Eight months ago I launched TubeSpark, an AI co-pilot for YouTube creators. Pricing it was the hardest non-technical decision I made.
Building from Brazil means your users are everywhere, but your costs are in dollars. OpenAI, Anthropic, Google, Groq — every API bill comes in USD. Pricing in BRL only would mean my margins swing with every exchange rate movement.
I ended up with a dual system:
International: Stripe in USD. Starter at $6.90/mo, Pro at $17.00/mo.
Brazil: LemonSqueezy in BRL with PIX. Starter at R$47/mo, Pro at R$147/mo.
The BRL prices aren't a straight conversion. R$47 is about $8.30 today, not $6.90. I eat the difference because a Brazilian creator making R$3k/month can't justify $6.90 the same way someone making $5k/month can. PPP stops being theoretical when it's your own country.
Why $6.90 and not $9 or $12? I tested three pricing options on a mock checkout page before writing a single line of Stripe code. $6.90 had 2.3x the conversion rate of $12. At my current API costs per user, $6.90 still gives me ~62% margin on Starter. Good enough to grow.
Pro at $17 unlocks 4 AI providers (GPT-4, Claude, Gemini, Groq) running in parallel. Each one costs me differently. Claude is the most expensive per token, Groq is the cheapest. Blended cost per Pro user is around $5.40/mo, so margins are tighter, but retention is much better. Pro users stick around 3x longer.
One thing I got wrong early: I launched with only yearly billing because I wanted committed users—terrible idea. Monthly converts 4x better for a product nobody's heard of. Nobody commits to something they found five minutes ago. I added monthly, and conversions jumped immediately.
The LemonSqueezy integration for Brazil took about two weeks longer than I expected. PIX payments are instant and free for the buyer, which matters a lot here. But refunds, subscription management, webhooks — everything works slightly differently from Stripe. I basically maintain two payment backends. Annoying, but the Brazilian market is worth the extra code.
Current split: about 65% international (USD), 35% Brazil (BRL). Brazilian users came mostly from organic Portuguese content. International users came from IH, PH, and SEO in English.
If I started over, I'd price test even earlier. Before the landing page, before the logo, before everything. The price shapes the product more than you think.
How do you price for markets with different purchasing power? Anyone else running dual payment systems?
"Going vertical with 'AI for YouTube creators' is such a smart move compared to the broad AI writing tools. It makes the marketing copy much easier to dial in. You should consider submitting this to this competition—entry is $19 and the winner gets a trip to Tokyo. Prize pool just opened at $0. Your odds are the best right now."
Round just opened 👉 tokyolore.com
aitubespark, your report about dual pricing (USD/BRL) is a guide for any Brazilian founder. The challenge of maintaining margins with costs in dollars and local revenue is real. Regarding the $6.90 vs $12 test, did you use any methodology like Van Westendorp to find the “optimal price point,” or was it purely an A/B test on the landing page?