Avoid mistakes during ERP selection and implementation

Nothing is more expensive than an incorrectly implemented ERP system

A company that has outgrown the garage stage can hardly manage without an ERP system. But choosing the right system is not as easy as choosing the right word processor: there are currently around 150 ERP system providers with more or less comprehensive systems on the German-speaking market.

What’s more, implementing the system into day-to-day business operations is not something that can be done at the weekend, but requires months of careful preparation. But which employee in a medium-sized company is experienced in the selection and implementation of ERP systems? In many cases, only large corporations can afford such personnel. SMEs, on the other hand, need to consult external service providers – both for the selection and implementation of the ERP system. Otherwise, he runs the risk of making at least one of the mistakes that we repeatedly encounter in our firefighting operations.

Market leader versus industry specialists

When selecting a system, for example, it doesn’t help much to rely solely on the market leaders having the best solution. Even if the market is clearly dominated by SAP in large-scale industry, the cake is far from being shared out in the SME sector. Although SAP is trying to place its R/3 or mySAP.com software product here via SME partners, it is encountering many reservations among SMEs and is obviously not making much headway. SAP’s acquisition of the Israeli system provider TopManage should also be seen in this context. The aim is to make faster progress with a separate SME package. A second software giant – with a stronger presence in the SME sector than SAP – is currently coming from a completely different direction and has chosen the field of ERP systems as a further area of business: Microsoft. With the acquisition of the Danish ERP system provider Navision, Microsoft is positioning its troops in Europe against SAP. But who would want to predict today that Microsoft will be able to continue its success story in this segment? The only thing that is certain is that the requirements in the SME sector are and remain so complex that it is always necessary to take a close look in order to avoid backing the wrong horse.

Project results partly miss the mark

The statistics show that this is not so simple: even if no valid figures are known, there is a public consensus that the original targets were not achieved in 50% of system launches. In 40% of all system launches, the originally planned cost budget is significantly exceeded. And we can only endorse that. In practically all projects over the past 10 years, we have repeatedly encountered the same mistakes and incorrect cost/benefit considerations when selecting and implementing ERP systems.

10 tips for ERP implementation

  1. Think about whether a new IT system is really necessary!
  2. Proceed systematically: First simplify, then select an ERP system, then introduce and train, and finally optimize.
  3. Let the experts design and radically simplify your order processing!
  4. Choose your ERP system to suit your organization and not by board decision!
  5. Select your system according to business criteria and not according to the number of functions available.
  6. Reduce implementation costs and risks by selecting less complex systems and limiting yourself to the essential modules and functions in an initial operating stage.
  7. Involve employees in the ERP implementation
  8. Make sure you have enough qualified application consultants!
  9. First you train, train, train, then you test, test, test!
  10. Don’t forget to optimize after the launch!

Selection error

You can make enough mistakes when selecting an ERP system to lose the fun of the subsequent implementation.

Wrong ERP philosophy

Surprisingly often we encounter the problem that an ERP system does not meet the basic needs of a user company at all.

A Rhenish company in the iron, sheet metal and metal industry produces customer-specific contract products from the slab to the finished product. An ERP system that exclusively supports customer-neutral warehouse production was introduced by Group decision. What used to be produced in a customer-specific production order from raw material to finished product is now produced in a finished goods warehouse via three storage levels and two to three production orders, independent of the customer order. Since the ERP system does not support any other form of organization, the company has no choice but to submit to humility. No one has ever calculated how much money this inverted process organization actually costs. However, nobody claims that order processing and system support work better today than in the past.

System change instead of system renovation

ERP systems are often no different for companies than cars are for families. After a few years, you get tired of the old car and find more and more arguments as to why you need a new one. The selection of an ERP system sometimes has little to do with business considerations.

Switching from an old to a new ERP system is not always unavoidable. In some cases, the old ERP system would still meet future requirements if the effort was made. If order processing were consistently simplified, many seemingly necessary but missing functionalities would perhaps no longer be required at all. It’s worth the effort, because refurbishing an existing ERP system only costs around a third of the cost of introducing a new ERP system. It may be several years before the supposed efficiency advantage of the new system has recouped the conversion and introduction costs.

System complexity too high

Virtually all living ERP systems are becoming increasingly complex as they are filled with more and more modules, functions and “little functions”. The managing director of a medium-sized, very well-known ERP system once told me that he had set up his own business with the firm intention of finally bringing the ERP system to the market that includes exactly the functions that users actually need and that dispenses with all the “IT rococo”. “And where do we stand today?” he asked and gave the answer himself: “Where everyone else stands. We try to offer everything, otherwise the market won’t accept us!”

