top of page
close-up-warehouse-view_edited_edited.jpg
Pelangi Logo.png

How We Solved Pelangi's Manual Sales Processing with AI Technology

Overview

Inventory Management, Order Processing Time, Conventional Sales Channel

Every year, Pelangi produces more than 400 new publications in five different languages - English, Bahasa Malaysia, Chinese, Thai and Bahasa Indonesia. With a large SKU in terms of variety and volume, Pelangi faces a huge challenge when it comes to managing an ever-growing inventory.

 

Apart from that, being a large educational publishing company means having to deal with large volumes of book orders before each school season starts. With manual data entry processes, Pelangi's order processing time becomes a problem when customer's purchase orders come in.

Lastly, Pelangi has a large variety of customers - big, small, urban, and rural schools. Many of their distributors still operate in conventional and manual ways which can be inefficient.

About Pelangi Publishing Group

Pelangi Logo.png

Pelangi Publishing Group is a regional educational publishing company that was founded in 1979 by a group of young teachers. The Company is based in Malaysia with local publishing offices in Thailand and Indonesia. Pelangi produces educational materials not only meant for Malaysia curriculum, but also for Thailand, Indonesia and Singapore curriculum. The educational products range from kindergartens to pre-university levels. 

Challenges

Home.png

Large and Complex

Inventory Management

Over 40 years and more than 400 new publications, Pelangi's huge inventory and complex SKU variety (Subjects x Levels x Series x Versions)  complicates how they manage their inventory.

Calendar charts.png

Slow

Processing Time

The few months leading up to a new school season, Pelangi is always under pressure to deliver high volumes of book orders to schools over a short order processing timeline.

Delivery Box change.png

Conventional

Sales Channels

Pelangi's sales channels cover big, small, urban, and rural schools. Many of their distributors still operate in conventional methods which involves large amounts of manual work.

Solution

Since customers’ purchase orders come in different formats, Pelangi had to manually extract from customers’ Purchase Orders (PO) and convert them into their Business Portal.

 

Axrail came up with a generative AI solution based on Amazon Textract and Amazon Bedrock . This has saved a lot of manual work to key information into their system, both in the Operation team and Sales team.

Here's how the Solution works:

  1. From the customer's end, sales entry documents will be scanned.

  2. The scanner will then trigger an email notification to Axrail where Amazon Textract will be used to extract the information found on the documents.

  3. Key information such as Customer PO, Customer ID, Customer Name, ISBN/Bookcode, Quantity, and Sample are collected.

  4. The information extracted from Amazon Textract is passed to Amazon Bedrock running Anthropic Claude 3 model for further processing.

  5. A Key-Value pair database will be automatically populated with the information processed above.

  6. Pelangi admins can then review and verify the extracted information on their web portal.

For most organizations today, the below picture is the normal daily sales activities, where customers will contact you and issue you a PO, and your operations team will manually input the PO into the existing ERP software like SAP or Navision. This process is very time consuming and prone to many errors.​

Manual Sales Processing.jpg

With Axrail Commerce (see picture below), our customer journey remains unchanged and PO will go into Axrail Commerce platform. We use Amazon Textract and Amazon Bedrock running Anthropic Claude 3 to extract the relevant information and automate data input into the ERP software like SAP and Navision.

AI Sales Operations using Axrail Commerce.jpg

Outcome

Clock dash.png

40% Reduction in Manual Processing Time

Pelangi's employees no longer need to manually extract data from customers' purchase orders, which can be time consuming when the volume is high.

user.png

25% Reduction in Manpower

With order entry automation systems in place, Pelangi is able to manage their customers' orders with a leaner team, seeing reduced back-end office manpower over 3 years.

Bar chart upward.png

Increased Customer Satisfaction

With better accuracy, the number of wrong supply cases and customer complaints have reduced, leading to happier customers.

Target.png

Increased Manpower Efficiency

Pelangi's sales team can now focus on sales activities instead of wasting time with manual data entry work whenever they receive customer orders.

Conclusion

Overall, AXRAIL's solutions have successfully solved Pelangi's manual sales processing challenges with AI technology,

resulting in reduced order processing time and manpower, as well as increased customer satisfaction and manpower efficiency.

With a team who is AWS certified and experienced in AWS and consulting, AXRAIL has been serving clients ranging from startups to enterprise, covering multiple industries like retails, media, transportation and well-known service providers. With this being said, an industry like Education is not a difficult industry to deal with!

 

We believe that Education is a very important industry that needs digitalization so that our future generations are able to benefit from this! Pelangi then entrusted us with their problems, believing that we are able to help them get better in their operations!

In conjunction with the AWS Region Malaysia launched, you can try the same document automation experience for 6 months at no cost worth up to RM30,000.00.

Why Work With Us?

With AWS expanded presence in the new AWS Region Malaysia, all of our cloud solutions are now hosted locally, delivering the same global standards right here in Malaysia. To celebrate, we are offering free 6-month trials on any of our innovative cloud services - an unparalleled opportunity to experience the power and flexibility of our offerings at no cost.

bottom of page