Aeon: Turning a preventive health concept into a product people trust with their bodies.
Preventive health products fail in a specific way
They accumulate clinical depth while losing the patient's ability to make sense of their own results. The data gets richer, the product gets harder to navigate, and users disengage before they reach the insight the product was built to deliver.
That was starting to happen at Aeon when I joined in late 2023.
The brief
Aeon's idea was clear and genuinely ambitious: give people access to full-body MRI scans and bloodwork analysis before they get sick. Preventive, precise, and delivered through a product that made complex medical data legible to a non-medical audience.
I joined to build the mobile app from scratch. The concept was defined, clinical operations were taking shape, and the team was moving fast. But the product scope was already expanding faster than the architecture could handle, and the real design challenge wasn't going to be the interface.
The architecture decision that shaped everything
The original product was organized around MRI scan results, with bloodwork as a secondary layer. That made sense when the scope was narrow. It stopped making sense the moment investor-driven additions entered the roadmap: metabolic risk scores, brain age, liquid biopsy, genetic risk factors. Each reflected a genuine clinical opportunity. Together they were fragmenting the product. Users were being asked to move across disconnected data views without a frame to connect them.
The recommendation I made was to stop building new tests and build a unified health picture instead. Rather than separate modules, every metric would sit within the body system it belonged to. Cardiovascular: scan findings, blood biomarkers, relevant risk scores, all together. Neurological: the same. The product became a view of your body, not a collection of reports.
That decision required the founder to accept a tighter launch scope. It was the right trade, and I'd make it again. Without it, every new metric Aeon added would have required a new surface. With it, additions slotted in naturally. The product could grow with the clinical offering instead of fighting it.
The shift from MRI-first to systems-based information architecture was the most consequential product decision of the project. It wasn't in the brief. It came from looking at where users would get lost as data complexity grew, and designing ahead of that. In hindsight, I should have pushed for it earlier. The original structure should have been challenged the moment blood biomarkers entered the scope, not after the product had already been built around it.
The gap nobody was designing for
A lot of the patient experience happened outside the app, and nobody was treating it as a product problem.
Aeon communicated with patients through email at several critical points: appointment confirmation, pre-appointment preparation, results notification. But the app wasn't connected to any of it. A patient would receive an email telling them their results were available, open the app, and land in a product that assumed they already knew what to do next. No preparation before the appointment. No expectation-setting between the scan and the results. No guidance once the data appeared.
I mapped their email communication against the app experience and found the gaps. Then I pushed to join their internal communication workshop to make sure the product and the communication were being designed as one experience. The output was a set of pre-appointment and post-appointment states in the app: what to expect from the MRI, how the blood test works, what to bring, realistic timing expectations after the scan, and a guided path toward results that reduced anxiety rather than amplifying it.
It sounds obvious in hindsight. It required someone to sit in both the product and the communication channel at the same time to see it.
What I'd do differently
The NPS response rate inside the product was low throughout the engagement. We flagged it and added it to the roadmap but never fully solved the feedback loop between what users experienced and what the product team knew. Better instrumentation earlier would have given us more signal, particularly around the pre-appointment states where we were largely designing on assumptions and workshop outputs rather than behavioral data.
The outcome
In late November 2024, the product detected its first significant health finding in a real patient. I shared it with the team not as a metric but as a reminder of what the work was actually for.
Within the first six months of operation, Aeon exceeded their revenue goals by 86%. The product detected 20+ significant health findings in that same period. They closed a CHF 2M pre-seed and CHF 8M seed round, with product strategy contributing to both.
