My first post here, so a quick introduction. I run OdiGo app, for getting around and out on the water in the Maldives. The OdiGo app is where you book a speedboat transfer or a ferry seat between islands, charter a boat for fishing or diving, buy and sell gear in a marketplace, and check the weather and sea conditions before you leave. It's on iOS and Android. Most of the people using it live here rather than booking from abroad, and that turns out to matter a lot for how they want to pay.
Until now we took money two ways. Card through BML Connect, and plain bank transfer where the customer uploads their receipt and we verify it. Cards work fine, but plenty of our customers don't reach for one by default. Bank transfer works too, and we've automated the verification so it's quick, but it's still several steps and a wait before a seat is actually held.
Payer replaces the card with a login. The customer picks Payer, goes to Payer's checkout, signs in with their mobile number and OTP or password, and the amount comes off their Payer balance. Then they're back in the app looking at a confirmed booking. No card to find, no long number to type, just the phone number and OTP or password they already know.
We've also wired it so we can switch Payer on and off from our own admin without shipping an app release, which was useful during testing and will be useful the first time something needs turning off in a hurry.
On the integration itself
Worth saying, since this forum is partly people deciding whether to build on Payer: it was one of the easier integrations I've done.
The documentation is the main reason. It's accurate, which sounds like a low bar until you've worked from docs that aren't ā the example requests matched what the API actually returned, the webhook payloads matched the ones that arrived, and I never had to guess at a field or find out the hard way that something was optional.
The sandbox helped as much. It's the same host as live, with the key prefix deciding which you're on, so what I tested is exactly what shipped, no separate environment behaving slightly differently, and nothing to reconfigure on the way to production beyond the key itself.
Hosted checkout meant no card or wallet credentials ever touch our servers, which took a whole area of risk off the table before we'd written a line. And the webhooks are specified properly: signing, retries, the lot. I built against the spec, turned it on, and it behaved exactly as described.
That's it. Happy to answer anything about the OdiGo side if it's useful to anyone here, and thanks to the Payer team for building something that was genuinely straightforward to work with.
ā Sidey, OdiGo