Your personal health scientist that works with the wearable you already use.

Say hi to start

We pick a peer-reviewed study with people like you, design the protocol around your wearable and your week, and tell you plainly whether it worked. Then the friend who texts you "did the magnesium work?" gets the answer back in a fork.

Works with

Apple Health

in testing

Browse a few real protocols.

Or describe your own goal in the middle. Cards like these get texted to your phone when you pair.

Describe a health goal.

Type a goal. Get the card back.

Same shape as the samples on either side, with citations and a duration. Yours to print, paste, or text us to run.

Anchored in

36M+ peer-reviewed papers
1.8M+ randomized controlled trials
9,000+ Cochrane systematic reviews

The science. Every protocol traces back to a peer-reviewed study, with the cohort, the dose, and the duration visible. You can read exactly what the researchers did.

The fit. Therefore adapts the protocol to your wearable, your timezone, and your week. Same study, different protocol per person.

The verdict. When the protocol ends, Therefore reads your wearable data and computes the outcome in one of ten precise states. The math is visible.

The library. Every protocol traces back to its source paper. Browse, see why a study was picked, fork it, run your own.

How it works

Text a goal. We read your wearable, search the research, and run the protocol over text.

You text a goal, we use your wearable data

i've been laying awake for 30-40 min before i fall asleep lately

your sleep latency has averaged about 38 min over the last two weeks. pulled three directions from the research.

We find the research, you pick the path

stimulus control has the deepest evidence. blue light blocking is the easiest. magnesium glycinate has 3 controlled trials across 151 people.

stimulus control

−19min

blue light blocking

−8min

magnesium glycinate

−17min

magnesium glycinate feels the most realistic

locked in. 200 mg, 30 min before bed, 21-day run.

Day 7 check-in, we read the early signal

baseline was 38 min, last 3 nights averaged 34. 6 of 7 nights on it.

is that anything?

early, but the direction's right. tends to compound. keep going.

Day 21 verdict, you decide what runs next

sleep onset dropped 8 min. the trials projected 8 to 15 at 8 weeks - you're tracking early at 3.

worth extending, or want to try something else?