Flush Business

There's a specific kind of shame that only happens in a supermarket checkout line.

The trolley.

The 24-pack.

The walk from the paper aisle to the register and the creeping certainty that every person in your peripheral vision knows exactly what you're about to do with your weekend.

The funny part is nobody is actually looking.

Everybody is buying the same pack.

Businesses keep trying to make toilet paper a luxury item.

Same biology, better typography, and an unspoken agreement not to discuss any of it.

In all honesty, that silence has a dollar value.

It might also have a medical cost.

As of mid-2026, four hardware companies had decided to start measuring what hits the porcelain.

Before I finished writing this, a fifth showed up.

Serious Data

A while ago, on a trip to Australia, I got very sick.

I'll spare you the details, but it ended with a doctor, an ice-cream container and a little plastic scoop.

I'll let your imagination do the rest.

This was after eating a McChicken that had been sitting in the fridge for three weeks, which was less a meal and more an avoidable experiment.

Doctors know what your gut output reveals.

Patients like me get uneasy discussing it, then even more uneasy when told to scoop it up, seal it in a container and carry it into a clinic.

There is a thing called the Bristol Stool Scale.

It classifies stool into seven types, from hard sheep-like pellets through to liquid yuck.

It is not a diagnosis on its own.

It gives clinicians and patients a shared way to describe stool form, which is useful when the alternative is pointing at a cartoon and saying, "sort of like that one."

Your gut does not dress things up, but people are still surprisingly inconsistent at describing what came out of it.

That is one reason cameras and computer vision are interesting here.

IBS affects roughly one in eight people in the US, and inflammatory bowel diseases such as Crohn's and ulcerative colitis affect millions more.

These are not fringe conditions.

Between clinic visits, a lot of people manage symptoms on memory, instinct and some version of "I just feel like shit."

At the same time, every prebiotic drink in the supermarket tells you it is good for your gut without showing whether it did anything to your gut.

The existing tool for collecting real data is still the stool sample kit.

I've done this twice now.

You get a collection kit and a paper sheet that drapes across the toilet bowl.

You produce, scoop a sample into a container, seal it, bag it, and walk it into the doctor's office carrying something you cannot believe you are carrying.

Then you hand it to a nurse who quite reasonably gets gloves before touching the thing your doctor specifically told you to bring in.

The whole process takes about 15 minutes and is probably the least dignified 15 minutes available through primary care.

The consumer version now arrives by mail.

Viome uses metatranscriptomic sequencing to analyse RNA activity in a stool sample, then returns gut scores, food recommendations and supplements in about two to three weeks.

Viome says it has tested more than a million microbiomes.

The business case is clever because the test happens once, while the recommendations and custom supplements can keep turning up every month.

The clinical version of the same basic truth is a test such as Cologuard.

It looks for DNA changes and blood associated with colorectal cancer risk, and Medicare covers eligible people for a multi-target stool DNA test every three years.

Stool still reveals things a blood test does not.

Then we come to yellow gold, which is the same story from a different angle.

Withings now sells U-Scan, a small urine analyser that sits inside the toilet bowl looking like a clam that has made a series of very strange career choices.

The Nutrio cartridge tracks hydration, vitamin C, ketones and bio-acidity.

The Calci version tracks calcium, hydration and bio-acidity.

Withings calls it the first toilet-embedded urine lab and sells the Nutrio starter kit for about US$380, followed by replacement cartridges.

The measurements are sold for general wellness and self-tracking, not diagnosis.

That distinction matters.

Your GP orders a urine test when something is already wrong.

Withings wants to run one before you have had coffee, then sell you the interpretation on a subscription.

A 40-Year-Old Idea

The smart toilet is not a 2025 invention.

Japanese patent filings for toilet-based urine analysis go back to 1982.

TOTO built toilets that could measure urinary sugar and hormone levels in the early 2000s, but unsurprisingly, demand was low.

They cost thousands, required replacing the whole toilet, and lived in a world with no smartphone to receive the data and no subscription model to turn it into a business.

The technology worked.

The rest of the market had not caught up.

