It’s not that things were handled particularly badly in the past. When Burggraaff joined Farmers two and a half years ago he noted that the organisation had in fact invested quite heavily in IT. The only problem was that the approach, typical of business at that time, had been fairly ad hoc. As issues arose, they were solved without too much thought of any encompassing plan. And so it had happened that the business grew successfully – until, that is, its lack of coherence inevitably reached a point where integration was becoming a challenge and there were too many platforms – including outdated ones – to maintain.
The first thing Burggraaff sought to clarify was what exactly the business wanted from its IT operations. “Nobody really knew what we were really doing there,” he says. “There was a basic misalignment between what we thought we were doing and what the business thought we were doing.”
What caused this misalignment? As already stated, it was typical of that time. But, Burggraaff says, it was also a result of people focusing on their own needs without thinking about the business as a whole. At that time Burggraaff’s IT team more or less ran its own operation. They had no real structure where they could bounce their ideas around, where they could see themselves in the broader business context. Confusing the issue was the fact that they were, in essence, responsible for the IT needs of two vastly different types of businesses – the Farmers credit card operation and the retail chain.
It’s not that the reporting structure was completely out of whack. Farmers previously had a general manager of IT who reported to the CEO. In that sense it conformed to the general consensus that CIOs and CEOs should work closely together. Burggraaff, however, does not believe that such relationships necessarily reflect the real influences in business. “You might be on a par with the other general managers or divisional managers but that doesn’t mean much if you are not perceived to be on a par with the different subcommittees where matters are decided and where issues are regularly discussed.”
It was readily apparent to Burggraaff that things had to change. The nine different platforms had to be whittled down. The seven operating systems had to be rationalised. The big IBM mainframe was obviously way too expensive for the size of the business.
Other events were forcing the change. The Farmers card financial business was split off and sold to Fisher and Paykel Finance, while the retail outlets went into private ownership. The change was liberating for Burggraaff, freed from two incompatible cost structures. Retailing IT could now be targeted at 1-2% of operational cost, as opposed to the typical 5% ratio for a financial structure.
Splitting up the business also meant Burggraaff could attack the cost of running the mainframe, supported by 25 Cobol developers . “No one in the world knew what we were running,” says Burggraaff. “The mainframe had been running for 17 years, with all the applications modified and maintained by internal staff.”
Burggraaff’s first objective was to reduce operational costs. He recognised the seasonal nature of retailing and aimed for more of an on-demand IT model. The period leading up to Christmas, for example, would use more resources than at other times of the year. It needed to be flexible enough to allow for that.
Contrary to many other IT operations, Farmers’ IT had become too conservative. New technologies were coming along that could reduce operational costs and speed development. The old mainframe Cobol had become cumbersome when compared with Java and .Net developments, plus the other new programming environments. However efficient Cobol was, developers would eventually turn away from it as the market for their skills dwindled. And, hey, the world had moved on.
Besides, incompatible systems meant higher costs and duplication of files. Burggraaff wanted to achieve one version of the truth, as he describes it, by replacing a myriad of different access and reporting tools with a single, integrated approach. The result, he expected, would be improved data reliability and integrity, plus better planning and forecasting across the business.
For the changes to happen, Burggraaff needed a new mindset among his staff. With that in mind, he introduced a range of Microsoft applications and services under what he called the New Technology Toolkit. They added up to a pleasant change for staff, many of whom had felt their options were limited by being stuck with old technology. “I really wanted to show the IT people that Farmers is a pretty cool place to work at,” he says. “Besides,” he continues in his occasionally quaint English, “nobody in the whole wide world is still doing Cobol in New Zealand.”
As a low-cost measure, the business intelligence team decided to use SQL Server 2000 to develop three pilot datamarts in the finance, supply chain and store operation areas. Together with Peter Mills from Datawarehouse Associates, a three-month training and support period was developed and executed to ensure people from across IT and the business had the same level of knowledge of data warehouse development lifecycles and the various tools on offer.
“The business depended on us developing something much more intuitive to use,” says Burggraaff. “The users needed a system that would give them access to information much closer to what they wanted, and much faster. I mean, everyone was querying the operational system. But the information was coming from everywhere, some of it from the mainframe, some from the other platforms.”
The developers soon got stuck into the BizTalk server tools and Microsoft Visual Studio. They’re still working on the various projects, of course. Integration via the .Net development platform is making it easier to integrate existing back-end systems.
