Useful AI needs honest economics
Why we price Swan Labs products around real costs and clear value, and what that means for the people who use them.
Most software pricing is designed to be confusing. Tiers are arranged to nudge you upward, "unlimited" plans quietly throttle, and the real cost of a product only becomes clear once you depend on it. AI products have made this worse, because the cost of running a model is invisible to the person using it and changes constantly behind the scenes.
We think that approach is short-sighted, so Swan Labs is built around a simpler idea: honest economics.
What honest economics means
It means three things in practice.
Clear pricing. You should be able to read our pricing page once and understand what you get, what it costs, and what happens when you reach a limit. No asterisks that matter more than the headline.
No artificial lock-ins. We don't want you to stay because leaving is painful. We want you to stay because the product is genuinely useful. That means easy cancellation and no manufactured switching costs.
Limits we can explain. Running AI models costs real money, and that cost varies enormously between a quick question and a long, complex task. Rather than pretend otherwise, we set allowances we can actually sustain and explain them plainly. When you reach a limit, we tell you why, and what your options are.
Why "unlimited" is usually a fiction
When a product advertises unlimited AI, one of a few things is true: it's heavily restricted in ways that aren't advertised, it's losing money in a way that won't last, or it's quietly routing you to the cheapest possible model regardless of quality. None of those are good for the person relying on the product.
We'd rather describe access as generous fair use, set genuine ceilings, and keep the product alive for the long term. That's less exciting as marketing copy. It's far better as a foundation you can depend on.
The trade-off we're making
Being honest about economics costs us some sign-ups. A plan that promises everything for a low price will always convert better in the short term than one that explains its limits. We're choosing the slower path on purpose, because the alternative ends the same way every time — with a price rise, a downgrade, or a shutdown that users didn't see coming.
Swan Labs is being built to last. Honest economics is how we intend to earn that.
Sources & references
Useful AI, built in Perth.
Try Swan Chat, or follow what we’re building next.