
eKYC ProWhen building onboarding flows, identity verification often requires checking a single user...
When building onboarding flows, identity verification often requires checking a single user identifier against multiple platforms. Whether you are validating a phone number against WhatsApp, Telegram, and VK, or checking email presence, the architectural choice between individual service calls and consolidated requests significantly impacts your application's state management.
Developers typically face three paths for identity verification:
If your application requires simultaneous verification across a defined set of services, the POST /v1/check/combo/phone or POST /v1/check/combo/email endpoints are the most efficient architectural choice.
service_type.service_types array, allowing for dynamic verification logic based on user tier or region.Integration design must account for the reality of distributed systems. In a Combo Check operation, individual services may occasionally return a timeout error while others complete successfully.
// Example of a partial success response structure
{
"data": {
"results": {
"ws": { "registered": true },
"vk": { "registered": null, "error": "timeout" }
}
}
}
Because the API returns completed services even if others fail, your application logic should treat registered: null as an indeterminate state. You can then selectively retry only the failed service using the standard /v1/check endpoint, rather than re-running the entire combo.
| Requirement | Recommended Approach |
|---|---|
| Need a single risk score | Unified Score API |
| Need raw registration signals for multiple platforms | Combo Check API |
| Need to isolate specific service retry logic | Per-call API |
| Need the lowest possible number of outbound requests | Combo Check API |
For most onboarding workflows, the Combo Check approach provides the cleanest balance between simplicity and control. By moving to a consolidated request model, you reduce the complexity of your integration layer while maintaining the ability to handle individual service failures gracefully.
Always ensure your client-side timeout is configured to accommodate the expected response window (at least 15s) to properly handle the parallel processing nature of these requests. For more details on implementation, refer to the official documentation.
This article was drafted with AI assistance and reviewed before publishing.