The Hardware Margin Nobody Talks About: Why Smart Rings Cost $30 to Build and $350 to Buy

TL;DR

Smart rings sell for $300 to $500. The parts inside cost roughly $30. That 90% gross margin is not a secret, but nobody talks about it because the real product is not the ring. It is your biometric data stream, and the hardware exists only to collect it.

What $30 actually buys you

A smart ring is a stack of compromises. The enclosure must be thin enough to wear, tough enough to survive dish soap, and metallic enough to feel like jewelry. Inside that shell lives a Nordic Semiconductor nRF52-series BLE SoC, a multi-LED photoplethysmography sensor, a skin temperature thermistor, a 3-axis MEMS accelerometer, a small lithium-polymer cell, and a flexible PCB to connect them. Add a charging coil and the titanium or plastic outer body, and you have a device that fits on a finger and lasts three to five days.

None of these parts are exotic. The nRF52840, a common Bluetooth Low Energy chip with an ARM Cortex-M4, costs around $3 to $4 in volume. A decent PPG sensor module (the green and red LEDs plus photodiode that measure heart rate and blood oxygen) runs $2 to $4 depending on whether you want clinical-grade accuracy or fitness-grade approximation. The accelerometer is a commodity: $0.50 to $1.50. The battery, because it must be a curved or oddly shaped lithium-polymer cell to fit inside a ring, costs more per milliamp-hour than a cylindrical 18650, but still only $2 to $4. The flex PCB, titanium shell, charging case coil, and assembly labor add another $10 to $15.

Add it up. At 1,000-unit volumes, which is where most small wearable startups start, the bill of materials lands between $25 and $35. At 10,000 units, you can push that below $20. I am using $30 as a shorthand because that is where Pulsyn's numbers sit at our 500-unit pilot run, and I would rather be honest about small-volume pricing than pretend we are already at Xiaomi scale.

The enclosure is where companies hide costs they do not want to explain. Oura uses a titanium shell with a DLC coating. Samsung's Galaxy Ring is titanium with a different finish. Titanium rings in bulk cost $5 to $10 per unit machined, depending on whether you buy from a domestic job shop or a Shenzhen OEM. Pulsyn uses a medical-grade polymer for the first run, which drops the shell cost to roughly $2 but means we sacrifice the jewelry feel until we switch to metal at higher volume. That is a real tradeoff, and I will get into it later.

Why the retail price is ten times higher

If the parts cost $30, why does Oura charge $299 for the Gen 3, or $449 if you want the Heritage finish? Why does the Ultrahuman Ring Air start at $349? The gap is not profit in the way most people think. It is the cost of building a company that sells data infrastructure disguised as a wellness device.

First, the obvious overheads. FCC certification for a Bluetooth device in the United States runs $8,000 to $15,000 if you use a certified lab and do not fail the first time. CE marking for Europe is similar. Biocompatibility testing (ISO 10993) for a device that touches skin 24/7 adds another $5,000 to $10,000. Firmware development, app development, cloud infrastructure, customer support, returns, and chargebacks absorb margin faster than most backers realize. A hardware startup at seed stage is usually burning $20,000 to $50,000 a month before shipping a single unit. Spread that across a few thousand rings and you are adding $10 to $30 of overhead to every unit before it leaves the factory.

Those are real costs, and they explain why a $30 BOM does not translate to a $35 retail price. But they do not explain a $350 retail price. The rest of the markup is the subscription business model, and that model only works if the hardware margin is large enough to subsidize the real product: the cloud platform that ingests, processes, and resells insight into your body.

Oura's subscription is $5.99 per month, or $72 per year. Whoop charges $30 per month with no device purchase option. Fitbit Premium is $9.99 per month. These subscriptions are sold as "advanced analytics" or "personalized coaching," but the unit economics reveal what is actually happening. A SaaS customer with a 12-month lifetime is worth $72 to $360 in recurring revenue. The hardware margin on a $349 ring with a $30 BOM is $319. If you add the first year of subscription revenue, Oura extracts roughly $390 from a customer in year one. The hardware is the customer acquisition tool. The subscription is the business.

This is not a conspiracy. It is standard consumer hardware economics, and it works the same way in printers (cheap inkjet, expensive ink), game consoles (subsidized hardware, expensive games), and smart home cameras (cheap doorbell, expensive cloud storage). The difference is that your heart rate variability and sleep stages are not printer ink. You cannot buy them back.

The subscription is not an add-on. It is the business model.

I have talked to investors who ask why Pulsyn is "leaving money on the table" by not charging a subscription. The question assumes that subscription revenue is free money, and that the only reason to skip it is naivety. The honest answer is simpler: Pulsyn's unit economics do not require it, and requiring it would change what we are building.

At a $30 BOM and a $160 retail price, Pulsyn's gross margin is 81%. That is lower than Oura's roughly 90% because we are not charging $349, but it is still higher than most physical consumer goods. Coffee beans retail at a 50% markup. Sneakers at 60%. An 81% gross margin in hardware is unusual, and it exists because smart rings are small, the components are cheap, and the assembly is straightforward. The ring is not where the complexity lives. The complexity lives in the signal processing, the algorithms, and the user experience, all of which we are doing on-device or with optional cloud AI that the user controls.

