How ONDA scores tools

Agreement of measured HRV (RMSSD) with an ECG chest-strap reference, drawn from peer-reviewed validation studies and side-by-side testing.

Sensor type (optical PPG vs electrical ECG), sampling approach, and how well a clean signal holds during movement and overnight.

How closely sleep staging and total sleep time track polysomnography or validated references — HRV is read mostly during sleep.

Access to raw and detailed data: export, an open API, and integration with third-party tools rather than a closed app.

Comfort and form factor for 24/7 wear, plus battery life and how much charging interrupts continuous measurement.

Clarity of the app, quality of the guidance, and whether metrics are explained rather than reduced to one opaque score.

Hardware price together with any mandatory subscription, weighed against what the device delivers.

Breadth, depth and quality of the guided library — meditations, courses and series across topics and levels.

Calibre of the instructors and the teaching itself — how clearly and credibly practice is taught, from beginner to advanced.

How well the app adapts to you — structured courses, progress tracking, recommendations and daily guidance.

Clarity and friction of the app itself — interface, navigation and how easily a session starts.

How much genuine practice is possible without paying — a usable free tier versus a locked product tour.

Subscription price weighed against what the library and the teaching deliver.

How far the app is grounded in research and credible teaching lineages rather than vague wellness claims.

How accurately the app measures sleep — timing, stages and disturbances — by accelerometer, microphone, sonar or a paired wearable.

Breadth and quality of content for falling asleep — soundscapes, sleep stories, guided audio and white noise.

Grounding in real sleep science — circadian methods, CBT-I, smart-wake timing — rather than vague relaxation claims.

Quality of the analysis the app gives back — trends, explanations and actionable recommendations, not just a score.

Clarity and friction of the app — interface, navigation and how easily it fits a nightly routine.

How much is genuinely usable without paying — a working free tier versus a locked sample.

Price — one-time or subscription — weighed against what the app delivers.

Cooling and heating range delivered (e.g. 13–43 °C at the bed surface), dual-zone capability, recovery time after a temperature change, and how the system holds the target through real ambient swings — the criterion the whole category exists to serve.

Hardware build quality (cover, hub, water hose / air duct), install footprint, water-tank or filter maintenance burden, and the multi-year reliability track record on the actual hardware shipped today.

Quality of the companion app, sleep-staging and HRV tracking on the device, integration with Apple Health / Oura / Whoop and the depth of insights surfaced — sleep-climate is increasingly bought as a tracker as well as a heater.

Cover vs full-mattress vs blanket, fit on existing beds, dual-zone (his/her) capability, and whether the system survives travel or rentals — install flexibility is the second-most-asked question after cooling.

Whether ongoing membership is required for full features (Eight Sleep), how locked the device is to the manufacturer’s app, and the realistic 3-year total cost of ownership. Eight Sleep’s subscription model is the category’s biggest editorial point of contention.

Hardware price plus subscription weighed against climate capability, build longevity and tracking depth. The category spans $500 (BedJet) to $5,000+ (Eight Sleep Pod 4 Ultra) — value is a real differentiator.

Quality and quantity of peer-reviewed studies on the device itself — not the vagus nerve in general — plus FDA / CE clearances and the conditions they cover.

What the device does to the vagus nerve — electrical (transcutaneous tVNS at the ear or neck), implanted, vibrotactile or infrasonic — and how transparent its dosing parameters are.

Range and customisability of programmes — session lengths, intensities, modes (sleep, stress, focus, recovery) and whether parameters are documented or hidden behind a black-box app.

Form factor, skin tolerance and how realistic daily or session-based use is — handheld, ear clip, neck collar, chest pebble or wearable patch.

On-device or paired HRV measurement, session logging and integration with health apps — does it just stimulate, or close the loop with measurable output.

Hardware price together with any mandatory subscription or prescription, weighed against the evidence and what the device delivers.

