Leading logistics firm SiCepat Ekspres scales to 40 million packages a month across Indonesia with MongoDB Atlas on AWS

INTRODUCTION

Explosive growth in e-commerce since COVID-19 drives complex data requirements for logistics

SiCepat Ekspres has emerged as the powerhouse in end-to-end logistics services in Indonesia achieving massive growth over the past two years due to the explosive demand in e-commerce since the start of the COVID-19 pandemic. Founded in 2014 to provide last-mile deliveries for small merchants, SiCepat has since expanded to serve all major e-commerce platforms. Its services now also cover warehousing and fulfillment, middle-mile logistics, online distribution and food deliveries. The business has 1,300 branches and 31,000 delivery drivers cater for Indonesia's fast growing population of more than 275 million people.

SiCepat focuses on e-commerce and social commerce for people who sell goods through their social media networks.

Testament to its success as a leader in technology-driven logistics, the company raised a $170 million Series B funding round in March 2021 and is poised to expand into other markets across South East Asia.

THE CHALLENGE

How to achieve 4X growth and scale while managing complex data in real time

The logistics industry in Indonesia is highly fragmented, which means higher costs for businesses. At the same time, demand for deliveries is increasing thanks to the growth of e-commerce, especially during the pandemic.

SiCepat runs most of its business through a single app that needs to incorporate all its complex logistics data, from orders with individual e-commerce partners, to pickup and delivery via individual drivers and branches in specific locations, all in real time.

SiCepat’s technology team started with a simple monolithic style of architecture using relational database SQL Server on AWS. However as the company grew and deliveries increased the team began to run into performance problems and scaling issues. Due to the underlying data model of the existing infrastructure, updates to the database or incorporating new features become increasingly difficult and time consuming.

In 2018, Reynaldi Oeoen joined SiCepat as CTO leading a development team of just 6-8 people. He realized they needed to move from a monolithic architecture to microservices and a more flexible data model to support SiCepat’s scaling aspirations and to continue innovating quickly. “Our tracking and order fulfillment app has a direct and critical impact on business revenue, logistics and customer experience. “Data is everything for us – from day-to-day business decision making, through to customer experience and on the ground operations. Everything relies on data so finding a database we could trust was absolutely vital,” Oeoen said.

THE SOLUTION

MongoDB’s modern, multi-cloud database platform solves multiple logistics challenges

Given SiCepat’s complex and rapidly changing needs, they turned to MongoDB for an integrated, modern, multi-cloud database platform to accelerate and simplify how they scale and build new features.

“We’d heard good things about MongoDB’s fully managed Atlas service, so we decided to test this for our partners' order data and some billing applications. It was an immediate success,” Reynaldi said.

After initial pilots proved successful in 2018, SiCepat began introducing MongoDB Atlas for more and more use cases. This immediately helped developers move faster and the applications scaled seamlessly.

Today, nearly a third of all of SiCepat’s data is managed on MongoDB Atlas hosted on AWS. Most important of which is the logistics application that sits at the heart of the business.

SiCepat has taken advantage of a number of operational features to support the business as it scales and ensures a great user experience. These include:

• Indexing – by taking advantage of MongoDB's rich indexing functionality, the team is able to keep data efficiently indexed for optimal query speeds.
• Atlas Online Archive – most data is only 'hot' (i.e. needed in a timely manner) during the delivery itself, after that, the data can be archived to lower-cost cloud object storage yet remain available for querying and analytics. Online archiving lets SiCepat set automatic rules to keep data available when it's needed but archive when it's no longer likely to be required.
• Continuous Cloud Backups – Atlas's Cloud Backups provide localized backup storage automatically so, in the event of an incident, the database can be easily rolled back to a previous state.
• Sharding – Sicepat are now looking at how to scale further and take advantage of native sharding in MongoDB – where data is distributed across multiple machines to improve performance at even very large data volumes.

“MongoDB has been fantastic and really delivered for us operationally. We continue to explore the full power of the platform as we scale. From optimizing queries right through to sharding, MongoDB is constantly helping us grow without slowing down our development," explained Oeoen.

The development side too has been enjoying the advantages of MongoDB's document model. When it came to building out the mobile application developers preferred using MongoDB Atlas because the support for object storage was more natural to the way they coded and speed up delivery.

Another significant advantage was that MongoDB's modern, multi-cloud database platform also extends to a native mobile database, MongoDB for Mobile (formerly known as MongoDB Realm). The development team used MongoDB Realm to build out the side of the application that would sit on the device itself. They're now exploring using Atlas Device Sync to automatically handle all syncing between the device and the main database even when there are single outages or data conflict resolutions required.

For almost all new innovative applications, MongoDB Atlas has become the default. For example, recently, the team built a clever new tool for executives to monitor key daily business metrics. They built a chatbot that scrapes company data and can give simple and clear answers to executives' questions such as – how many packages are out for delivery today or are any deliveries delayed. The app and all of its data is built on Atlas.

THE RESULTS

Seamlessly scaling to millions of customers and a huge spike in activity during the pandemic

MongoDB Atlas has enabled SiCepat’s tech teams to focus on building and enhancing its app features for its six million customers. Not only has the business been able to quadruple the number of packages it is capable of delivering to as many as two million per day, but the organization has also reduced processing time by more than 98% (from three hours to around two minutes).

The stakes are high and the volume of data running through SiCepat’s system is best put into perspective through the lens of its monthly package deliveries which have doubled from 20 million to 40 million over the past year as more people shop from home during the pandemic.

"We have an infrastructure maintenance team and, to be honest, they have to spend a lot of time supporting and upgrading our relational database estate. But we can proudly report that MongoDB Atlas is the one service they never need to worry about,” said Oeoen.

SiCepat is part of a group with eight sister companies in different industries, all with rapidly evolving technology needs.

“We are on a journey to become a one-stop shop for e-commerce, upgrading our legacy systems and developing a system for functional data to integrate across all eight companies in very different industries. MongoDB’s modern, multi-cloud database platform is playing a key role to make vital data part of that equation in a simple and stable manner,” said Oeoen.