Royal Mail puts big data "building blocks in place" with Hortonworks Hadoop platform

royal mail
© Royal Mail

Royal Mail says that turning to Hortonworks was like the “Big Bang” for them in terms of an advanced data analytics strategy

Share

For a company that is tasked with safely delivering billions of letters and parcels across the UK and beyond every year, it is comforting to see Royal Mail taking a steady, sensible approach to big data. (See also: what is a graph database?)

Speaking at the Hadoop Summit in Dublin this week, director of the Technology Data Group at Royal Mail, Thomas Lee-Warren told Computerworld UK that its Hadoop investment is the foundation of a drive to gain more value from internal data.

This has involved using Hortonworks Hadoop analytics tools to transform the way it manages data across the organisation, freeing the analytics team to deliver insights on proprietary information held in its data warehouse.

“We have a lot of the building blocks in place, key partners in place and want to start delivering on those ROI’s, because we can see them, and now we are chasing them,” he said.

Read next: Why simplicity - not speed - is key to enterprise Hadoop strategies

Lee-Warren is responsible for business intelligence, commercial insight and analytics at the historic company. “I am tasked with changing the concept of data, which is something which appears on reports, into something which changes the fortunes of the company.”

He says that Royal Mail has a relative scarcity of resource to throw at gaining value from its data, so it has to focus on specific returns on investement (ROI).

Now it has put the infrastructure and tools in place for its data insights team of around fifteen employees to investigate “every part of our business and how we engage with our customers,” says Lee-Warren. “We are looking at how people are using our services. Analytics can be really important in those areas.”

Big bang

"We have a lot of data,” says Lee-Warren. “We are about to go up to running in the region of a hundred terabytes, across nine nodes.”

Royal Mail decided on a Hortonworks deployment of the open-source Apache Hadoop software to manage this data following an extensive tendering process. “It was almost like the Big Bang, before Hortonworks there was nothing.”

Lee-Warren speaks about how he wanted to free up his data insights team from spending ninety percent of their time “ferrying data backwards and forwards” from its Teradata data warehouse, to spending ninety percent of their time exploiting that data and making it available to the rest of the business.

Read next: How Hortonworks helped British Gas unify its customer data in "a true open source environment"

“We’re accelerating that whole process, we’re not having to spin up projects just to get data. We are able to accomplish a huge amount of work with single individuals,” says Lee-Warren. “We see Hortonworks as our advanced analytics platform.”

Churn modelling

One project Lee-Warren’s team has completed is around churn modelling, trying to cut down on customer attrition, “so looking at how we could help our business colleagues identify those customer that were more likely to churn in particular industries.”

Using Hortonworks DataFlow (HDF), Royal Mail “went in, got the data, maybe not as close to the edges as other people are working, and moved that into Hortonworks DataPlatform (HDP). We then used Hortonworks to run the analytics, come up with the churn models and then used HDF to actually send that back into our visualisation.” From there the marketing and sales teams could identify where attrition rates were high and take action to change it.

Read next: Jaguar Land Rover is using Hortonworks' Hadoop technology to harness its growing data volumes

The advantage of using HDF, says Lee-Warren, is “the speed. And we are doing a lot of experimentation and I think the really exciting bit will be how we integrate this within our more conventional tool set.”

Skills gap

The skills gap for big data experts was a common topic during the Hadoop Summit, but it’s not an issue Royal Mail has faced.

Lee-Warren said: "I don’t know if I am lucky, and it may be because we have a very attractive brand, but we’re not finding it difficult to attract strong talent. A lot of the time I think data scientists get locked into a way of working that they find difficult and they like new challenges all the time, and we can provide that."

Conclusion

Royal Mail may not have delivered any eye popping projects yet but Lee-Warren and the rest of the Royal Mail group appear to have taken a very sensible route to deriving value from its large data volumes. By building out its infrastructure, partner network and focusing on hiring the right talent the building blocks do appear to all be in place.

As Lee-Warren said: “So really it’s watch this space and in a year or so we will have some exciting stories to tell.” We will keep you posted.

“We’re accelerating that whole process, we’re not having to spin up projects just to get data. We are able to accomplish a huge amount of work with single individuals,” says Lee-Warren. “We see Hortonworks as our advanced analytics platform.”

Find your next job with computerworld UK jobs