In the 21st century, shoppers don’t just rely on word-of-mouth recommendations anymore — they lean on the Amazon reviews of their 200 million closest friends. Reviews are one of the most effective ways to boost the brand’s conversion, credibility, and overall eCommerce presence. Reviews greatly influence conversion. In fact, 22% of shoppers won’t look anywhere else once they’ve identified an Amazon product they want to buy, and reviews are a major push when it comes to purchasing decisions. Research shows that 84% of shoppers trust online reviews as much as a personal recommendation, and 91% of shoppers occasionally or regularly read online reviews. ARequesty is a service for Amazon sellers that allows requesting reviews for the entire Amazon store in 1-click. It adds the “Request a Review” button to orders in the “Orders” page on Amazon Seller Central and keeps track of reviews already requested. Additionally, it allows adding the “Request Reviews For Entire Store” button to request reviews for all orders in the store. Our solution Since we joined in the middle and the client used one of the boilerplates to build the project initially, we were limited to the stack that isn’t popular for multi-tenant applications per user: Flask, PostgreSQL, and SQLAlchemy. It was a great challenge for geeks like PLANEKS, and we brainstormed and tried a lot of different solutions. The final one was to create wrappers for requests to the DB, handle each request, and select the current user's appropriate schema. Obvious solution is the optimization of the background task that performs requests to another service. The worker ran for many users in the system, so we decided to use an asynchronous approach for fetching data from external API for users. Additionally, we used requests to DB asynchronously. It helped to seriously decrease the time of execution. Quite classical situation - client’s previous developer didn’t communicate and reported well while working on the project, which resulted in the slow and incorrect implementation of certain features. PLANEKS regularly helps clients in such situations and solved the challenge by applying an internal framework for communication between the client and a dedicated Python developer. As a result, the client was always up-to-date about current progress, blockers, questions, and reports, which freed him from worrying about essential needs that companies should provide by default. Dealing with 3rd party APIs is a regular task for developers, while Amazon MWS requires a deeper marketplace understanding. Having a strong experience in various eCommerce automation, especially for marketplaces like Amazon, eBay, Etsy, Depop, Mercari, Poshmark, PLANEKS provided a dedicated Python full-stack developer with relevant skills. As a result, the implementation of features went smoothly. Results: The client outlined the requirements and explained the technologies and APIs in play. PLANEKS helped map out the challenges into smaller tasks, provided a top-notch dedicated developer who was a great fit for the job, and took care of the project from start to deployment, end-to-end.