A.P. Moller - Maersk - Empowering the Pricing Support Team Through Data Analysis

Name:

Vivek Jany, Senior Specialist, India
Vengadesh Nadanasabapathy, Senior Specialist, India
Vignesh Raj KG, Specialist, India
Manoj Kumar A, Senior Process Expert, India
Nick Ulrich Pederson, Senior Data Scientist, Denmark
Gokul Pankaj Seshathrinathan, Data Scientist, India
Julia Evelyn Larsen, Data Scientist, Denmark

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

 

Business Challenge:

At Maersk Chennai GSC, our Pricing Support Team provides end-to-end support to the Global Ocean Pricing and Insurance product managers from A.P. Moller – Maersk. We are a team of 40+ individuals who constantly work on challenging the status quo while upholding the KPIs embedded on us. We aim to go all the way to render maximum efficiency to assist our stakeholders with their pain points in the business.

Our team is part of the upstream process; hence quality is an absolute priority. A small error in a pricing project would cause a downstream impact on thousands of contracts and affect the bookings for a long period of time. This would result in a bad customer experience, major revenue impact, and time-consuming manual corrections with the help of various teams around the world.

Up until mid-2021, we were reliant on the IT team for any data requirement for our business needs. It was taking days to get the data and a few more days of manual complex analysis with help from multiple users. Finally, the end product had to be audited manually once again to ensure 100% quality.

Some of the business challenges we were trying to solve through data analysis:

  • Missing rates for demanded corridors: Without the rates, we were missing out on the business from our customers.
  • Correction of improper product filing: Rates were being filed incorrectly causing revenue leakage.
  • Correction of downstream impact: End-to-end correction was required upon any upstream process changes.

Though we had a repository of all the relevant datasets refreshed and maintained accurately, we didn’t have a tool to communicate with it and customize it according to our needs. We realized it was high time we start looking for one, and it was at that moment, that we came across Dataiku.

 

Business Solution:

We have an in-house Data Science team who helped us set up access to Dataiku. Initially, this was for a couple of users in our team, but it’s gradually increased to up to 10. They collaborated with the Dataiku team and arranged webinars and platform training In the end, it only took us a week to start working on our use cases.

  1. Missing rates for demanded corridors: We connected our data repository through HDFS connection and imported the required datasets, created recipes in the Dataiku tool, and extracted the most demanded corridors that were quoted last season. Then we cross-checked with our pricing projects to see if rates are available for them. We filtered the missing corridors and reached out to our stakeholders to update the respective rates.
  2. Correction of improper product filing: We started analyzing the improper product filing data in the active customer contract lines and found multiple anomalies which wouldn’t have been caught until it is too late. For example, there were cases where the product was filed with incorrect rates or filed as optional when the request was to file it as mandatory, etc. We created recipes to identify those scenarios and fixed them in our system.
  3. Correction of downstream impact: Whenever there is a change in our upstream process, we must mitigate the downstream impact to the utmost minimum. By creating recipes and setting up automated scenarios, we ensured that our process is 100% compliant as expected.

With Dataiku in hand, the way we work has become smarter and more efficient. We have automated our audit process and streamlined it. Nowadays, when we look at a new opportunity/initiative, the first thing we do is to check how we can use Dataiku to support it.

 

Business Area: Product & Service Development

Use Case Stage: In Production

 

Value Generated:

Post gaining analytical knowledge and ease of access to various data points, we changed our approach from reactive to proactive. It got us numerous appreciations and recognitions from our stakeholders and helped us contribute to our organization’s goals and targets.

  • Missing rates for demanded corridors: Millions of USD of revenue were saved for our company through this proactive data-driven initiative and enabled our end customers to make 1000+ bookings in a year.
  • Correction of improper product filing: One Million USD revenue saved by correcting 170k abnormal customer contract rates.
  • Correction of downstream impact: With the automated setup, we correct roughly 100+ affected bookings every month proactively. We also go all the way by ensuring that it will not impact any future bookings.

As we arrested these challenges by integrating this proactive approach into our business routine, we established a strong relationship and trust with our stakeholders and became the first point of contact whenever they approach a hurdle.

The data science knowledge from Dataiku also helped us expand our team’s scope to new businesses. For example, we collaborated with our Pricing Managers and optimized our pricing projects, and shared multiple reports using Dataiku which helped their pricing strategies.

 

Value Brought by Dataiku:

Dataiku has been an integral part of our journey now as the impact it has created cannot be put into just words and numbers. The Dataiku Academy was self-sufficient in training new users, and the opportunity to apply for their certifications for free is commendable. We have also created an internal community where we help each other with various scenarios and queries and keep upskilling together.

With regards to solving our challenges:

  1. Missing rates for demanded corridors: Before using Dataiku, we were spending weeks extracting the data through various other tools and had to analyze it manually. Due to tight deadlines and manual interventions, there were delays and accuracy issues. Dataiku processed tables with more than 2 billion lines to identify the missing rates and handled the analysis part by itself in a matter of minutes.
  2. Correction of improper product filing: The Contract lines dataset in our repository contains more than 1 billion lines and has complex columns with a long list of arrays. So, it was impossible for us to analyze them manually. But Dataiku handles the data analysis part with ease and gave us a neat overview of what we expected.
  3. Correction of downstream impact: We used to spend 100+ hours each month extracting data and analyzing them further. With Dataiku, we were to automate the reports by creating scenarios and saving the respective time and effort.

Saving time while maintaining accuracy has been the key competence of Dataiku. It has also helped our team gain more knowledge, visibility, and empowerment within the organization.

 

Value Type:

  • Increase revenue
  • Save time

Value Range: Millions of $

Comments
Kathir
Level 1

Good one... Congrats..!

VaivaM
Dataiker

Incredible work by @Vivek_Jany and the team! 👏

Santo0304
Level 1

That's some brave work by just not stopping and going beyond to find a solution for the business Challenge through Data Analysis. Fantastic & Keep it up!! 

@ManojKumar @Vigneshraj@Vivek_Jany @VNA089  @NickPedersen, @KatrinaP

Share:
Version history
Publication date:
07-09-2022 10:04 AM
Version history
Last update:
‎09-08-2022 02:54 PM
Updated by: