Trading Without Friction

"Easy" crypto apps hide the complexity and charge a premium spread for the privilege. "Pro" apps put everything on screen and assume you brought your own confidence.

The industry treats that as a segmentation choice. It isn't. Beginners pay hidden fees and never learn enough to graduate off them, which is the point. Pros pay a confirmation tax built for somebody else. Both audiences are being charged for the same decision, and neither one asked for it.

The call the whole thing rests on: one platform, one shared core, and the mode is a setting you flip.

The easier build is two products, and it's easier because it lets each one have its own roadmap. It also traps people. A beginner on a beginner product has to leave and start again somewhere else to grow, and that's the moment every one of these apps loses the customer they spent the most to acquire.

Making it a setting means a beginner can see Pro exists before they're ready for it, and a pro can flip to Beginner to walk a friend through a trade without logging out of their own account.

The most-fought decision on the project. The full cost of the trade, spread included, sits next to the action button on every order. Not in a drawer, not behind an info icon, not on the confirmation screen after you've already decided.

It's unpopular internally for an obvious reason: the spread is the revenue, and putting a number next to a button is the fastest way to make someone reconsider pressing it. The hidden version only works while people don't know. The day they find out, you have taught them the product was designed against them.

The second rule was plain language as a constraint rather than a copy pass at the end. If I couldn't explain a thing in one sentence, it either got explained inline or it came out of beginner mode entirely. That killed features, which is what a real constraint does.

The first beginner mode was smothered. Confirmations on everything, a tooltip on every term, and an onboarding tour that would not end. It tested badly, and not for the reason I expected. People didn't feel safe. They felt talked down to. Safety you can't switch off reads as an accusation.

The fix was to explain on hover, confirm only above a threshold, and otherwise get out of the way.

Pro mode failed in the opposite direction and I'd made the same mistake in reverse. I stripped it down for cleanliness and took out things pros actually used. Density is a feature for that audience. I was treating my own taste as a usability principle.

Nothing here shipped. There's no live product to go and press, and I'd rather say that at the top of this section than let the screens imply otherwise.

What exists is the reasoning: the teardown of who each model is really charging, the decision to make mode a setting instead of a second product, the cost-next-to-the-button rule and the argument for holding it, and two prototypes that failed in opposite directions and taught me more than the ones that worked.

The gap I never closed is the intermediate trader. They are wrong in both modes: past hand-holding, not yet fluent in an order book. A real v2 needs either a third mode or per-control customisation, which is a different and harder problem than the one I solved. That's the honest edge of it.