DFS - Managing the IT progress
To guarantee security and provide IT services efficiently the system house of Deutsche Flugsicherung (DFS - German Air Traffic Control) uses IT benchmarks as an instrument to evaluate the services of administrative IT in terms of quantity, complexity, quality and unit costs. The benchmarking process is an established one which is run on a regular basis.
Deutsche Flugsicherung (DFS) is one of the leading global providers of flight security. The requirements of the customers (airlines and passengers) for the highest possible security and quality levels not only relate to the flight security systems which monitor over three million flights per year in German and neighbouring air space; DFS also regularly analyses its classic corporate IT, known as business support systems.
The first benchmark project in 2004 had a genuine “Aha effect” remembers Andreas Winkelmann, Manager of the internal system house at DFS. The aim was to make costs transparent and verify the complexity of the business support systems (BSS). At the end of the day, according to Winkelmann, the classic administrative IT was always overshadowed by the flight security systems. “The benchmark enabled us to show that BSS were extremely important to DFS and that they had to be in the top league within the market“, continues Winkelmann in describing the effects of the project.
On the other hand the first benchmark with Maturity was not aimed at achieving cost leadership. “Complexity, customer focus and service quality always have to be viewed in correlation to the unit costs”, emphasises the system house boss. In addition Winkelmann’s colleague, Jörg Kundler, was given a firm negotiation base for business and IT services. The Manager of the IT Services Department in the system house controls the IT service management (ITSM), the clients and the system operations of the data centre with a staff of 75. During the introduction of an internal performance account review it became necessary to hold discussions about costs with internal customers and to evaluate the services by means of neutral benchmarking in market conditions. The aim was to analyse the service focus which had been introduced previously and the range of the BSS service modules. In addition to the defined performance, quality and complexity of the services, the required prices were compared with the latest market data. “We can now monitor our unit costs, evaluate developments and include market trends ourselves using the continuous process”, reports Kundler.
Benchmarking data at the touch of a button
After what is now the third benchmark project, DFS is now able “to supply the required data almost at the touch of a button“, says the service manager. As a result of the experience and expertise that the company has now gathered, IT can now provide information quickly, enabling it to complete benchmarks relatives speedily, or at least much more quickly than on the first occasion. System house boss Winkelmann compared a one-off benchmark to a “flashing light”, something that only provides a rough guide. However, the real objective is not only to establish the location but also the route towards continuous improvement, “The first flashing light should provide us with a clue to how we are doing relative to the market, whilst additional benchmarks enable us to see how we are developing relative to the market.”
Although a benchmark can provide information on the performance of the department, it provides nowhere near enough information to enable an internal IT department to provide transparency to its customers showing that its services are being provided at market prices and also that the services are tailored specifically to the needs of the company. To achieve this a regular process must be established. One important and specific finding, for example, was that even after repeated benchmarks differences from the comparison measure of the peer group still occur. These have to be analysed in detail and their causes must be identified clearly.
This meant, for example, that it was necessary to adjust the cost structures after the introduction of the focus on service. In addition, an improved IT service architecture was constructed, explained service specialist, Kundler, “Previously we worked with flat-rate values and based our work primarily on a consumption and customer based calculation of the IT services“. If, however, the cost structures are changed, there will inevitably be consequences in the benchmarks. The costs of major projects also had a direct effect on the prices of IT services. When compared to the market, however, these expenses were levels and only a specific percentage of the project’s costs could be included.
Identify and eliminate negative influences
“We will not be able to be the best organisation for every two-year comparison in every department”, admits IT manager, Winkelmann. That is why it is even more important to identify the reasons for differences. “The main thing is to identify the negative influences and eliminate them”, to which the benchmark makes a valuable contribution. At the same time the market continues to develop and the modelling of the services is also changing. The managers must therefore find the ideal range of services before deciding on the pricing models - depending on the situation and requirements in transparent and direct form or in bundles. All of this has an effect on the current benchmark data.
The choice of peer group for comparison purposes in the benchmark project was by no means a trivial matter, since after all the work carried out by DFS is in fact unique in Germany. As a result of this the recorded figures must be normalised against the complexity and the services, the SLAs and the quantities. ”We have a certain quality and using it we can find a comparison with other IT organisations”, says Kundler. Using this normalisation of the quantities, the service quality and the complexity of the services, the services are therefore standardised to such an extent as to make a comparison possible regardless of the specific features of the company. The main factor in this is that a clearly structured method is used in the benchmark, transparency and reproducibility are maintained in full and the benchmarker has a statistically significant database.
Price and performance
For Andreas Winkelmann, the boss of the system house, it is also important to tell the people who do not work in IT, namely those who work in other departments, about the findings of the benchmark process. Although IT is always interested in identifying its problem areas and rectifying them, “it is also important that we can show that our services have levels of quality, speed and costs that at least bear comparison with the rest of the market.” The accusation that IT is too expensive can now be answered clearly with objective arguments. “We know what we produce and what the customer would normally have to pay in the market”, says service manager, Kundler.
An analysis of the own strengths and weaknesses has another positive effect – namely on the IT personnel. Initially they were sceptical about the benchmark, but the instrument has now become part of the established procedure. Ultimately it is now possible to demonstrate how good the IT experts at DFS really are and what they do, reports Kundler. “Every benchmark makes us more confident and self-assured”.
Download the case study IT benchmark at DFS as a PDF file