The symptom
“Our systems do not talk to each other”
Connect Disconnected Systems
What it’s costing you
Retyping is the visible cost. Disagreement is the real one.
- The same record typed into two or three systems, with three chances to differ
- Numbers that disagree between departments, and meetings spent reconciling them
- Work that waits because the next team cannot see what the last team did
Why it usually happens
Each system was the right buy. The gaps came free.
- Each system was bought to solve its own problem, by the department that had it, and each purchase was sensible on the day
- No vendor sells the space between products: the accounting vendor stops at accounting, the point of sale vendor at the till
- Integration is invisible work, so it loses the budget argument to visible features until the retyping becomes somebody’s whole job
The target operating state
One record, entered once, read by every system.
- 01In the published boutique engagement this fix put a counter, a website and two marketplaces on one record: a sale anywhere deducts from the same number, and the first person to notice a stock problem stopped being a customer
- 02One record for each thing, entered once, read by every system that needs it
- 03Integrations between the platforms you already run, before any replacement is discussed
- 04Handovers that happen in the system rather than in a phone call
A worked example
Six channels, one stock number.
Before
A Dubai boutique sold two catalogues through a counter, a website, Amazon and Noon, with each channel running on its own copy of the stock figure, reconciled by hand.
After
A custom ERP and POS put every channel on one record: a sale anywhere deducts from the same number, so a size sold out at 9am is sold out everywhere by 9:01.
Services that may be involved
Platforms that suit this
Questions
Do we replace our systems, or connect them?
Often connect. Integration is defined work on the systems you already run: which records flow, in which direction, what happens when a message fails and who owns each side. Replacement is the answer only when a system genuinely cannot carry its half.
Who owns the integration when something fails?
That is agreed in writing per engagement, before anything is built: direction, error handling and ownership are defined, so a failed message is somebody’s job rather than everybody’s surprise.
What does connecting two systems cost?
It is scoped rather than priced from a list, because the cost lives in the data and the edge cases, not the connector. Every connected system adds scope, and undocumented third-party APIs add risk we name up front.
Where does the Health Check fit?
It maps the joins first: where data breaks between systems and what the break is costing, in writing, so the quote is for work that actually exists.
Ready to make them talk?
If that sentence sounded familiar, the Health Check maps it properly before anything is bought.