The full story
The challenge
The morning shift started the same way every day. Someone opened Amazon Seller Central, worked through the overnight orders, and typed each one into the internal system by hand.
On a quiet day it took an hour. During peak periods it could take three. While that was happening, the fulfilment floor was waiting.
Meridian Fulfilment had built a reliable operation. Their internal order management system was well-established and the team knew it well. The problem was the gap between that system and Amazon. The two did not talk to each other, and the only bridge was a person at a keyboard.
Errors were not common, but they were not rare either. A wrong quantity, a misread address, a skipped line during a busy morning. Each one meant a customer complaint or a return. The team handled them professionally. They would rather not have created them in the first place.
The bigger issue was capacity. As Amazon order volume grew, the manual entry window got tighter. The business was effectively hiring people to manage data entry. That is not a scalable answer.
The solution
We integrated Meridian’s internal order management system with Amazon’s SP-API. The principle was straightforward: orders confirmed on Amazon should appear in the internal system without a person in the middle.
Amazon’s SP-API is not a simple connection. Authentication, rate limits, data mapping, and error handling all require careful implementation. We worked through the technical requirements methodically, tested every order type the client handled, and built a clear logging layer so that anything requiring attention was immediately visible.
The data mapping mattered. Amazon’s order structure does not match every internal system out of the box. We wrote the translation logic to fit Meridian’s specific fields and business rules, so that when an order arrives in their system it is already in the format their processes expect.
Status updates flow back the other way too. When an order is marked as dispatched internally, the tracking information pushes back to Amazon automatically. Customers receive shipping updates without anyone logging in to send them manually.
We ran the integration in parallel with the existing process for two weeks before switching over fully. Any edge cases that appeared during that period were resolved before the manual process was retired.
The result
The morning data entry shift stopped. Orders from Amazon appear in the internal system in real time, already formatted and ready for the fulfilment team.
The staff who had previously spent their first hours at a keyboard moved to work that required their judgement rather than their typing speed. Fulfilment errors dropped because the data arriving in the internal system came directly from the source rather than through a manual transcription step.
Peak period planning became simpler. Higher order volume no longer meant proportionally more admin. The integration scales with the business without any change to the team’s daily routine.
If your operation has a gap between Amazon and your internal systems, that gap has a cost. API integration can close it.
What changed
After go-live, the shift showed up in the week first. Then it showed up in the numbers. Here is what that looked like on the ground.
from the moment Amazon receives the order
between Amazon and their internal system
from go-live onwards
How we approached this