In fact, the users are just as much to blame for this development as the providers. More and more frequently, companies are selecting systems that are far too complex, while only about 20% of the system functionality and 50% of the costs are needed to handle 80% of the tasks to be performed. Some extensive specifications and catalogs of requirements from user companies could be summarized in one sentence: “One S-Class with all the extras, please!”. On the other hand, ERP system providers are still striving to differentiate themselves from potential competitors by offering specific functionalities. Buyer dreams and the desire to differentiate have led to ERP systems being sold on the market today that the manufacturers themselves barely understand and that no user, not even in large corporations, can reliably master.

Many application consultants are no longer fully familiar with their own ERP module. Their knowledge becomes very thin when it comes to the question of what the actions of their own module do in other modules of the ERP system. What right do you have to expect users to know the interrelationships that take place in an ERP system if even the system provider’s consultants are afraid of pressing the wrong button? However, if the users no longer understand the “mindset” of the ERP system, this considerably slows down the efficiency of the entire company and causes drastic problems during system implementation and subsequent system use.

Remember: anyone who thinks they need a particularly complex ERP system is just too lazy to think about simplifying their business processes.

Lack of cost-benefit analysis when selecting a system

Viewed dispassionately, the purchase of an ERP system is nothing more than the purchase of an operational transportation system. The latter is used to transport materials and goods, the former to transport information. Both represent investments that have to be made. It would be best to design material and information flows in such a way that the support of a technical system is not required.

If the investment is unavoidable, then at least a relative investment comparison should be carried out between the system alternatives under consideration. The first step is to determine realistic acquisition, introduction and operating costs. If the internal personnel costs are taken into account in the introduction costs, it can be roughly assumed that the costs have been reduced by a quarter. Software, hardware, external consulting and internal personnel costs each account for 25% of the costs.

Unfortunately, it often happens that system providers, having beaten the competition to the punch, suddenly realize that the one or other functionality they have been talking about all this time only works with the addition of another module that has not yet been considered or offered. It is often discovered afterwards that the originally planned hardware equipment does not provide the required performance. I know the current case of an automotive supplier where the system provider suddenly realized that additional modules worth 1.5 million euros were needed.

There is also the case of a medium-sized machine manufacturer. Once the implementation project had begun, the software provider realized that the calculated AS400 configuration was not sufficient. IBM had to pay an additional €60,000 for the hardware configuration that was actually required. Thanks to the careful drafting of the contract by the machine manufacturer’s purchasing department, IBM was allowed to send the invoice to the system provider. However, this is the exception rather than the rule, as there is usually a corresponding reservation in the contracts. Nor should we forget the managing director of a 1000-strong company who was told by a system provider that he would need 230 man-days for a release upgrade, which were not covered by the maintenance contract.

However, the mere comparison of acquisition, implementation and operating costs between the system alternatives under consideration only represents the upper part of the shoe and says little about the actual cost-effectiveness of the systems. Only when you add the bottom to the shaft, in the form of the different benefits that result from the different systems, does it become a shoe on which you can walk towards the right system decision.

The greater benefits of comprehensive ERP solutions compared to other systems often have to be purchased at such a high price that the additional expenditure will never be amortized.

Introduction error

Misinterpretation of the “standard”

Many companies want to reduce the duration and costs of a system implementation by “consistently introducing the standard and dispensing with any customization programming. Doing without customization programming may often make sense, but believing in a standard is a mistake. The standard does not exist. By choosing any serious ERP system, you are opting for a kit from which you first have to assemble your ERP system. There are an infinite number of building templates for this, from which you must select the appropriate template and adapt it to your requirements. If you buy a set of Lego bricks, you have not yet bought a house.

Cementing inefficient order fulfillment processes:

Years ago, when many companies were just about to introduce their first ERP system, experts warned that the use of software alone would not lead to superior order processing. Unfortunately, this misconception has still not been eradicated. Many companies still believe that switching to a new ERP system cures a sick order processing system. However, 80% of the difficulties in order processing lie outside the IT systems. It cannot be emphasized clearly enough: the introduction of an ERP system alone neither improves order processing nor reduces running costs, it merely improves the transparency of processes; and this transparency must be used skilfully.

Anyone who does not carefully consider what their future order processing should look like before they have to map it with the ERP system is making the classic, most common and perhaps most expensive mistake when introducing a system. When you build a house, you don’t start thinking about the layout of the rooms when the bricklayers are on site. Certainly don’t leave it to the bricklayers to decide for themselves how the walls and windows are to be installed after a quick look at the family. The results would be geared more towards the needs of the bricklayers than those of the future residents. Usually, you consult an experienced architect who has already planned numerous houses and knows a large number of different solutions with their advantages, disadvantages and risks.

