We are a group of friends who have known each other since primary school in Aalborg, Denmark, and every year we go away for a long weekend with our families. This April it was a farm holiday, and during that weekend Jakob brought up an idea he had first suggested years ago.
Jakob Schöllhammer Knudsen is a doctor, and for a long time he had been talking about the same gap. People collect more and more data about their own health, and they still have no good way to find out what works for them. Should you stop drinking coffee after noon? Does the evening walk do anything for your heart rate? Most of us try things like that, and most of us never find out.
Research doesn’t always help either. Many everyday habits have never been properly studied at all. And when something has been studied, the study tells you how it worked on average across a large group of people. An average can hide a lot. Even when something works for most people in a study, some get little or nothing out of it, and the study can’t tell you which group you belong to. Knowing that something helps many people still doesn’t tell you whether it helps you.
Years ago, it would have been hard to build a product that answers that question for one person at a time. This time it looked possible. Smartwatches and rings already measure sleep and heart rate around the clock, and AI assistants have become good enough to help someone plan a proper experiment in an ordinary conversation.
Two of us also happened to have the time. Bjarke had left WorkTango in December, wanting more time with his family and a new challenge. Peter had left the University of Copenhagen a year earlier to become an independent researcher with a fully flexible calendar. Jakob, who runs his own company, joined part-time to bring the clinical side. All three of us were keen, so we started building it.
Questions we couldn’t answer ourselves
Part of why the idea stuck is that two of us knew the problem from our own lives.
Bjarke: I’ve had lower back trouble for a while. I started using a standing desk, but on and off, depending on my habits and whether I remembered. I noticed my back seemed better on days I had been for a run, so I asked an AI chatbot about it and got a routine of morning exercises and stretches. Some mornings I did them, some mornings I forgot. The pain came and went, and I honestly can’t tell you whether any of it helped. Judging that from memory is a bit of a crap-shoot: you end up leaning on your own biases. With a background in the natural sciences, I would much rather run a real test.
Peter: I have tracked my sleep and heart rate for years. I want a long and healthy life, so I’ve tried running, meditation, breathwork and sauna to bring my resting heart rate down and my heart rate variability up. My watch shows the numbers moving, but it can’t tell me whether any particular change caused it, and I never tested one thing at a time. So I still can’t say which of them made a difference. In my work I see how easily people read cause into coincidence, and it turns out I do it too. N1 Trial is a natural response to that.
What N1 Trial does
There is an established method for exactly this kind of question. It is called an N-of-1 trial, and doctors and researchers have used it since the 1980s. You live with the change and without it in planned blocks of days, several times over, and measure the same thing all the way through. You become your own comparison, and no single good or bad day decides the answer.
The hard part has always been the effort. Planning a good trial takes know-how, and keeping track for weeks takes discipline. N1 Trial takes care of the tedious parts:
- You plan your trial in a short conversation with an AI assistant, or start from one of more than 200 ready-made trials.
- Your smartwatch, Oura or Whoop does the measuring in the background. For things a watch can’t measure, such as back pain, mood or headaches, you answer a quick question each day.
- When the trial is over, you get an answer and see how sure it is. If the data can’t tell yet, the app says so.
What we won’t compromise on
The first thing is honesty about what your data can and can’t say. A result that sounds more certain than it is would be worse than no result, because you might keep paying for a supplement or dropping a habit on the strength of it. So every result shows how uncertain it is, and when the data can’t tell yet, the app says that instead of guessing.
The second is your data. Bjarke learned at WorkTango that privacy and data protection are far easier to build in from the first day than to add later. Your trials and health data are stored and processed in the EU, and we don’t sell data or share it with advertisers.
N1 Trial is for everyday habits. It isn’t a medical device, and the AI assistant is built not to tell you to start, stop or change a medication. That is a conversation for you and your doctor.
Who we are
Bjarke Grouleff is a computer scientist from Aarhus University with about 15 years of building software. He co-founded WorkTango, built the first version of the platform himself, and led its engineering as CTO through the sale to Vista Equity Partners in 2022. N1 Trial is the third company he has co-founded. Here he is responsible for the app and everything underneath it, with security and privacy high on the list. Outside work it is family, travelling and hiking, as far out in nature as possible.
Peter Fisker has a PhD in economics from the University of Copenhagen. His research is on causal inference and treatment effects: working out whether something really caused a change, and how big the change was. That question is at the heart of every N1 Trial result. He works as a consultant for the World Bank. At N1 Trial he is responsible for the method: how trials are planned, how they are analysed, and what the app can honestly conclude from one person’s data.
Jakob Schöllhammer Knudsen had the idea in the first place and is our co-founder, part-time. He is a medical doctor with a PhD in pharmacoepidemiology, the study of how medicines work in real patients, and helps us with trial design, including for people living with conditions such as diabetes, where exercise and food show up in blood sugar from one day to the next. He is also the founder and CEO of Donor Network, sits on the board of Novatech and is an active investor.
We have known each other since we were children, and we trust each other. We also argue well, and most decisions get tested between us before they reach the app.
Where we are now
N1 Trial is a small company based in Aarhus, Denmark. A small group of invited testers is using the app now, and we are learning a lot from them about what fits into a normal day and what doesn’t.
If you have a question about your own habits that you would like to put to a proper test, sign up for early access and we will write when it’s your turn. And if you want to talk about the idea or the method, email us at info@n1trial.com. We would love to hear from you.