Home / Drivers / Systems that won't cooperate
D/09

You bought the tools.
You're still joining them up.

Each platform may perform its own job. The operation fails in the space between them, where people re-enter data, reconcile disagreement and discover too late that something never arrived.

What changed

The software estate grew. The operation never became one system.

Businesses buy applications at different moments for sensible reasons. Finance, sales, warehouse, ecommerce and service teams select tools that fit their work. Over time, the customer or order journey crosses all of them.

When the connections are weak, people become the transport and recovery mechanism. They download, reformat, re-enter, check and chase. A record can be correct in every local system while the overall operation still fails.

Another integration attempt often begins with moving data from A to B. The more important work is agreeing which system owns each fact, what a successful handoff means and who must act when the expected event does not occur.

What is at stake

The same event can create exposure or advantage.

The fear

The business cannot see or recover the failure.

Customers discover missing work first, teams dispute which record is correct and managers add more checking because no one can trust the flow end to end.

The ambition

The systems behave like one accountable operation.

Each event has a clear source, owner and state. Failures are visible to the person who can recover them, without asking staff to inspect every successful transaction.

Warning signs

How the constraint shows up.

01

The same customer, product or order is keyed into several systems.

02

Teams exchange CSV files or copy information between dashboards.

03

A failed interface is found through a customer complaint or month-end reconciliation.

04

Nobody can say which system owns a disputed field.

05

Previous point-to-point integrations became difficult to monitor or change.

06

A new platform is being proposed mainly because the existing ones do not communicate.

The usual response

Moving data is not the same as completing the operation.

01

Add another point-to-point integration

This can solve a bounded flow, but repeated links become fragile when ownership, retries and change are not designed consistently.

02

Replace everything with one suite

A common platform may reduce interfaces, but it can also trade integration complexity for process compromise and migration risk.

03

Automate the manual transfer

Robotic steps can remove typing while preserving ambiguity about source, success, exceptions and recovery.

Before commissioning software

Define the operational promise carried by each connection.

A dependable integration is a product boundary with users, failure modes and evidence — not only a technical pipe.

Q/01

Which system owns each important record and when is that ownership transferred?

Q/02

What acknowledgement proves that the receiving system accepted the right outcome?

Q/03

Which failures can retry automatically and which need human judgement?

Q/04

Who can see, explain and recover an exception without calling a developer?

The right answer may be to change the process, configure a product, integrate what you already have, modernise one part or build an owned capability. ORBN helps make that decision before it helps deliver it.

Own how you operate

Own the joins that make the operation whole.

The company may continue buying specialist systems. It should not surrender visibility and change control over the flows that connect its customers, work and money.

OWN / 01

Clear source-of-truth and lifecycle definitions for important records.

OWN / 02

Monitored interfaces with safe retries, visible exceptions and named recovery owners.

OWN / 03

Documentation and contracts that prevent one vendor change from becoming an operational surprise.

Relevant evidence and capabilities
Other reasons the question begins

What else changed?

What changed?

Make the systems work
as one operation.

Show us the handoff people are carrying and what happens when it fails. We will help define the boundary, ownership and smallest useful integration.