole interpretation layer has converged on $6 to $10 a month. Willingness to pay is real. It is also concentrated, and concentrated where the marginal health gain is smallest.

Where we sit

Sahha reads both halves. Four of the connections in our integration list run against a device maker’s cloud API and the rest read what a user has already routed into Apple Health or Health Connect from a device they own. Alongside that, the SDK reads the phone itself, and for the majority of any customer’s users who own no wearable, the phone is the only sensor there is.

We are not neutral about this, so read the argument with that in mind. But the shipment numbers are not ours, and they say what they say: the health data that products are built on is being produced by a premium minority that is growing slowly, while the cheap end of the market that once promised to reach everyone is contracting. The people a health product most wants to reach will not be arriving with a ring. They already arrived with a phone.

The short version

IDC forecasts wearables up about 2% in 2026 to 626 million units, with hearables at 65% of volume and smart glasses up 41%, while smartwatches fall 2.8% and wristbands 6.8%. In Q2 2026 Counterpoint counted smartwatch shipments down 4%, the first contraction in a year, and Omdia counted basic bands down 9% against advanced smartwatches up 6%. Counterpoint has sub-$200 smartwatches down 9% and the $200 to $400 tier up 48% over 2025, with average prices rising 5% to 6% a year. India fell 17.6% in smartwatches; China took a record 38% of the world’s. The devices carrying full health data are about a quarter of units, and their buyers are younger, wealthier, more urban, healthier and better insured than the people who do not buy them. Policy written for “every American” is colliding with a market that has stopped making the device that could reach them. The phone is the sensor the other half already has.