Customers are increasingly demanding access to real-time data, and freight transportation provider Estes Express Lines is among the rising tide of enterprises overhauling their data operations to deliver it.
To fuel self-service analytics and provide the real-time information customers and internal stakeholders need to meet customers’ shipping requirements, the Richmond, VA-based company, which operates a fleet of more than 8,500 tractors and 34,000 trailers, has embarked on a data transformation journey to improve data integration and data management. Like many large organizations, prior to this effort, data at Estes Express Lines was spread across disparate data sources, which meant that each agile project team had to write its own code to access data from those source systems.
“Besides impacting customer experience, the absence of a seamless data integration and data management strategy was adversely affecting time to market and draining valuable human resources,” says Bob Cournoyer, senior director of data strategy, BI and analytics at Estes Express Lines.
Data woes impact business success
With shipping concerns coming under greater scrutiny, Estes Express Lines customers are increasingly interested in up-to-the-minute details about their shipments, such as expected charges, delivery time, and whether their goods are damaged or not. While the company had a data warehouse, it was primarily used for analysis. As it was batch updated every 24 hours, it didn’t work in real-time.
“Since the data was living everywhere — in the cloud, on prem, in multiple databases throughout the organization and even on desktops at some point — we were unable to fulfil the needs of our customers. It was frustrating for both the customers and those serving them,” says Cournoyer.
Pulling data from multiple sources and then sharing it in a common way was also taking a toll on the company’s IT department. “Our cloud-based systems are very specific and disparate in nature. For instance, we had Salesforce CRM to manage our customers and Oracle ERP for our back-office functions. A lot of times data from all the different systems needed to be combined into one, which was a tedious process. Users couldn’t self-serve themselves and we had to assign a resource to them to satisfy that need,” says Cournoyer.
Under the old system, IT would have to write ITIL processes to source the requested data, which would then be moved to another database to be accessible to the business user, as opposed to giving a direct connection to the actual data source. “Every time somebody made a new request for a new piece of information, we had to touch the code and go through the entire testing lifecycle. It was frustrating for the business, to say the least,” Cournoyer says. “At one point, I had 15 people on my data team and seven of them were engaged only in data analysis.”
Those data bottlenecks also led to delayed time to market. “Whenever we needed to deliver a solution that was going to add value to the business, we had to build in all the extra time needed to source data and do data analysis, potentially write code. Depending upon the complexity, this could add six to eight weeks to a project,” he says.
In addition to these challenges that urgently warranted a data management platform, Estes also had a mission to reduce technical debt. As Cournoyer says, “We didn’t want to keep digging the hole deeper. Copying and moving data has its own costs associated with it and we wanted to do away with it.”
Future-proofing Estes Express’ data strategy
Considering these challenges, Cournoyer set about developing a data strategy aimed at making data available to internal business users and IT systems in real-time without creating any technical debt.
“To start with, the entire IT department was reorganized. The data team was decoupled, and all the data analysts were formed into agile teams so that they could support whatever the data needs would be. We then started our exploration for a platform to solve the data problem,” Cournoyer says.
Estes Express Lines evaluated all the big players, including IBM, before deciding to leverage Denodo’s logical data fabric to access all its enterprise data and have it available in one central location.
“Before deploying the solution, we decided to do a six-week proof of concept. We picked a couple of key areas of our data that were the most requested in the company and virtualized them, which formed about 10% of our entire data universe. We built and delivered some APIs on top of it within the six-week timeframe, and we did it with the internal team that had never seen the system before. That’s how easy it was to learn and use the new solution,” he says.
At the end of the six weeks, Cournoyer and his team “were able to approve two or three key concepts back to the business,” and the proof-of-concept work was rolled over to the next project. “During this time, we were able to map over 50% of all our data and started to use some of the more advanced features of the product. Now, a year and a half later, we’re well versed in it,” he says, adding that the freight transportation provider now has “well over 90% of the data in the organization completely mapped.”
While Estes Express chose an on-prem implementation because it still has a large presence of operational data on premises, the data fabric covers all the company’s internal and cloud-based data sources, delivering real-time data consistency by establishing a single source of truth.
Ramping up CX, slashing time to market
With the logical data fabric in place, powered by data virtualization, Estes Express is now able to manage, integrate, and deliver data to any user, in real-time, regardless of the location and format of the source data.
“Our customer care representatives now have information at their fingertips and no longer fumble or search for it. This ability to deliver value back to our customers and to our internal customers as well has been huge. Unprecedented insight into where shipments are and how they are moving through systems provide an optimal customer experience,” Cournoyer says.
“We measure the sentiments of our customers through a third-party company. They have come back and told us our numbers have gone up. Besides, we can analyze customer scores and perform sentiment analysis to adjust offerings to better the customer experience,” he says.
The new data strategy has also reduced the time to market. “It used to take us weeks, and months in some cases, to deliver solutions. We can now do it in days and even in hours. Reduction in time to market helped us deliver data faster to applications business users and has also reduced our labor cost by 10%,” he says. By enabling centralized, consistent data to all projects, post deployments issues have also come down, saving the company time and resources.
The IT department no longer needs to move and store data, which has reduced the company’s technical debt by cutting down the number of SQL databases, lowering license and storage costs.
The new strategy has also helped Estes Express bring API development back in-house. “We were paying a third-party company to build APIs for us. It used to take us six to eight weeks to get an API but if the requirements changed in the middle of that cycle, they had to go back and reset. With this new data platform, we built a couple of APIs in two hours. I don’t know how to put a number on that but our reliance on third parties to build APIs has gone way down, which has been a huge cost savings for us,” Cournoyer says, adding that the data fabric–based strategy has also laid the foundation for the company’s new data governance program.
Data Governance, Data Management
English