Venda

VENDA is a full-cycle shopping platform that is completely built with Python.
The company makes online store from scratch, including SEO,
work with orders, traffic purchase, etc.

Location
United States
Industry
B2B & B2C
Team size
5
Partnership period
24 months
Achievement

Venda technology

Python
Django
PostreSQL
MySQL
Mongo DB
Celery
Key facts
Ready, steady, go
The POC version was created in four months by a small team of developers
Full-stack Python
The system is built completely with Python both for backend and frontend
From scratch
The company makes an online store, including SEO, work with orders, traffic purchase, etc
CRM-side
Venda acts like a CRM, and all the data is transferred to the CRM via IPI
More than 500 people involved
Including Python programmers, support, advisors, sales managers, etc
How Venda was created?

VENDA is the result of 10 years of support provided by our company and we were the pioneers of this direction. The POC version was created in four months by a small team of developers and eventually there were more than 500 people involved in this project, including Python programmers, support, advisors, sales managers, etc. We’ve made a system that generates feeds with products and sends them on different platforms like eBay, Amazon, Google Market, internal and affiliate systems, and more. The average number of feeds was 20. For the first time, the number of clients was around 170.

How we improved Venda?
VENDA grew exponentially, and then we faced the first issue: the technologies wasn’t powerful enough at that time.
Problem definition
We faced the limit of 100,000 products per day. At some point, it wasn’t enough, because there was a lot of clients who could send up to 5,000 products on a daily basis.
Our solution
The solution was to update the system in order to send at least 1,000,000,000 orders on different platforms. Eventually, our system was able to cover more than 2,000,000,000 orders, so the problem was solved. It was a serious technical solution, which we applied.
Database maintenance
The second solution was to change the database structure: we switched from MySQL to Mongo’s NoSQL Database and changed the architecture of the whole application, so all the products could be formed more effectively. We succeeded.

Data analysis & the final steps

Another task was information collecting and analytics.
The CRM stage
We noticed, that we can collect all the data about clients and orders from every platform that we worked with, including our website and online stores of our clients. In fact, the project acted like a CRM, and all the data were transferred to the CRM via IPI. This solution was beneficial both for us and VENDA.
The success
The CRM part of the project started to be actively used, so we focused on technical needs and problems: we provided the uninterrupted work process and make the CRM able to work in the loaded mode (when products are sent and the sewage is synchronized).
Final steps
Eventually, the company was sold to NetSuite, which belongs to Oracle in the present and today it is completely integrated with Oracle’s internal infrastructure.
You have an idea, and we hold an advanced technology stack for its realization.