The five-year plan
One of Burggraaff early tasks has been to set up a three-to-five-year plan. As one would expect, it sets targets for retiring the mainframe and restructuring the application base. Work already under way includes introduction of an SAP enterprise resource planning system, coming fully on-stream by next July, at which time the mainframe will finally be retired.
Burggraaff has no doubts that the .Net technologies will handle the new requirements for other aspects of the business. He is well versed in their use, and that experience has probably been a key factor in his decision to go for it, as opposed to settling on Java. “I knew what capabilities I was looking for,” he says. “We explored Java and .Net as possible options for the technology platforms and focused on things like the difference between them from a developer’s perspective. We also asked how much investment had been done in either environment and how many contractors were available for either platform.”
It soon became apparent to Burggraaff that .Net was in ascendance and that it was difficult to find good Java programmers. “Certainly there are a lot of them around but it is difficult to get access to them in the Auckland marketplace. Also, it’s important to be able to see the involvement and interests of the companies behind the toolsets. In this instance, Microsoft was way more active in helping us understand the pros and cons. On the other hand, we know of other people who have been through exactly the same exercise and come out in favour of Java. There are definitely personal opinions at issue here. Whatever religion you follow, you are sure to hold strong beliefs in that religion.”
Then there was the difference in price. Microsoft definitely had the advantage here, says Burggraaff. That made it even easier to go down the Microsoft path. “To be honest, it would have been more interesting if what the IT team had seen of Java had held more attraction for them.”
As for ongoing licensing issues, both teams took different approaches but in the end there were no major differences. If you understand a product’s road map for the future you can anticipate any changes early on, Burggraaff believes.
Another consideration for the Farmers team was open source software. They looked, but Burggraaff continues to have doubts. “It’s nice, yes, if you are at a university or if you want to build your own website. But support is always going to be an ongoing issue, which means you are always going to veer towards commercial products. Either that, or build your own team around your system.”
Burggraaff cites point of sale software as an example of what is wrong with open source. Linux versions of POS systems are available, he says, but companies that succeed with it typically set up their own teams of 10 to 20 people in support of the environment. Nor will you find any SAP systems certified for open source.
Farmers has already begun to build an enterprise data warehouse using SQL Server 2000 and its data mining component, Analysis Services. The success of three pilot datamarts begun in 2003 using SQL Server 2000 led to plans for a data warehouse to combine data from a variety of different systems across the business. SQL Server has recently been upgraded to a 64-bit system and the organisation is using SQL Reporting Services for the production of management reports, previously created with Excel.
Burggraaff talks about the user tools emerging from the new development. Some of them are aimed at picking up discrepancies in business operations so that changes can be made in real time. Another is a promotions tool to manage the flow of information between marketing, merchandising and external promotions.
The big iron, at least in terms of its role in the business, is the SAP system. Burggraaff is happy enough about the fact that it is used widely around the world. If a product represents best practice, he’s all for it. “We implemented SAP Financials last year and we are now at the end of the design phase for the WMS and Retail components of SAP,” he says. Other applications include Compass for merchandise, Dallas for warehouse management and PINT for point-of-sale integration, with BizTalk Server as the middleware to integrate SAP with the existing legacy applications.
Farmers worked with partner Orbiz to implement the first pilot to link its legacy systems with SAP. The business is now developing two additional projects: the first to automate requests to Mainfreight for stock transfers between stores and the second to support the exchange of campaign information with its marketing agency.
The new SQL Server-based loss prevention datamart was developed to analyse Farmers’ point-of-sale transactions for the presence of potential fraudulent transactions. Industry research in Europe and the US has shown that a large percentage of retail stock losses are due to internal fraud, making fraud prevention a major area of focus for the retail industry.
“We are now increasing company-wide awareness of the tool to leverage it as more of a preventative measure, and we plan to create a data mining application to conduct even more sophisticated POS transaction-based analysis,” says Burggraaff.
Another datamart focuses on improved vendor management in the supply chain area. It supports the analysis of supplier performance by reviewing and tracking the delivery of bespoke customer orders (for whiteware or furniture, for example) and identifying any delays. The resulting information can be used to take corrective action and suggest future improvements in the supply chain process.
More datamarts, some with additional .Net applications, are being planned in diverse areas such as:
X Workforce management, the analysis of store sales and appropriate staffing levels.
X Merchandise planning, such as analysis of stock rotations, purchase orders and discounting of products.
X Labour scheduling via the combination of time tracking data with sales data.
X Store KPI scorecards for measurement of aspects such as how individual stores are operating via performance dashboards.