While diving into this, I found a Stanford paper in Nature Biomedical Engineering describing a toilet-mounted system capable of running a ten-parameter urine analysis, classifying stool on the Bristol scale with computer vision, measuring urinary flow and tracking a range of other signals.

The clinical advantage is obvious: unlike a wearable, you cannot forget to put your toilet on.

Talk about the full Monty.

To identify which household member produced which data, the Stanford team put a fingerprint scanner on the flush lever and used a second camera to identify people by the unique creases of their anus.

This is not a joke.

It appeared in a peer-reviewed academic journal, not a startup pitch deck.

The paper politely calls it an analprint.

As it turns out, your arse is as committed to biometric authentication as your face.

The technology has been circling labs and patent filings for more than four decades.

What failed each time was not the sensor.

It was everything required to turn the sensor into a product people would buy, install and keep paying for.

Toilet Business

Eight Sleep did not invent sleep tracking.

Fitbit was automatically tracking sleep in 2009.

Beddit was measuring sleep through an under-mattress sensor before Eight Sleep launched, and Apple bought it in 2017.

Sleep Number had smart-bed features before Eight Sleep sold its first Pod.

What Eight Sleep built was the right sleep monitor at the right time with the right business model.

The Pod costs more than many mattresses and Autopilot adds a monthly subscription.

It sounds ridiculous until you remember that people happily subscribe to software for their television, doorbell, car, watch and now apparently their bed.

If you can subscribe to your bed, why not your toilet?

Oura sells jewellery that happens to be a computer.

WHOOP sells a screenless recovery strap where the subscription is effectively the product.

Garmin came from the buy-once, cry-once sports-watch direction, then added Garmin Connect+ because recurring revenue eventually comes for everyone.

Each product has a primary reason to exist that is not clinical data.

Eight Sleep is temperature and comfort.

Oura is a ring.

Garmin is a sports watch.

The health monitoring becomes the hook that keeps the subscription useful.

None of them lead with biometrics.

The former CTO of WHOOP, John Capodilupo, has now co-founded Throne, a company that clips a camera and microphone to your toilet and tracks what comes out.

There is a pattern here.

Take a universal human behaviour, find sensing technology that has existed long enough to be reliable but has never been packaged properly, make the monitoring passive, add a subscription, then start with biohackers and health fanatics before pursuing clinical validation.

Sleep.

Activity.

Heart rate.

Recovery.

Toilets?

TOTO was early because the sensor was only one part of the equation.

The smartphone, subscription model and culture of tracking personal health data did not exist yet.

Now they do.

Ramping Up

This is not an off-the-wall thought experiment anymore.

Products from Withings, Kohler, Throne and TOTO all arrived or opened for sale within roughly a year, and Vivoo announced another one while the category was still learning how to explain itself.

The category arrived without much of a press cycle because the stigma is still very hard to beat.

Withings took the chemistry route.

U-Scan sits in the toilet, detects urine, draws a sample and runs it through sensors inside a replaceable cartridge.

No camera.

No optics.

It does not need to see anything because urine analysis can be done with chemistry.

But it is urine only.

Stool is not in scope.

Kohler, the bathroom company, launched a health division and made Dekoda its first product.

Dekoda clips to the rim and uses optical sensors to provide hydration and gut-health patterns, including alerts for signs of blood.

The blood alert matters because that is real information worth taking to a doctor, even though Kohler is careful to sell Dekoda as a tracker rather than a diagnostic device.

It costs US$599 and requires a membership starting at US$6.99 a month.

Kohler initially said the data was protected with end-to-end encryption.

A security researcher pointed out that it was not end-to-end encryption in the normal meaning of the term because Kohler's systems could decrypt and process the data.

Kohler removed the claim after TechCrunch called.

Worth remembering the next time a health gadget promises privacy in the marketing copy.

Throne takes a more direct route.

It is a US$399 clip-on sensor with a two-megapixel downward-facing camera, microphone, ambient-light sensor and motion sensor.

It tracks stool form and frequency, hydration, urinary flow and bathroom habits, then charges US$6 a month after the first three months.

The camera and microphone combination is deliberate.

