 
      DATA MIGRATION IN IT PROJECTS
Data migration is one of the aspects of each major IT project. Successful data migration requires planning and execution of a number of significant activities. There is actually a common consensus regarding the importance of the problem, since without data migration (particularly those referred to as the master data) registration of transactions is not possible and the system cannot go live. The question which arises here is why data migration is often a source of so many project issues and how to prevent the complications … The project experience of Hicron Consulting shows that the following critical aspects must be taken into account: 1. Correct assessment of the migration scope (a reviewed, reliable list of all objects subject to migration) The first step which must be taken when migrating data to a new system is to define their scope and design the adequate structures. The data in question reflect the business operations of the entire organization, including not only product indexes, customer data and formulas, but also the standards conformed with, delivery dates, order sizing parameters, transport requirements and many more. It is of crucial importance to identify during the project all business areas from which data have to be extracted and transferred to the new system. 2. Responsibility for migrated data categories; supervision and coordination of work Preparation of data for migration is one of the most time-consuming and labor-intensive stages of any system implementation. On the other hand, successful data migration is a precondition for the success of most of the following stages of the project, therefore utmost care is recommended during the process. The responsibility for preparing data for migration rests mainly with the company employees. When delegating people for the job, attention should be paid to their competence in respective areas, always choosing the most experienced and knowledgeable people. Furthermore, the complex interrelations between data items need to be taken into account, as well as the fact that some data are used in more than one area, which must be addressed when preparing the breakdown of responsibilities for data preparation. Another important aspect is the awareness of the project significance among the company employees and senior management. In order to ensure the success of the project and meet the planned go-live date, the task of preparing data for migration should be high up on the list of priorities, among other routine jobs. An alternative method of data preparation for migration reducing the necessary time investments is to engage in the process external consultants who will be in charge of the subsequent system implementation. Knowing the structure of the new, implemented solution and having access to the legacy IT system, the consultants will be able to provide considerable support in the data migration process and relieve the burden off the shoulders of the employees who will be able to focus on the achievement of the business goals and tasks. 3. Appropriate time management During the preparation of the source data, it is worthwhile to test their quality and coherence in the source application by conducting tests of the business processes on the migrated data in the test system. In this way we can check whether the data are well prepared, sufficiently detailed and accurate, in order to be able to make the potential corrections still in the source application. This approach will help us avoid last-minute data cleansing or even migration of incorrect data to the target system. Incomplete or inaccurate data preparation may lead to delays in the project start or cause problems during the system implementation. 4. Preparation and migration of data – do it yourself or get support? It is often the case that consultants are not engaged in the initial stages of the implementation work, drawing on the assumption that the company employees have a better knowledge of the databases and source applications and can manage to do the job on their own. In the case of more complex projects, however, engagement of experienced consultants with good knowledge of the target system is worth considering. The benefit is reduced workload of the company staff and a faster migration process which translates into a timely production start of the system. Such an approach will also give us assurance as to the quality and accuracy of the data migration process. 5. Excel is not enough: tools supporting data migration Prior to selecting an implementation partner, it is good to make sure what kind of data migration supporting tools the prospective partner offers. In the case of large-scale complex projects, when the repository of data transferred to the new system contains results of several years of the company’s operations, data migration tools will be very useful. It is especially true of large companies with multiple locations and a complex organizational structure. Excel will not do. Properly structured formulas which eliminate data duplication, ensure refining and cleansing of data on the daily basis, allow for saving and validation of the data prior to their transfer to the target application and enable reversal of migrated records, will speed up and simplify the process of data preparation for the implementation project. 6. Test data migration during the solution acceptance testing – “spontaneous” migration is unacceptable! After populating the system with migrated data we cannot go live immediately. Acceptance testing of the solution is required. It is the last chance to introduce the potential corrections. By executing an appropriate number of the company’s business processes in the new system prior to its production start, we will be assured that the system functions properly, is correctly configured and there are no errors during the process execution. An ideal solution is to involve at this stage the key personnel members from the respective business areas who will assess the system performance and identify the potential non-conformances 7. Migration according to the procedures A well run project is a project in which there is no room for ‘spontaneous’ migration, without prior tests and predefined procedures. The success of the entire project is, to a high extent, conditional on the stage preceding the proper implementation i.e. data migration. With a properly designed action plan, we increase the chances for timely completion of the project and eliminate the possibility of potential errors during the process configuration. The procedural approach and attention paid to all critical factors mentioned above is of particular significance for companies with complicated, multi-site organizational structure, distributed and heterogeneous sources of migrated data, a wide scope and high complexity of migrated objects or a large volume of historical transaction data which need to be captured. Projects implemented in such companies face a particularly high risk associated with data migration. Hicron consultants have extensive experience in data migration acquired during the execution of over 200 implementation projects. In some of the projects, Hicron experts were responsible for the management and/ or execution of complex data migration jobs. On the basis of the implementation experience gained during the several years of consultancy work, Hicron developed its own data migration methodology (which forms part of the overall Hicron Consulting Implementation Methodology), as well as tools facilitating the migration process. The data migration services offered by Hicron include: |
