An in-lab sleep study is the gold standard. SomniSense is for the other 364 nights.

The lab night you might already have had

If you've ever spent a night in a sleep lab, you know what it's like. You arrive at 8 p.m. A technician glues sensors to your scalp, your face, your chest, your finger, your legs. You're told to "sleep normally" while wearing the equivalent of a small octopus on your head. You wake up the next morning, drive home, and a few weeks later you get a report.

It's the gold standard for a reason. Polysomnography measures a dozen channels we can't get from a phone — EEG, EOG, EMG, oxygen saturation, leg movement, body position. The sleep technician scoring your night went through years of training to do that scoring well. The data is real and the diagnosis is real.

It also costs $1,500 to $3,500 if you're paying out of pocket, takes weeks to schedule, and gives you exactly one night of data — usually a night where you didn't sleep that well because you had sensors taped to your face.

That night is essential when you need a diagnosis. It's not very useful for tracking what's happening on a Tuesday in March when you had two glasses of wine and slept on your back instead of your side.

The 364 nights nobody measures

This is the gap I built SomniSense for.

The lab can tell you whether you have a sleep breathing condition. It can't tell you what your nights look like the rest of the year. It can't tell you whether the side-sleep training you tried for two weeks actually moved your numbers. It can't tell you whether the four glasses of wine at your friend's birthday party show up as four glasses of wine in your breathing pattern. It can't tell you whether your numbers improve when you finally lose those last 8 pounds.

SomniSense is a way to track those 364 other nights. Validated against the lab on n=70+ paired nights, so the trend you're seeing isn't fiction. But trend, not diagnosis.

Where each one fits

The most underrated workflow: PSG + SomniSense together

Almost nobody talks about this, but it's the workflow that helps the most people I've talked to.

Step 1. You're tired, snoring, or your partner's worried. You start with SomniSense for two weeks to get a baseline of what your typical nights look like.

Step 2. If your BRI runs above 15, you take that data to your primary care doctor. They write a referral for a sleep specialist or an HSAT.

Step 3. You do the formal study. You get a real diagnosis with real numbers from a real polysomnography setup.

Step 4. You start treatment — maybe CPAP, maybe a positional device, maybe just lifestyle changes. SomniSense quietly keeps running, showing you whether the treatment is working over weeks and months.

Most people skip Step 1 because they didn't know it was an option. They wait until something is undeniable, or until a partner refuses to share a bed anymore. SomniSense's role is to make Step 1 cheap enough that you do it sooner.

What SomniSense is NOT trying to be

• Not a replacement for in-lab PSG. The lab is a different kind of measurement.
• Not a replacement for an HSAT (Type-III home test) when one is medically indicated. HSAT has chest band and pulse ox we don't have.
• Not a clinical diagnosis. We use the same per-hour metric your specialist uses, but a number from us is data, not diagnosis.

What it is: the cheapest way to know whether to ask for the formal study, and the only way I know of to track what's happening on the other 364 nights of the year.