Data conversion, or data migration, is the task of bringing over relevant data from your legacy system to your new CRM. This is one of the most crucial steps for a successful AMS implementation. Bad data will result in a failed project, no matter how good the features are within your new platform.
As you continue your AMS implementation, here are a few considerations regarding the overall data conversion process.
Timing
The data conversion process needs to begin at the start of the implementation phase and run throughout the entire process. This is because there are a lot of items to be thought through and planned. A few of the most important are detailed below:
Data Quality
A new AMS is a good opportunity to both get a handle on your data quality and take concrete steps to clean your data. Consider launching a data governance effort to help you – the collection of processes, roles, policies, and metrics will ensure the effective and efficient use of all information.
Clean in the legacy, not the new
It is strongly recommended that you clean your data in the legacy system and not rely on bringing bad data into the new AMS and cleaning there. We can assure you that once you move to the new system, you will be risking its stability and performance.
Additionally, the first 6 months post-go-live will be spent learning the new AMS. Some customers are sold on the fact that the new AMS will help them with cleansing historic data. Our experience has been that this has not always worked well.
How Much?
One of the most important questions that we have to answer is “how much data needs to be migrated?” Here are a few considerations:
- How clean is your data? Consider bringing in only clean data. Often, the farther back in time you go, the worse the data tends to be. Trying to bring this data over will only result in time spent cleaning and transforming for a very low return.
- Are there regulatory requirements? Do you have regulatory requirements around the number of years of data that needs to be maintained in the system? If so, make sure you are aware throughout the process.
- Do you have a data warehouse? If you have a data warehouse, the data you migrate can consist of only the bare essentials. You can access the rest of the data from the warehouse.
- Evaluate Value, Not Time! Clients decide how much data to bring over by using an arbitrary number of years that they are comfortable with – let’s say 5 years. This is not the best method. Instead, think of it this way: I will bring in all my years of certification and membership data since it is required by my members and provides value, while I might not bring in more than one year’s worth of merchandise (hats, cups, etc.) data.
- Do you own the data? If the answer is yes, and you have access, then consider leaving most of the legacy data in the database and tying a reporting/BI tool to it. This will give you easy access to your past data while not burdening the new AMS.
- What is the Cost? There is typically a data storage cost with newer cloud-based systems. Evaluate the ROI of the data you choose to migrate.
Extract, Transform, Load
Get help if required. Most system integrators (SI) will not want to take on the responsibility of extracting data from your legacy system. This could be because they do not have experience with your legacy system or do not want to take on the responsibility of extraction. If you do not have internal staff who can pull data effectively, hire a vendor that is an expert in data extraction. Extracting data might also include transforming data into a format that the SI requires. This requires certain skills that less experienced teams may not possess.
Most systems integrators will help with the loading of the data if it is provided to them in the format they desire. However, we have also had some clients take on data conversion themselves – and succeed at it! The key to remember here is the competency of the people executing the conversion process as a whole.
How many data cycles?
We recommend a minimum of at least 2 data cycles. The first data cycle is used to get the methodology established and the second is the final data cycle. Taking the time to iron out the problems in the first cycle will be the key to having a quick and clean final data cycle.
Reconciliation
During the running of your first data cycle, confirm you have the requisite auditing capabilities in place to ensure that you are comfortable the new system represents your information accurately. Since the structure of your legacy system could be vastly different from the new AMS, finding good auditing methods during your first data cycle is important.
One simple tactic is to rely on a set of similar reports in both systems. Another is to write automated check routines that compare data. Again, a topic for its own blog.
Stay Tuned For More
Implementation is well underway, and there is finally an end in sight! It is important to continue to be detail-oriented and focused as your team gets into the later stages of the project – don’t let all your hard work go to waste because you run out of steam.
Step 9 of Your AMS Journey will examine the User Acceptance Testing (UAT) and training phases of your implementation process. In the meantime, remember that fusionSpan is here for all of your CRM questions and needs!