Mean absolute relative difference (MARD) of the underlying sensor versus reference plasma glucose, sensor lifespan, drop-out and warm-up behaviour. Most consumer CGMs ride on either Abbott Libre 3 or Dexcom G7 — both well-validated, with G7 marginally more accurate in independent comparison.

How the app interprets glucose curves — meal scoring, food-by-food impact, time-in-range, fasting and overnight metrics, AUC and variability. This is the category’s real differentiator and the metric biohackers are paying for.

Form and quality of human or AI coaching included — registered dietitian, certified coach, AI agent or none. Substance of the guidance, not the marketing claim.

Clarity of the app itself plus integration with the wider stack — Apple Health, Google Fit, Oura, Whoop, MyFitnessPal — so the CGM data composes with other biomarkers rather than living in a silo.

Subscription terms, ability to pause or buy single sensors, whether raw glucose data can be exported, and how easy it is to leave without losing access to historical insights.

Total monthly or annual cost — hardware, sensors and any coaching subscription — weighed against insight depth and the credibility of the coaching layer.

Sensor type and count (EEG channels, fNIRS optodes, tDCS electrode pairs), sampling rate, dry vs wet electrodes and how well a clean signal holds during a real session. For non-measurement devices (tDCS) this scores stimulation parameter transparency.

Breadth and depth of guided sessions — meditations, focus exercises, neurofeedback protocols, clinical programmes — and whether the experience evolves with the user over months of use.

How the app interprets the brain signal — EEG band breakdowns, focus and calm metrics, session-by-session progression, sleep-stage analysis where applicable. The translation from raw signal to actionable feedback.

Headset weight, fit, electrode pressure and how realistic 20–60 minute sessions or overnight wear is in practice.

Clarity of the app, friction of starting a session and whether data integrates with Apple Health, Google Fit and other wearables.

Raw-signal export, SDK availability, third-party app ecosystem and how locked the user is into the manufacturer’s interpretation. Biohackers and researchers value this disproportionately.

Hardware price together with any required subscription, weighed against signal quality, content depth and developer access.

Measured power density (mW/cm²) at the manufacturer-specified treatment distance, cross-referenced against independent meter readings rather than marketing-stated peak figures. The single biggest source of consumer confusion in this category.

Spectrum coverage — most consumer panels combine red 660 nm (skin) with near-infrared 850 nm (deeper tissue); some add 630, 810, 830 or 940 nm. The mix and the published peaks matter more than the LED count.

Hardware build quality, independent EMF measurements at typical treatment distance, and driver flicker rate. Cheaper panels often skip the driver hardware that keeps both clean.

Treatment area for the panel size — full-body, half-body, targeted or handheld — and ergonomics for unattended sessions (stand, hanging hardware, door mount).

Quality of the published photobiomodulation literature backing the device’s claimed indications, FDA registration / Class II clearances where applicable, and how honestly the manufacturer represents what its panel does versus the published evidence.

Hardware price weighed against measured irradiance × coverage area × build quality. Premium pricing is acceptable when independent verification matches; thin when it does not.

How cold the unit can hold (target range, minimum temperature), how fast it recovers after a plunge, and how it behaves in summer heat. The single criterion the category exists for.

Tub materials and insulation, exterior finish, expected lifespan and warranty terms. Cold-plunge hardware lives outdoors or in damp environments — the build either holds up for a decade or starts failing in year two.

Ozone or UV sanitisation, filter quality and how often the water actually needs changing in practice. The maintenance pattern that determines whether the plunge stays usable for daily protocols.

Indoor vs outdoor, footprint, install requirements (power, drainage, level surface) and how realistic the device is in a typical home setup.

How honestly the manufacturer represents cold-exposure benefits relative to the published research, and whether the included protocols are grounded in real science or in marketing language.

Total cost of ownership including install, electricity and water — weighed against chiller capability, build longevity and the realistic daily-use pattern.

Traditional convection vs full-spectrum IR vs near-IR incandescent — and within IR, which emitter type (carbon, ceramic, incandescent) and what wavelength bands the device actually puts out. The single biggest source of consumer confusion in the category.

