By Henry Ho, Founder of Geode
A note on why Geode uses one account and one plan across iPhone, iPad, and Mac, instead of splitting the user’s audio workflow by device.
When we decided that one Geode plan should work across iPhone, iPad, and Mac, not everyone on the team agreed.
The concern was reasonable.
Geode is a local-first transcription app. Many local tools are sold per platform. A Mac app is a Mac app. An iPhone app is an iPhone app. If a user wants both, they buy both.
That model is simple. It also avoids one obvious problem: sign-in.
A user who just downloaded a local transcription app usually wants to try it immediately. Asking them to create an account adds friction. We knew that. We also knew it could hurt first-use conversion.
So the easier path was clear: no account, separate apps, separate purchases.
I decided against it.
The reason is that Geode is not built around devices. It is built around the way audio work actually happens.
A recording may start on iPhone because that is the device in your pocket when the conversation happens. Later, the serious work may happen on Mac: higher-accuracy transcription, speaker separation, summaries, editing, review.
That is not two separate products in the user’s mind.
It is one job.
If Geode forced users to buy the iPhone app and the Mac app separately, we would be charging them twice for completing the same workflow.
I do not think that is the right model.
This decision also fits how we think about devices.
We do not pretend every device should do the same thing.
iPhone and iPad are better for capture and mobile transcription. They are with you when a lecture starts, when an interview happens, when a voice note needs to be saved, or when audio comes from another app.
Mac is better for deeper local work. It has the hardware and screen space for long recordings, speaker separation, on-device summaries, editing, and review.
A cloud-first product can hide these differences by sending everything to servers. A single-device local tool can avoid the question altogether.
Geode cannot.
If we believe local AI matters, we also have to respect what each device is actually good at. The right answer is not to make every device identical. The right answer is to let each device do its best part of the workflow.
That is why one account matters.
Not because we wanted to add a login screen. We did not.
The account exists so the user’s plan can follow them across devices. Without it, the workflow breaks. Record on iPhone, continue on Mac, review later on another device. That should feel like one Geode experience, not a set of separate purchases.
This also shaped how we built cross-device transfer.
We chose local-network discovery and peer-to-peer transfer instead of relying on iCloud or routing files through our own servers by default.
That was not the easiest engineering path.
Cloud sync would have been simpler to explain. It would also fit the pattern users already know: upload everything, then access it everywhere.
But Geode is local-first. If a recording moves from your phone to your computer, it should not have to pass through our server just to make the workflow convenient.
So we chose the harder path.
One account across devices.
One plan for the full workflow.
Local transfer where possible, without making the cloud the default middle layer.
I know this can feel unfamiliar at first. Some users may wonder why a local-first app asks them to sign in. That is a fair question, and we need to explain it clearly.
The answer is not that Geode wants to collect more from the user.
The answer is that Geode is trying not to split the user’s work by device.
A product should not make someone pay again just because the work moved from the pocket to the desk.
That was the decision then, and it is still the decision I would make now.