Skip to content
AspirecoStart
All answers

Integrations

Why do software integrations keep breaking after launch?

The short answer

Because the systems on the other end keep changing. Every connector is a promise to keep up with somebody else's API, and most integration failures are quiet: an event that never arrives, a field that changes meaning, a credential that expires.

Updated

The far end moves

Vendors change APIs, deprecate fields and add requirements on their own schedule. Asisly, our own voice receptionist, reaches eighteen third-party connectors — eighteen far ends, each with a roadmap that does not mention you.

Failures are quiet by default

An integration rarely crashes loudly. It drops an event, retries into a duplicate, or keeps running on a credential that stopped working. So every webhook router we build carries retries, idempotency and a failure digest that reaches a person: a stuck event becomes a line someone reads in the morning, not an order that never happened.

Credentials expire

Expired and rotated credentials are a common cause. Business OS brokers secrets centrally, so a credential rotates without redeploying every system that uses it.

Budget for keeping up

Integrations get quoted as one-off work and then behave as an ongoing liability. Scope the keeping-up along with the build.

Asked next

How do we find out an integration has stopped working?

Something has to tell you. Monitoring that reports what ran and what failed, and a daily digest a person actually reads, turns a silent failure into a visible one. Without it, the first alert is usually a customer.

Should integrations be covered by a maintenance retainer?

Often, yes, because the work does not end at launch. Retained capacity covers monitoring, digests and escalation for an automation estate, sized to what it looks after.