Burggraaff says he did consider outsourcing parts of the Farmers IT operations but he decided against it. It was more important to get things back into shape internally. Outsourcing could not be a consideration until the IT environment was rationalised. “You have got to know what you have got before you consider that,” he says.
Farmers has around 60 stores catering with somewhere between 2000-3000 devices under IT control. New technologies such as voice over IP are on the radar but waiting in the queue, perhaps two years away. Meanwhile, the wide area network is in the process of being upgraded. Satisfactory telecommunication rates have been negotiated with Telecom, which reduces the need for VoIP even further. The technology awaits a business case.
The joy of shopping
“I certainly believe retailing is all about the customer shopping experience,” says Burggraaff. “From a supply chain point of view it means that if I go into a particular store to buy a shirt, it’s important that the shirt I want is there. If it’s not there, I will go somewhere else. For good supply chain operation you have to know where you stock is. You have to know your forecasts, and you need good delivery of goods. That’s basic. Retailing could do much better in New Zealand.”
For Burggraaff it’s a case of playing catch-up with Europe. The new point-of-sale system is already last year’s innovation in Europe. Farmers’ IT is targeting a condensed timeframe, which hopefully will put it on a par or eventually even ahead of its European counterparts.
“We have lost the advantage with eft-pos,” says Burggraaff. “Certainly, we use eft-pos to pay for everything but in terms of technical capability we are not ahead. In Europe you will find things like kiosks and gift cards – plastic card vouchers.”
Burggraaff is mindful of the emotive context behind the customer experience. A shirt is a relatively simple product without too much emotive weight but it is a different matter for a customer buying products such as a leather couch or electronic goods. In these latter situations access to information can make or break a sale. The sales people need innovative tools to improve prospects and the customers need whatever support they can get. A customer needs support on his or here journey through the store, whether that comes from kiosks, the salesperson or online preparation.
“It’s all being explored in Europe. At some stores you can put your shopping list online. Then, when you enter the store, you either use your own device or pick up a device that matches your list with what is available in the store.” It can do more than that, of course, such as offering additional information and other products that might appeal. It can even tell you where to go – whatever you want.
Burggraaff laments the lack of competitive pressures in New Zealand. If one company introduced the latest sales innovations, others would have to follow. The problem is that everyone is holding back. That begs an obvious question: if one retailer showed a little more courage, perhaps it could give itelf an unassailable lead – for a while, at least.
“Wouldn’t it be beautiful if you had shop people who were more customer-focused and service-centric, if they were armed with tablet PCs or whatever technology they needed to improve the flow of information? Wouldn’t it be great if the store manager carried a tablet PC for immediate access to last week’s sales or today’s sales? Wouldn’t it be great if staff could carry an online device to check immediately whether an item was in stock? After all, that’s why people are coming to the store. Wouldn’t it be nice if they could carry their virtual performance dashboard while they are on the shop floor? Most shop floor staff don’t have access to that kind of thing.”
Burggraaff continues with his rhetorical questions. “If you had lots of customers and the lines at the sales counter were getting longer, wouldn’t it be nice if replenishment staff were notified and told to open up new point-of-sale queues?”
The single vision
For the innovations to happen, Burggraaff believes, the basics need to be set up properly. That’s his focus at the moment. The business is no longer being split two ways by having to service its card operations. Having a single, unified vision, means he can focus on what is best for the retail business.
Of course, every vision needs to be challenged. Burggraaff says this is happening under his revitalised governance structure. “Now we have a goal to work towards, and to remain healthy that goal should always be challenged,” he says. “As new things come up, we challenge them and come up with a business case. We have created an IT steering committee, and governance is well established. If a business case is established for a new idea and it is going to make us money, we go for it.”
A big difference is that in the past not enough thought would be given to infrastructure. If a new application ran, say, under Linux, IT staff would just go for it. Now everything has to fit into the big scheme. If it doesn’t fit, that’s bad luck.
It’s also important for IT to show how its work is impacting positively on the business. If no one knows what it is doing, says Burggraaff, IT can be seen as just a black hole. Money goes in, nothing comes out.
Burggraaff’s native directness undoubtedly helps his case. Besides working in the Netherlands, he has been involved in projects in Paris and Sweden. We all have our different ways of doing things, he says. Kiwis like to talk things through. The French are more action-focused. The Swedes take a more consultative approach. And the Dutch – well, they just get on with the job.
Join the CIO New Zealand group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.