Computer vision can classify what is visible while acoustic analysis adds another signal for urinary flow and what the camera may not distinguish cleanly.

The household identification problem is handled through Bluetooth proximity or a personal button.

That is much more elegant than Stanford's analprint database, although Throne's own privacy notes admit a nearby paired phone can occasionally start the wrong person's session.

Nobody said cameras in toilets would be simple.

TOTO is also back.

The company released NEOREST models with a Stool Scan function in Japan in August 2025, automatically measuring stool condition and sending it to the TOTO Wellness app.

It is still taking the full-toilet-replacement approach rather than the retrofit clip.

Higher price, higher installation friction, but once installed it is not falling off the rim or running out of battery.

Vivoo then appeared at CES 2026 with a US$99 urine-focused sensor and app subscription.

It was available for preorder rather than sitting on shelves everywhere, but four became five before anyone finished writing about four.

There is also a quieter version of this market that has been running for years.

Toi Labs has deployed its TrueLoo smart seat in more than 50 senior-living communities.

It optically tracks stool and urine patterns so care staff do not have to follow residents into the bathroom or rely on memory and manual logs.

That is less exciting than a CES launch, but it may be the stronger proof that passive toilet monitoring solves an actual problem.

Snapshot Versus Stream

Viome is the snapshot business.

Send in a sample, wait two to three weeks, receive RNA analysis, food recommendations and a list of supplements it can sell you.

That can be useful, but it is roughly equivalent to taking your blood pressure once a year and declaring yourself monitored.

ZOE, Tiny Health and others sell versions of the same mail-in idea to different audiences.

Toilet hardware is trying to be the stream business.

It asks what your body did this morning and how that compares with last Tuesday.

These are different products with different jobs.

Viome answers: what is active in my gut right now, and what should I be eating?

Throne or Dekoda answers: what changed in my daily output?

The truth is that useful care probably needs both.

The behavioural stream for ongoing monitoring.

The molecular snapshot when something changes enough to justify a lab.

Bowlsthetics

Then there is the problem of making any of this look normal in a bathroom.

Throne looks like a paperback book clipped inside the toilet rim.

It runs for about a month per charge and ships with a 13-foot USB-C cable because bathrooms were not designed around charging a camera attached to the bowl.

Kohler's Dekoda sits on the front of the toilet like a large white clam that every guest will notice.

The question either device generates is not necessarily, "I want that."

It is the other question.

Compare that with an Oura Ring or an Eight Sleep Pod visible on a bed.

Both can generate genuine curiosity and product desire.

A camera clipped to your toilet generates the same surprise with a completely different emotional outcome.

The Oura Ring is jewellery that happens to be a computer.

Throne is a camera that happens to be watching the bowl.

To Bidet, or Not to Bidet

In Japan, bidet toilet seats are already normal.

A trip to Korea made me realise how glorious they are and that Westerners are basically pigs for refusing to make them standard, including in public bathrooms.

The US started catching up during the COVID toilet-paper shortage and bidet sales did not collapse when everybody went back outside.

Unlike Peloton.

Unlike Zoom.

A premium electric bidet seat is already plugged into the wall, fitted with a pressure sensor to detect occupancy, positioned directly over the bowl, and bought primarily for comfort and hygiene.

Put health sensors into that form and the monitoring becomes a subscription hook rather than the pitch.

Imagine opening a dinner-party conversation with, "I have a device that monitors my bowel movements."

I can guarantee that will not be a successful opener.

Now try, "I got a Japanese toilet. It's incredible. And it also tracks your gut health."

That just might work.

Nobody has built the complete version of this yet.

TOTO is closest, but it is a full toilet replacement costing thousands and requiring a plumber.

The missing product is a retrofit bidet seat that combines optical stool analysis with chemical urine analysis, fits a normal toilet, and can be installed without explaining an analprint scanner to an electrician.

Whether that is a gap worth filling or a gap that exists for a reason is the expensive question.

Hardware margins on a powered seat are not the same as margins on a small clip.

GLP-1 in the Wild

Then GLP-1 drugs arrived and made the customer problem much larger.

J.P. Morgan estimates around 10 million Americans were using GLP-1 treatments in 2025 and forecasts roughly 25 to 30 million by 2030.

