When I joined The Pokémon Company International's e-commerce team in 2021, the online store only existed for the US and Canada. My first major project was helping expand to the UK — the company's first international storefront.
The Problem
Launching a new country storefront in Fluent OMS required running dozens of GraphQL mutations by hand like retailer setup, location config, product catalogs, payment routing and some of them in sequence, every time. During the UK launch, this took weeks of incremental work. And doing it by hand meant copy-pasting values into each mutation manually, which created its own problems: misnamed warehouses leading to another update mutation. This had to be done for test environments and then production.
With Australia, New Zealand, and Germany on the roadmap, this process was going to repeat.
The Insight
For the UK launch, this was just the work. You learn the system, you run the queries, you verify each step. But somewhere in the middle of it I kept thinking "we're going to do this again." Australia is next. Then New Zealand.
I brought the idea to the team: what if all of this configuration lived in a file, and a script handled the execution? The team had been thinking along similar lines and we aligned quickly. A coworker pitched a perfect name, FAST: Fluent Automated Setup Task. A new storefront is just a new file.
What We Built
A JSON configuration file updated per market containing all storefront-specific values, and a Python script that reads the file and executes the corresponding GraphQL mutations in order. The JSON handles the what, the script handles the how.
Impact
- APAC and Germany storefronts: one file update + one script run
- Setup time reduced by at least 2 sprints per launch
- Eliminated manual copy-paste errors and naming inconsistencies
- Reused for test environment provisioning