In this podcast, BI 2014 Singapore speaker Fredrik Ohlsson of SAP customer Tetra Pak describes how the global packaging firm has addressed the challenge of ungoverned business intelligence across its organization.
Specific topics covered include:
- The benefits and drawbacks of the analytic underground
- How Tetra Pak addressed the challenge of ungoverned BI
- The organizational changes Tetra Pak made to support self-serve BI
- Tips on governance strategies for BI in a global organization
- A review of the SAP BI solutions Tetra Pak is using in its current landscape
Dave Hannon: Hello and welcome, this is Dave Hannon with SAPinsider. Joining me on the podcast today is Fredrik Ohlsson, the technical program manager for business intelligence at Tetra Pak, Southeast Asia. Fredrik will be presenting at our upcoming BI 2014 conference in Singapore, which is coming up on October 13th through the 15th in Singapore. Welcome Fredrik, thank you for joining us.
Fredrik Ohlsson: Thank you.
Dave: Fredrik, I wanted to start off by asking a little bit about Tetra Pak as a company, for those who aren’t familiar can you provide us with a quick overview of what the company does as well as your role there at Tetra Pak?
Fredrik: Tetra Pak is a leading packaging and processing company, it’s a multi-national, it’s an 11 billion Euro multi-national, we’re present, we have a presence in 170 markets. We have offices in 100 markets, so it’s a truly global organization.
I have actually two roles, I’m the technical program manager for business intelligence, so I’m heading the rollout of BI in the group. But I’m also the enterprise architect for business intelligence and mass data management, which is more a strategic role where I’m making sure we have a tools roadmap, a technology roadmap for BI and master data, but also that we’re having a sustainable solution globally.
Dave: Ok, great. I know that Tetra Pak has been on a BI journey for quite a while now but more recently you’ve been focused on addressing the challenges of the analytic underground. I was hoping you could sort of explain that challenge for us a little bit?
Fredrik: Yeah, it’s, analytic underground, that’s what we call the, that’s what we call the ungoverned BI and that is something that happens when the globally-approved BI tools fail to deliver the business need, the line of business, they will go out and buy their own BI tool, and they will set up their own BI team. And that is what we call the analytic underground.
There are good things and bad things with the analytic underground; the good thing with the analytic underground is of course that it addresses the specific BI needs. Since the tools are bought locally, you can really get them specific to what you need, and normally also the analytic underground, the local BI tools, they tend to be more agile and more flexible, because the development has gone closer to the business, so the line of business, they like them, but there are some negative aspects of this analytic underground as well. For instance, we lose the economy of scope, having common tools around the world gives us the flexibility when it comes to people moving around. If you know one tool and you can search for information you want to repeat it and move to another place in the organization, or a place in the world and you can easily pick up that tool and you can search for information there.
Dave: So how did you address this challenge within your organization specifically?
Fredrik: We realized that the current toolsets we had didn’t deliver the need we had, the global needs. First of all, we realized that we had a too-centric approach, we had development in two central BI development centers, we had one process, addressing all the needs, and that was a very IT-centric process. We also had tools that were very IT, you had to be IT-savvy to develop tools.
So we realized we had to change the approach, and it was not only to buy a new toolset, we realized that first we need to have the skills in the line of business to create good BI, we had to have another process that addressed both enterprise BI, which is business intelligence we need to make sure that everybody speaks the same language in the company in terms of measurement and definitions and so on, but we also needed an agile, flexible process to address opportunities and situations where we needed repost quick. But we also needed a new toolset, we needed a toolset that addressed big scale corporate BI, but also self-serve BI, the BI that can be done by the end users, on the fly in the line of business.
Dave: Ok, great. In addition to the sort of technology side of things, as you’re preparing for that move to self-serve BI, what organizational changes did you make?
Fredrik: What we did was that we started to look at the line of business and what was missing there because we realized that we had to move the action closer to the business. We had to move development from the development centers to where the decisions were taken, which is the line of business, we realized that we had a knowledge gap there. We had a lot of people that liked creating reports but we realized they were not very good at it. We had a lot of what we call the “cool” analytics but not much effective analytic—a cool analytic is something that is appealing and seductive and it looks good, but it is not actionable. So we had to put people in the line of business that understood analytics, but we also needed people that were tool-savvy, that could actually help them use the BI tools in the right way, we also needed people that understood data.
So in the end, we came up with a new role called BI expert, which is a person that knows analytics, tool-savvy, person that also knows the governance around BI, and when I say governance around BI, it’s about what BI do we need in the organization, and what BI should be managed globally versus locally, right, so we address the right needs. But the person also needs to understand what data we have, in the organization, but also what data can be used externally as well.
So that was the role that was missing and that is not a role that is easy to deploy. So we started a training change management program that has been continued now that has been running for a year now, to identify people and to train them. So that’s an organizational change, a new role.
Dave: Ok, great. Lastly I wanted to talk a little bit about the SAP tools that you’re using as part of this transformation. Can you give us sort of an overview of which BI solutions you’re using, and how you’ve decided where they are best applied?
Fredrik: The answer is that we’re using different tools for different purposes. Starting with the big scale analytics, what we call the enterprise BI, we’re using mainly Design Studio, Design Studio 1.3, for management reports and free analysis reports. As a complement to that we’re also using Analysis Office, it’s a tool that integrates with Excel, and a lot of the users, they use Excel and we needed that capability.
For self-serve BI, we use mainly Lumira. Lumira is a tool that we use as a desktop optimization tool, so users they can take the data they have in Excel, in text files, in SQL databases, and they can put it into this Lumira tool, and they can easily mash up datasets and they can create visualizations for presentations and for their own analysis needs.
Apart from self-serve BI we also use Analysis Office. We have a wealth of information in the corporate business warehouse, and Analysis Office allows users to tap into that data, they can connect to queries, they can extract the data in Excel, in a secure way, because we reused access in the corporate business warehouse, and they can use all the functionality they have in Excel to do the analytics they need.
So those are the tools that we use mainly; we also use, for more canned reports, we use Webi, and Webi’s used in the line of business so the BI experts that we train, we also train in Webi, and that is to address simple report needs that are required on corporate data sources and external data sources as well.
Dave: Ok, sounds like there’s a lot of exciting things going on there. Fredrik Ohlsson, technical program manager for business intelligence at Tetra Pak, Southeast Asia spoke at our annual BI event in Singapore. Thank you very much for joining me today on this podcast!
Fredrik: Thank you!