These drugs can slow gastric emptying, and nausea, constipation and diarrhoea are common enough to appear prominently in the prescribing information.

Real-world studies also show a lot of people stop within the first year, although side effects are only one reason alongside cost, access and insurance.

The gut response varies between people and after dose changes.

Right now, most patients manage that between appointments by remembering what happened and trying to decide whether it was the medication, the meal or just Tuesday.

A toilet sensor with a GLP-1 mode is an obvious thought experiment.

Establish a baseline before medication, compare bowel patterns with dose timing, and flag a meaningful change that is worth discussing with a clinician.

That is not the same as diagnosing a side effect, and making a specific medical claim would make the regulatory conversation much harder.

It may also explain why nobody has shipped that mode yet.

Missing Seats

The products that launched in 2025 and 2026 have solved at least part of the sensing problem.

The next problem is turning the data into something people actually want.

Think about what Eight Sleep delivers.

The value is not that you can open a dashboard and admire your sleep temperature.

The cover changes the temperature before you wake up, and you feel better in the morning.

The data is mostly invisible.

The result is the product.

Every current toilet product still leads to a dashboard.

• Your hydration score is 68.
• Your Bristol Stool Scale average is 4.2.
• A trend line says your gut was less stable in February.

What does that mean?

What do you do with it?

The current answer is usually some version of look at the graph, tag what you ate, or speak to a dietitian.

What people actually want from this category is not more data.

They want an explanation.

• Why do I feel bloated every Tuesday?
• Why has my energy been low for the last three weeks?
• Is the medication doing anything?
• Is the new diet working?
• I recently started a GLP-1. Is this change normal enough to monitor, or worth calling my doctor about?

The gut is one of the most common sources of symptoms people struggle to explain.

The people willing to pay US$500 for a toilet sensor are probably the same people who have already tried several other things and still do not have an answer.

Gut data that requires interpretation is homework.

Most people do not consistently do homework, especially when the homework involves photographing the toilet.

The stigma compounds it.

Buying an Oura Ring means telling people you track your sleep.

That is a completely normal sentence in 2026.

Buying a toilet monitor means telling people you have a camera classifying your bowel movements.

Nobody is volunteering that at dinner.

Eight Sleep is not a mattress-cover brand.

It is a sleep-fitness brand.

The cover is the detail.

The categories that broke through, sleep, recovery and activity, all landed on products people could talk about without mentioning the raw data.

Nobody has found that story for the toilet yet.

The sensing problem and the story problem need separate solutions.

Right now, every company is trying to solve both with the same piece of plastic clipped to the bowl.

So What?

At least five consumer toilet-health products launched or were announced within roughly a year.

Two came from established bathroom and health-hardware companies.

One came from WHOOP's former CTO.

TOTO returned to an idea it could not commercialise 20 years earlier.

The quieter care-market version is already operating in senior-living facilities.

This is not a fringe area of business anymore.

The unit economics make sense on paper.

The retrofit products range from about US$99 to US$599, with several adding US$6 to US$10 a month or ongoing cartridge plans.

Adjacent categories have already proved that people will pay for hardware and then subscribe when the product changes behaviour or makes them feel better.

What has not been proved is whether toilet data changes behaviour.

Whether the dashboard becomes something people return to after the first funny week.

Whether the stigma loosens the way sleep tracking did, or stays uncomfortable enough to keep the whole category niche.

The market is not small.

IBS affects roughly 12 percent of people in the US.

GLP-1 use is growing from millions to tens of millions.

People with Crohn's and ulcerative colitis still manage long stretches between clinic visits using memory and instinct.

These are people with a recurring, disruptive problem that current products do not fully solve.

They are not wellness tourists.

The sensor is finally good enough.

The business still has to make people care after the novelty flushes.

Shout out to all the legends I've ripped info from for this piece:

NIDDK | PubMed | Viome | Medicare | Withings | Nature | Kohler | TechCrunch | Throne | TOTO | Vivoo | Toi Labs | Garmin | J.P. Morgan | FDA | Crunchbase | CNN Business