A.P. Moller - Maersk - Leveraging Dataiku to Improve Global Invoice Quality and Reduce Customer Pain Points

Name: Alora Pedro Vaz

Title: Transformation Manager

Country: India

Organization: A.P. Moller - Maersk

A.P. Moller - Maersk is an integrated container logistics company working to connect and simplify its customers’ supply chains. As the global leader in shipping services, the company operates in 130 countries and employs 80,000 people. The Maersk GSC, which is an enabler of Maersk’s transformation plays a critical role in delivering customer outcomes and enables decision-making and prioritization for Maersk based on its end-to-end process view. With a strength of more than 12,000 employees, Maersk GSC is spread across India (Bangalore, Chennai, Mumbai, and Pune), China (Chengdu), and the Philippines (Manila), with a small hub in Morocco (Tangier).

Awards Categories:

  • Value at Scale
  • Moonshot Pioneer(s)
  • Most Impactful Transformation Story

 

Business Challenge:

At Global Service Center (GSC), we are a team of 13 individuals who constantly work on challenging the status quo and assisting our stakeholders with their pain areas in the business. Multiple systems coordinate and exchange data to ensure the right invoice is triggered to the customer.

An issue with data in any of the systems would result in incorrect invoices being sent to the customer, thus resulting in increased E2E time for settlement of invoices, working capital, customer dissatisfaction, customer efforts to raise disputes, etc.

The following are some of the issues which we were trying to resolve:

  1. Identify shipments with rates different than the agreed rates.
  2. Identify the causes leading to incorrect rates in the system.
  3. Plug gaps by correcting rates in the system and shipment.

 

Business Solution:

In March 2021, we initiated the project by creating the recipes. Later, data scientists Nick Ulrich Pedersen and Gokul Pankaj Seshathrinathan helped to set up the pipeline in production. Below is a list of things that were enabled through Dataiku.

  1. Visibility on top issues which were impacting invoice accuracy.
  2. Pipelines helped us with the shipments with incorrect rates due to different issues.
  3. We coordinated with different teams to ensure correct rates are updated in the system and shipments.
  4. Enabled users to change focus from manual analysis to collaborating with stakeholders to resolve issues through detailed analysis.
  5. Enabled to create analysis to monitor the efficiency for system fix.

 

Business area enhanced:

  • Lessen the load on analytics as recipes could be built with limited IT knowledge.
  • Helped internal pperations to update correct details in the system.
  • Created visibility to Sales on incorrect rates.
  • Product development by fixing bugs.
  • Enabled data visibility to other teams e.g., Risk/Compliance/Internal Audit.

Use case stage: In Production

 

Value Generated:

This use case helped us to reduce customer pain areas. Our achievements are as listed below:

  1. We were able to expand the check for incorrect rates to 5,000+ satisfied customers.
  2. Helped improve Global Invoice Quality by ~0.3%.
  3. Helped reduce customer pain by raising disputes for ~100K shipments with incorrect rates.
  4. Helped weekly reconcile/correct invoices discrepancies in millions.

 

Value Brought by Dataiku:

Multiple systems interact with each other to ensure the correct invoice is triggered. Extracting data from individual systems was time-consuming and also cumbersome to manage.

Huge data for billions of shipments:

  1. Helped to integrate data across systems.
  2. Enabled us to gain insights and create value out of data across customers.
  3. Enabled us to expand logic globally across ~5,000+ customers.
  4. Enabled improved Global Invoice Quality by ~0.3%.

 

Value type:

  • Reduce customer pain to dispute invoices.
  • Helped avoid revenue leakage.
  • Improved invoice quality.
  • Increased customer satisfaction.

Value range: Millions of $

 

Share:
Version history
Publication date:
03-09-2023 03:13 PM
Version history
Last update:
‎09-15-2022 05:13 PM
Updated by:
Contributors