Next-gen ecommerce platform
eJam is a company that designs, advertises, and sells a variety of products (e.g. garden lights, washers, toys, etc.).
The system aggregates data which is downloaded from advertising platforms, binds them with assets such as videos and images that were uploaded previously and allows the user to toggle the display, filter the results, and group data. It also allows for the creation of new ads on the FB platform, bypassing the Ads Manager. This process of creating new ads is tailored to your specific requirements, making it faster than using Ads Manager.
What we proposed to the client – our approach & methodology
We explained to the client that when it comes to managing projects, we have very high standards. When we give them a team, we guarantee that they will create a product based on the requirements that the client provided. There was no unique methodology. We had a board in Jira where the client would put in the tasks, set them in priority order, and we would do them one by one.
- There were no workshops or live meetings.
- There were no daily meetings.
- There were weekly status updates combined with demos.
- Most communication took place in Slack. There were also uploaded daily updates on the progress of the project.
The team was changing:
1 Fullstack + 1 Backend
1 Fullstack + 2 Backends
1 Fullstack + 1 Backend
On the customer side, the PM responsible for organizing the work and delivering the requirements was involved mainly.
The Challenge: How did we deal with it? What solutions did we apply?
- In the beginning, the biggest challenge was certainly the short deadline for the MVP delivery (about 2 months) and the integration with an external API (Facebook Marketing API), with which we had no previous experience. It also turned out that the initial requirements provided by the client were changing during the MVP development.
- Then, the next challenge was the ever-changing team, with a lot of time going into onboarding new people.
- There were also problems with regressions, but we were adding new integration tests all the time and managed to get it under control.
- There were also some challenges with integration with additional API (Google Ads API), optimization of queries, and data modeling in the database.
Techstack: what technologies were used & integrations
Frontend: React.js (with support for redux, react-query, redux-saga, styled-components)
Infra: AWS (at the beginning regular VM, then migration to ECS)