Data migration plan
Which data migration strategy is right for you?
Based on a preliminary investigation of the data migration with a data migration plan as deliverable, we can provide you with the right answer.
Data migration plan
Type 2 Solutions uses a standard data migration approach for retrieving, transforming, validating and loading data.
Preliminary study
We start with a preliminary study of the data migration with a data migration plan as deliverable.
The data migration plan includes:
- The data migration strategy
- The action plan
- A risk analysis
- The data migration process
- The control framework
- The data quality optimization process
- The planning, the required resources, and the tools
Data migration approach
In order to be able to work with the new system(s), data will have to be migrated from the legacy systems.
Two approaches
There are two approaches for migrating the data: all at once in a single run (live with a big bang) or divided into multiple subsets over a period of time.
Live with a big bang
Migrating all data at once is called going live with a big bang. This is preferable from a data migration perspective.
The initial situation only needs to be made consistent once, the check only needs to be done once, there is no need to take into account conflicting data in the target system and the fallback scenario is much simpler.
Advantages
Also, from a system perspective it seems obvious that the big bang scenario is preferable.
Not only is the number of portals, interfaces and systems that have to be kept up and running during the transition phase of a divided migration much higher (100%), special processes, interfaces and perhaps systems have to be set up for the transition phase.
More transparent
Finally, the big bang scenario is often more transparent and straight forward for both the customer and the organization. During the migration period, both the legacy systems as the new system are unavailable. This means that it this period is critical and should be kept as short as possible.
For these reasons, several trial migrations are performed with representative data prior to the “real” data migration so that the lead times and data quality issues can be precisely determined. An additional advantage is that this data can be used during testing, so that the results of these test scenarios are representative of the final production situation.
Data migration example
John Wittekamp, Chief Technology Officer at DELTA Fiber Netherlands emphasizes the importance of test runs during a migration:
“In particular, performing another test run was typically something that was very much stimulated by Type 2 Solutions to ensure that you achieve the best quality you can deliver.”
Migration in a 24/7 environment
A good example of the latter situation is the data migration at the N.V. Westerscheldetunnel.
A big bang scenario was impossible here because the physical conversion per toll lane took several weeks, but the toll collection process in the other lanes had to continue as usual.
The entire migration did entail more than just the upgrade of the old lanes and the replacement of the existing hard- and software. It also comprised the conversion and migration of all data collected since the official opening of the tunnel in 2003.
The main provision was that the toll collection process had to carry on unimpeded 24 hours a day, and that no transactions could go missing. Further, only one lane would be allowed to be inoperative for upgrade in each direction at any given time.
Finally, the migration plan had to be coordinated in a way that end users would not have to work with both back-office systems in parallel.
The migration was deemed the most precarious operation of the entire project.
Which strategy is right for you?
Based on a preliminary investigation of the data migration with a data migration plan as deliverable, we can provide you with the right answer.
The data migration plan includes:
1. Strategy
The data migration strategy.
2. Action
The action plan.
3. Risks
A risk analysis.
4. Process
The data migration process and the data quality optimization process.
5. Control
The control framework.
6. Planning
The planning, the required resources, and the tools.
Data migration plan
Drawing up a data migration plan takes 10 working days on average. First, there will be a preliminary investigation of the data migration which looks into the source and target system.
Scope of the migration
You always need to start with the target system. Within the target system, you identify the data sets (business objects) that are within the scope of the data migration. An example of a data set is clients or relations, contracts and prices.
Form and content of the migration
The target system determines the form and content of the data migration. To give an example: in the new system, you redefine the ‘business client’ and ‘private client’.
In the present system, a business client is a company. In the new system, you call a business client ‘any entity that purchases business products from us’.
In that case, you have changed the definition, which impacts the way that you migrate the data.
Your benefits
Why would you start with a preliminary investigation with a data migration plan as your deliverable?
The answer is very simple: because you need to know, as does any organisation, what it will cost, who you need to hire, how long it will take, and what the risks are.
For the execution of a data migration plan you need to pick an expert. It should be someone who has extensive experience with data migrations and excellent knowledge of data analysis and database platforms.
Migrating data is a specialism whereby the quality of the data is crucially important. Find the right people and arrange for a (control) framework.
Less is more!
With the proper tools, one expert can accomplish more and achieve better results than five generalists.
Meet our data experts
We are happy to get to know you.
Contact us
Leave your contact details if you wish to receive more information. We will get in touch with you within one business day.
Frequently Asked Questions
When do I start considering migrating data?
After you have decided that it is time to put a new system in place. Start thinking about the data migration before you choose the new system.
Why should I do so prior to choosing the new system?
Data migration factors into your choice for a new system and supplier. Compare it to moving house. If you are unsure as to whether the furniture that you care about is going to fit in the new home easily or only by pulling out all the stops, this will affect your choice for a new home.
Which requirements do I set for the new system and the supplier?
Your requirements for the new system are that you can migrate data into it, that it supports your operational processes and that you can bring the data that is now available to you to the new system or systems. In addition, you will want to know which data migration tasks will be the supplier’s responsibility and which tasks you will be expected to take on during the implementation.
What is the programme manager's role in this?
When a programme manager is set to implement a new system, they will want to know beforehand how much effort it will take to run tests and train users, and how long it will take to migrate the data. Say you want the migration to be completed within one weekend, but the loading of all the data is going to take 6 weeks. It is clear that this is not feasible. In brief, you will need to give this some thought beforehand, so the data migration will form an integral part of the programme and you can make a budget available for this purpose.
Why is data migration important?
Data migration is important because your new solution needs to fuse your ongoing data flow with your historical data. You face the challenge of finding and extracting your old data. The data must be checked for quality and fitness to ensure it conforms to your new setup. Only then can your operations rely on the new solution.