Cabin wood (cedar, hemlock, basswood, eucalyptus), joinery, insulation and expected lifespan. Sauna hardware that holds up matches the wood to the use case rather than to the marketing budget.

Independently measured EMF at the seated treatment position, with documentation. Cheap IR saunas can run high EMF without the user knowing; the premium brands publish numbers.

Cabin (1 / 2 / 4-person), portable, blanket or barrel — and the install requirements for each (power, footprint, ventilation, outdoor weather rating).

Quality of the published sauna and IR-therapy literature backing the device’s claimed benefits, FDA Class II where applicable, and honest representation of indication versus marketing language.

Hardware price plus install plus annual electricity — weighed against heat-source quality, build longevity, EMF profile and form factor fit.

Measured magnetic field intensity at the treatment surface (microTesla / Gauss), pulse-rate range, and how documented the dose is. Marketing-stated peaks are common; usable continuous output is what matters and is rarely published.

Whether the specific waveforms and frequencies used are tied to published research (Bemer biorhythmic pulse, FDA-cleared bone-healing waveforms, Schumann-resonance frequencies) or are proprietary patterns without independent backing.

Hardware build, control-unit electronics, coil quality, included applicators and warranty. PEMF hardware lives 5–10+ years when built right; cheap mats degrade in 12–24 months.

Range and customisability of programmes — preset protocols, intensity steps, session lengths, app control where present. Black-box single-button devices score lower than systems exposing real parameters.

Full-body mat vs coil/loop applicator vs wearable patch — and the range of included applicators (whole-body mat plus pillow plus spot applicator vs single mat only). Form-factor flexibility drives real-world session frequency.

Hardware price weighed against measured field strength, waveform research backing and applicator range. The category spans $299 (wearable) to $10,000+ (clinical coil systems) — value gap is large.

Breadth and depth of guided breath sessions — count, lengths, themes (sleep, focus, energy, anxiety, performance) and how the library evolves with use over months.

Range of documented breathwork modalities covered — box breathing, 4-7-8, Wim Hof rounds, Tummo, holotropic, Buteyko, coherent breathing, cyclic sighing — vs single-technique apps.

How far the app cites and follows published research (Stanford cyclic-sighing, polyvagal, Buteyko CO₂ tolerance) rather than vague wellness claims, and whether instructor credentials are real.

Clarity and friction of the app itself — interface, visual or audio guidance, session-start friction, Apple Watch / wearable support.

HRV / breath-rate tracking, Apple Health and wearable integration, and whether the app closes the loop with measurable physiological output — most apps just guide; a few measure.

Free tier viability and subscription price weighed against library breadth, technique coverage and evidence depth. The category spans free (iBreathe, Breathe2Relax) to $150/yr (Othership) — value gap is large.

Measured power density (mW/cm²) at the skin during session, cross-referenced against independent meter readings rather than marketing-stated peak figures. Mask form factor delivers lower irradiance than panels by design; honesty about the dose is what separates serious devices from spec-inflation.

Spectrum coverage — most face masks combine red 633 nm (skin surface, collagen) with near-infrared 830 nm (deeper tissue). Some add 590 nm amber (pigmentation) or blue 415 nm (acne). The mix and the published peaks matter more than the LED count.

LED count and spatial distribution across the face — coverage of forehead, cheeks, jaw, neck. A 100-LED mask with poor distribution treats less skin than a 60-LED mask with even coverage. Neck flap inclusion is the 2026 spec war.

FDA Class II clearance, published peer-reviewed studies on the specific device for the claimed indications (fine lines, collagen, pigmentation, acne), and how honestly the manufacturer represents the published evidence. The category is full of marketing claims weakly tied to what the hardware actually does.

Flexible silicone vs hard plastic, weight, strap design and whether the mask can be worn unattended during a session. Hard-shell masks are cheaper to build but compromise daily use; flexible-silicone is the 2026 reference.