The subscription model in this industry exists because companies need to justify venture capital valuations. A hardware company with one-time sales and an 85% gross margin is a good business, but it is not a SaaS business, and SaaS multiples are higher. A $10 monthly subscription turns a $160 hardware sale into a $280 lifetime value customer over two years. That lets you raise a Series A at a higher valuation, spend more on customer acquisition, and grow faster. The problem is that the customer does not benefit from the subscription in any way that a well-designed on-device algorithm cannot replicate. You are paying for cloud infrastructure that exists to make the company investable, not to make your sleep better.

Whoop is the most honest version of this model. They do not pretend the hardware matters. You cannot buy a Whoop strap; you lease it with a subscription. The device is free because the device is irrelevant. Whoop is a data analytics company that mails you a sensor pack. Oura is slightly less explicit: you buy the ring, but without the subscription you get a fraction of the features. Fitbit is the most confused: they sell you hardware, then gradually move features behind a paywall that did not exist when you bought the device.

All three approaches share the same DNA. The hardware is a loss leader or a low-margin acquisition channel. The data is the product. You are the supply chain.

What Pulsyn is doing differently

Pulsyn's BOM is $30. We sell the ring for $160 on Kickstarter and $200 at retail. That gives us a gross margin of 81% to 85%, which is enough to cover certification, firmware development, customer support, and a small team without charging rent on your sleep data.

The way we make that work is by removing the biggest cost most wearable companies pretend does not exist: the cloud platform. Oura, Whoop, and Fitbit all run massive backend systems that ingest every heartbeat, every SpO2 reading, every temperature fluctuation, and process it on their servers. That infrastructure is not free. AWS or GCP bills for a wearable company with a million users run into the hundreds of thousands of dollars per month. Those bills are passed to the user as "personalized insights" and "trend analysis."

Pulsyn processes everything on the phone. The ring sends raw sensor data over BLE. The phone runs the signal processing, the sleep staging, the HRV analysis, and the readiness scoring. No cloud required. The optional premium tier adds cloud AI for users who want deeper context windows or longer trend history, but the base product is fully functional without it. We are not "leaving money on the table." We are refusing to build a table that charges rent.

There are tradeoffs. On-device processing limits how much historical data we can store. It means the AI model must be smaller. It means we cannot push algorithm updates to every user instantly without an app release. I am not sure if these limits will matter in two years. On-device chips are getting faster, model quantization is getting better, and users are getting tired of privacy policies they cannot read. For now, the tradeoff feels correct.

The hardware itself is also simpler in ways that matter. Pulsyn does not use titanium for the first run. We use a medical-grade polymer that costs $2 instead of $8. We do not have a wireless charging case with an LED battery indicator. We have a USB-C dock that costs $4 to make. We are not building jewelry. We are building a sensor that happens to be wearable, and the design reflects that honesty. When we move to metal at higher volume, the price will stay the same because the margin will absorb the extra cost.

The honest limits of cheap hardware

I want to be clear about what a $30 BOM does not buy. It does not buy perfect quality control. At 500 units, our Shenzhen manufacturer is doing manual visual inspection on flex PCBs instead of automated optical inspection. That means some units will have LED alignment issues that affect PPG accuracy. We have budgeted a 5% return rate, which is high for consumer electronics but realistic for a first hardware run from a new supplier.

It does not buy infinite battery life. The nRF52840 plus a PPG sensor running every few minutes will drain a 20mAh curved cell in roughly 72 hours. That is worse than Oura's claimed 7 days and worse than RingConn's 5 to 7 days. We are trading battery life for on-device compute efficiency and component cost. The tradeoff is real, and some users will care.

It also does not buy instant scale. If Pulsyn sells 10,000 units, our per-unit BOM drops to roughly $22. If we sell 100,000, it drops to $18. But getting from 500 to 100,000 units requires capital, inventory risk, and manufacturing relationships that take years to build. I am not claiming we have solved hardware scaling. We have not. We are at the very beginning of that curve, and the $30 number reflects that honesty.

Finally, a low BOM does not mean a low total cost to the user if the device is disposable. Smart rings are sealed units. The battery degrades. The sensors drift. Most users will replace their ring every two to three years, which means the effective annual cost of a $160 ring is $53 to $80. That is still cheaper than Oura ($299 plus $72 per year, or $123 effective annual cost in year one, $72 every year after), but it is not free. Hardware wears out. We are not pretending otherwise.

About the author

James Hoffmann is the founder of Pulsyn. He has been building the Rune 1 smart ring in Phoenix, Arizona, with a supply chain that spans Shenzhen and Tucson. He writes about hardware economics, on-device privacy, and the wearable industry at getpulsyn.com.

References

1. Nordic Semiconductor nRF52840 product specification, nordicsemi.com. The chip is listed at $3.50 to $4.20 in 1,000-unit volumes through authorized distributors.
2. Maxim Integrated (now Analog Devices) MAXM86161 PPG module datasheet. Reference pricing for clinical-grade optical heart-rate modules in wearable form factors, 2024.
3. Oura Ring Gen 3 pricing and subscription terms, ouraring.com, accessed May 2026. Hardware starts at $299; monthly subscription is $5.99.
4. Whoop membership pricing, whoop.com, accessed May 2026. No hardware purchase option; subscription is $30 per month with a 12-month minimum.
5. Samsung Galaxy Ring launch pricing, samsung.com, 2024. Initial retail price was $399.99 in the United States.
6. ISO 10993-5 biocompatibility testing estimates from Pacific BioLabs and Toxikon, 2024-2025 pricing for skin-contact medical devices.
7. FCC Part 15 certification cost estimates from MET Labs and Element Materials Technology, 2024-2025.