Information Age (IA): In the UK, Heinz produces 15 million items every week, including 3 million cans of baked beans and 1.5 million bottles of ketchup. With so many food products heading for customers all over Europe, how does Heinz make sure that stock levels are not running too high or too low?
Daniel Haverly (DH): In a business of Heinz UK’s size and complexity, this is a real challenge – but it is also an absolute necessity. There are several factors at work here: our principal goal is to minimise the amount of stock we are holding in our warehouses, because it costs us money to store it – we want it out and sitting on supermarket shelves. Also, we want to minimise wastage, so we need to have a pretty accurate idea of what products we do have in stock and what their sell-by dates are. If we know that, we can allocate stock [to the supermarkets] in the right sequence.
On the flip-side, the one thing we do not want is to find ourselves unable to supply our customers with the products that they want, when they want them. Our key customers are the large supermarket chains – they are key to our business and we want to be sure they get a very good level of service from us. So what Heinz is looking for is the ability to be able to hit the right balance between forecast sales and actual orders. We need to make sure we are producing in sufficient quantities and have sufficient stock to satisfy orders without having stock levels too high. This is a major operation in visibility: the more information our people have and the more visibility they have, the better Heinz can manage customer service levels.
IA: So how is Heinz giving its employees access to that information?
DH: We have been using various business intelligence (BI) tools for the past ten years. But in the past, what Heinz tended to do was use different BI tools for each individual project or business area. We had different BI tools in different regions, and different BI tools for different applications. We had tools from Cognos, Business Objects and Oracle, among others. It was just a case of whichever tool the person controlling that business area thought was the best solution for that area at the time. What we were lacking was a coordinated approach, and to address this, the company decided to standardise globally on business intelligence tools from Cognos.
IA: What do you perceive to be the advantages of this co-ordinated approach?
DH: Well, if you are using a number of BI tools, you immediately have a problem with data standards. For one thing, you have different units of measurement, so you’ll have one group that measures in tonnes, one group that measures in dozens, one group that measures in units. Heinz needs consolidated data standards so that we can establish a common understanding of our business, and so that we can make meaningful comparisons between different parts of the company and from country to country and region to region.
IA: And how is this being implemented at Heinz UK?
DH: In the first instance, we decided to roll out Cognos to one particular project, Heinz UK’s National Distribution Centre (NDC) for ‘ambient’products – foods that are packed and stored at room temperature in tins, sachets, packets and so on. The NDC, opened near Wigan in February 2002, is a new venture for the company, complete with new business processes.
For reporting purposes, Heinz wanted to give users up-to-the-minute information on stock levels, customer orders and lorry loads. In order to provide all this information, we need to draw data from a number of different systems. For a start, we use a number of ERP (enterprise resource planning) systems internally. In Italy and Spain, we use BPCS (an ERP package from SSA Global Technologies), and in the UK we use Prism (from Invensys/Baan). In addition, we also analyse data from a warehouse management system belonging to Wincanton, the company that operates the NDC on our behalf and manages all our logistics.
This was clearly a big integration effort but, in fact, it proved to be far less complicated than we expected. It was more a question of configuring interfaces and configuring tools to make reporting and analysis easier for users. Because the Cognos web-based products (Impromptu and PowerPlay) provide a common desktop experience for end-users, we have been able to hide all the underlying technological complexity from users. They now access business intelligence tools through a standardised-view, browser-based interface that is becoming very familiar to them.
IA: And how up-to-date is the information that users access?
DH: Well, that depends. There are two aspects here: First, Heinz employees require operational reports that provide details for the day-to-day running of the NDC. Using integration tools from DataMirror, we provide frequent updates to a transactional data store that holds all the base data for these operational reports. Depending on a number of other issues, these updates are performed somewhere between every 15 seconds and an hour. As a general rule, the data users see is typically about one minute old.
Second, Heinz employees need to run KPI [key performance indicator] reports that gauge the NDC’s performance against longer term goals. These are held in a series of analytical data stores – akin to individual data marts for each reporting area. This data is updated overnight, so for KPIs, the information users see [dates from] the end-of-play on the previous night.
IA: How long did this project take and what benefits did Heinz experience as a result?
DH: From nothing, a team of six people – a mix of consultants and newly trained internal staff – developed the reporting function for the NDC in under six months. New efficiency gains were realised after just two months of using the Cognos reporting tools. In particular, we were able to qualify the results in terms of substantially reducing stock discrepancies and managing stock more efficiently.
Our eventual aim is to use Cognos products across Europe as consolidating the common KPIs across our global operations becomes a business priority. This will enable all business users to use Cognos business intelligence portal as a ‘one stop shop’ for information, whether the information is local reporting, regional reporting or regional or global analysis. The key area here is the compilation and use of data standards so that we have consistency in definition across all of Heinz’s global businesses.