Hardware price weighed against measured irradiance, wavelength mix, clinical evidence and wearability. The category spans $169 (handheld wand) to $650+ (premium silicone masks) — value gap is large.

Does the tape or device stay on through a full night without falling off, with painless removal? Beard-friendliness for mouth tape, nostril-fit for dilators. The core daily-use criterion.

Full mouth seal vs porous mouth tape (allows emergency exhale), external nasal strip (dilates nostrils via leverage) vs internal stent (mechanically holds airway open). The mechanism dictates use case and risk profile.

How honestly the brand represents the published evidence (sleep-apnea caution is real — undiagnosed OSA + full mouth seal is contraindicated), and whether FDA registration / Class II clearance backs claims. The category is full of viral wellness marketing weakly tied to clinical reality.

Strip vs cross vs single-piece vs dilator vs collar. Each form factor trades comfort, seal effectiveness and reapplication friction differently. Cross designs allow corner-of-mouth breathing if needed.

Adhesive composition (silicone gel vs acrylic vs hydrocolloid), hypoallergenic certification, skin-reaction reports in user feedback and lip-irritation patterns. Skin-tolerance issues are the #1 reason users drop the practice.

Cost per night including subscription (Hostage Tape model) or per-pack pricing (Somnifix, Breathe Right) vs the durability of the adhesive and reusable form factors. The category spans $0.30/night DIY medical tape to $13/month subscription brands — value gap is large.

Stall force (pounds of pressure before the motor stops) and amplitude (mm of stroke depth) — the two specs that define percussion dose. Premium devices deliver 60+ lbs stall and 16 mm amplitude; budget devices deliver 30 lbs and 10–12 mm. Marketing-stated peak figures versus real continuous output is the category's biggest confusion.

Hardware build quality, motor durability, attachment-head count and quality, and warranty terms. Premium brands offer 2-year warranties with 5-6 attachments; budget brands offer 1-year with 4-6 attachments.

Hours per charge under realistic use plus measured noise level in decibels. Premium brushless motors run 50–55 dB and 2-3 hours; cheap motors run 60+ dB and 1-2 hours. Both metrics drive daily-use friendliness.

Bluetooth integration, guided recovery routines via app (Therabody, Hyperice), pressure sensors, OLED displays. The premium tier differentiator beyond raw motor specs.

Weight, handle design, multiple grip positions, mini / travel form factors and how realistic single-hand self-treatment is over a 20-minute session. Heavy premium devices win on motor but lose on extended use.

Price weighed against measured stall force, amplitude, build, app and ergonomics. The category spans $99 (Bob and Brad, Renpho) to $599+ (Theragun PRO Plus) — value gap is large and budget devices have closed the spec gap meaningfully.

HEPA grade (H13, H14 medical), activated-carbon depth (lbs of carbon), PECO / photocatalytic claims and pre-filter quality. The single biggest source of consumer confusion — many "True HEPA" claims hide thin carbon layers and inflated efficacy figures.

Clean Air Delivery Rate (CFM) for smoke, dust and pollen — AHAM-certified figures preferred over manufacturer-stated peaks. Verified room coverage (sq ft) at 5 air changes per hour (4.8 ACH minimum for asthma/allergy protocols).

Hardware build, fan motor durability, noise level at low / medium / high speeds (dB), and how quietly the device can run at sleeping speeds for bedroom use. Cheap fan motors run 50+ dB at medium speed; premium devices stay under 35 dB.

Built-in PM2.5 / VOC sensors, app integration with auto modes, real-time air-quality display, HomeKit / Alexa / Google Home support. The premium-tier differentiator beyond raw filtration.

Filter replacement frequency, per-replacement cost and 3-year ownership filter cost. Premium devices with 12-month filter life can cost less long-term than budget devices with 6-month filters.

Hardware price plus 3-year filter cost weighed against measured CADR, filtration tech, build and smart features. The category spans $99 (Levoit Core 300) to $1,200+ (Molekule, IQAir HealthPro Plus) — value gap is large.