Your Data History Is the Product

Most tools in this space are designed around the moment.

You enter some numbers. You get an answer. You move on.

We made a different bet.

The real asset is the timeline

When someone has been on levothyroxine for years, the most valuable thing they have is not their current dose or their latest lab. It's the pattern across dozens of data points over time.

That pattern is what makes a projection meaningful. A single TSH value in isolation tells you almost nothing. The same value in the context of the last six months of doses and weights tells you something real.

So we built the product around that reality from day one.

Every entry is just another row in an append-only log. There is no "edit your dose from last month." There is only adding what actually happened. This isn't because we're dogmatic about immutability. It's because the history only stays trustworthy if it isn't silently rewritten.

What this enables later

When the time comes (and it will), a patient or their new doctor should be able to take their full history and do something useful with it. That might mean handing it to a specialist. It might mean feeding it into a more sophisticated model in the future. It might mean nothing more than having an honest record when insurance or an institution asks questions.

Designing for that use case early changes a lot of small decisions:

• We favor clear, exportable formats over clever internal storage.
• We make the raw event log visible to the user instead of hiding it behind summaries.
• We treat every piece of entered data as something that might need to survive the product itself.

This is slower and less "delightful" in the short term than a slick calculator that lets you tweak everything. We accept that trade-off.

The quiet power of boring data

Most of the value in myTSH will eventually come from the boring, consistent records people build over months and years — not from any single clever feature we ship.

That's why the blog posts, the disclaimers, and the product itself all keep coming back to the same idea: this is a tool for seeing trends in your data over time.

If that sounds less exciting than an app that "optimizes your dose," you're right. It's also closer to how this actually works in real life.