Many companies think they know what their future order processing should look like or believe that their current business processes are fine. The first questions asked by the application consultants when the system was introduced made it clear that no thought had ever been given to the necessary details. As a result, many companies leave it to their system provider’s application consultants to organize order processing. It should be clear whose interests the new order processing will be geared towards, and that this will turn logic on its head.

Instead of adapting order processing to the ERP system and the ideas of the application consultants, the ERP system should be selected according to the requirements of order processing and configured accordingly by the application consultants. The concept for future internal and cross-company order processing must therefore be worked out before the system is selected. The future organizational processes then result in key requirements for the ERP system to be selected.

Saving the wrong amount during ERP implementation

The introduction of a new ERP system is expensive, there is nothing to be said about that. But nothing is more expensive than saving on organizational consulting, IT consulting and employee training. A rule of thumb that has been confirmed time and again in recent years is: if order processing was not systematically analyzed and consistently simplified before the system was introduced, it contains a savings potential of one million DM for every 60 people involved. If you design your order processing before selecting and implementing a new IT system, you will significantly reduce the amortization period of the system investment and drastically lower the implementation risk!

A southern German fittings manufacturer was only able to realize 40% of its monthly turnover in the first month after switching to its new ERP system. The system was missing essential data that had to be mercilessly updated for each individual order, because otherwise the ERP system could not create production orders, print work plans and work cards, plan stocks and create shipping documents. The Managing Director concludes: “This month has cost us more than the entire introduction of the ERP system.

Lack of user training

The first pilot of the American Air Force received the order from his commander: “…and teach yourself to fly!”. A similar attitude prevails in many companies with regard to user training. These are far too short and are often carried out according to the train-the-trainer principle: A few key users are trained by the IT consultants and then pass on their knowledge to the other employees in the company. Unfortunately, we often find that this training method falls short. In order to avoid a crash landing at system startup, expensive retraining is then carried out at short notice.

Remember: It is not enough to memorize which “buttons” you have to press in which order to obtain or enter information. Users need to understand how the buttons are “wired” together; what goes on behind the surface of the IT system.

The example of a medium-sized West German retail company shows what can happen. After a new merchandise management system went live, the warehouse management system collapsed. For weeks, no invoices could be written to major customers; despite running up stocks, the company was barely able to deliver and was on the brink of bankruptcy.

Lack of project management or lack of experience in project management

Every ERP implementation is a complex project that needs to be carefully managed. This starts with the right scheduling and the right project organization and ends with the right tracking of task execution, deadlines and costs. ERP providers are aware of this problem and have all developed more or less detailed project procedures. Not all of them work equally well and not all good application consultants are good project managers. In 15% of all ERP implementation projects, the schedule is exceeded by 30% and in 40% of projects the planned budget is exceeded by 10% to 30%, according to a study by the market research company META-Group.

ERP implementation projects are practically never without tensions. The views of implementation consultants and internal employees are sometimes quite different, with the truth often lying between the two views. As the project progresses, the necessary communication diminishes and at the end of an implementation project, consultants and companies are usually happy not to have to see each other every day.

Most implementation projects lack a neutral moderator who understands both the language of the future users and the language of the IT consultants, who is able to moderate in the event of differences of opinion and who is able to maintain orientation away from the day-to-day business of the project.

It is part of the nature of complex projects that unexpected problems arise, leading to delays and cost increases. No ERP implementation project is spared from this. An investment or cost reserve is therefore essential, but is managed carefully enough by very few companies. It is not enough to monitor deadlines and track costs. Project progress must be set in clear relation to cost development.

Failure to optimize

Experience shows that the path to ERP implementation is planned with good intentions and paved with compromises. In the end, IT consultants and companies are happy that the system is running at all. It is not only in order to keep the number of construction sites low that control procedures and control parameters, e.g. for parts scheduling, are usually stored on a flat-rate basis and not calculated specifically. Virtually no application consultant has this specific expertise. Implementing an ERP system and setting it up optimally are two different things. Nor do you hire your bricklayer as an interior decorator.


Usually, after two to three years, a company realizes that it is using a golden typewriter rather than a real ERP system. In most cases, inadequate support for the company’s own business processes, a lack of or incorrect automation of scheduling processes and poor data quality stand in the way of success. This is no disgrace, because no ERP system has ever been introduced “in one go” with all the functionalities originally required. But if you don’t optimize your ERP system after two to three years, you are constantly losing money due to poor throughput times and delivery readiness levels despite high inventories.

Picture of Prof. Dr. Andreas Kemmner

Prof. Dr. Andreas Kemmner

Talk to us!

We are there for you personally and will be happy to advise you individually on our services and solutions.