Recently I ran into an issue where I couldn’t top up my balance on sms-activate for several days.

This wasn’t a one-off glitch — payments simply didn’t go through, and there wasn’t much visibility on when it would be resolved. For anyone relying on SMS verification services for testing, onboarding flows, or QA, this kind of dependency failure is surprisingly disruptive.

It made me reflect on something we often underestimate: how fragile “infrastructure-like” third-party services can be, especially when they’re widely used but lightly abstracted.

Over the past months, I’ve been working on a small alternative called SMS-Act, initially just to reduce my own operational risk. The goal wasn’t to compete on scale, but to focus on a few principles I personally cared about: • Clear scope: one-time, temporary numbers only (not long-term rentals) • Transparent pricing (credits refunded automatically if verification fails) • Simpler UX for developers and testers who just need a code, fast • Fewer “black box” behaviors when something goes wrong

I didn’t plan to talk about it publicly, but the recent downtime made me realize many people here might be facing the same issue and quietly scrambling for alternatives.

I’m curious: • How do you handle SMS verification dependencies in your stack? • Do you build internal fallbacks, or just switch providers when things break? • What’s the biggest pain point you’ve had with services like this?

If anyone’s interested, I’m happy to share what I’ve learned building SMS-Act so far — or just discuss better ways to reduce risk around these kinds of external services.

(No hard pitch here — genuinely more interested in how others approach this problem.)