Every co-parenting app assumes two paying parents. I built mine for one.

Every tool in this space is a court file dressed as an app: certified messaging, expense splitting, a log built to be shown to a judge. They are heavy, and they all assume two cooperating parents who both pay.

That assumption is false in most separations. One parent wants a tool. The other one just wants to know when the kids are coming.

So I built Nestido around a single user. The other parent exists in the data model as a member with no account at all: they show up in the calendar and in the statistics, children can be handed over to them, and the invitation can come later, or never. That is the nominal case, not a degraded one, and it is the decision that shaped everything else.

Three things I did not expect.

1. Removing the daily gesture breaks your own metrics. You state your custody rhythm once (alternating weeks, 2-2-3, 5-2-2-5) and the months ahead fill themselves in. From then on you only record what differs. It works, and it means a well guessed family writes nothing for weeks. My activation number sat at zero while the product was doing exactly its job. I had to put four numbers underneath it (prediction accuracy, families served, views over 7 days, families silent for 30 days) before the dashboard told me anything true. If your product succeeds by removing an action, do not measure that action.

2. A week is seven nights, not 168 hours. Adding 24 hours is not adding a night: across a DST transition, two identical intervals can count zero or one. So all arithmetic happens in the family timezone, loops start at noon and never at midnight (midnight does not exist in some zones on some days), and only storage is UTC. A one night error does not produce a stack trace, it produces a parent waiting at an empty school gate. Longer write up: https://dev.to/mollenthiel/a-week-is-seven-nights-not-168-hours-dates-in-a-shared-custody-calendar-3of4

3. The paywall must never touch what makes the product useful. I wrote two pricing models before this one, and both charged for something only a minority needed. The one I kept limits time, not entry: everything is free forever for one parent, and the paid plan only widens the window you can look back and forward through. Nothing is ever deleted, the free plan just shows less.

No messaging, no expense splitting, no court log. It is a calendar, not a case file, and I am careful never to let it look like evidence.

It is live and free, in sixteen languages: https://nestido.com. Zero paying users, and roughly two human visitors a day once you filter the bots out of the logs, so I am squarely in the "nobody knows it exists" phase. Happy to be told what I got wrong.