When it comes to making CRM projects much easier to implement, we certainly suggest that you consider bringing together data in Microsoft’s common data service, Dataverse that will allow this data to be readily available across Dynamics 365, Power Apps, Teams as well as any other Microsoft application.
However, you will find that there are many different occasions where businesses need their CRM system to connect up with different external data sources like ERP systems for example. In these cases, being able to seamless integrate is vital in building a single customer view across all the various touchpoints and joining up business processes that will help to provide your business with a competitive advantage in the market.
It goes without saying but when it comes to integrations these will make or break a CRM project, therefore this is why in our latest blog article, we will take you through 5 game changing tips & tricks on how to complete CRM integrations successfully.
1) What is the difference between an integration and a migration?
Before we get stuck into things, let’s make it clear what the main differences are when it comes an integration and a migration.
Firstly, a migration is where data is moved from one system to another. You may find that a migration is tested multiple times throughout a project, however the process is only completed once by extracting, transforming and adding data into the new system. There are a number of ways in which a migration can be handled, you could use spreadsheets or a much more straightforward solution that will address any problems with the data before moving the data across to the new system.
On the other hand, an integration essentially connects your CRM system with a third-party application which will help to maximise the benefits for your business. Unlike migrations, integrations involve a constant dialogue by transferring data back and forth between these systems which removes any duplication and enables routine workflows to be better handled.
Better still, you could use an integration for the purposes of a migration where data will flow into a CRM and then be refreshed at set points throughout the year. However, the only problem with this is the fact that data issues need more specialist migration tools to review and resolve these issues.
2) The different types of integrations
You will find the way in which two systems are joined together largely depends on a business’s requirements and what needs to be delivered. More precisely, why is this data needed in the main system?
In many instances, you will need to engage with data to be able to adjust this to bring it more in line with a business process. Often in these situations, Advantage would work with your business to produce an integration that will seamlessly write data into Dataverse so that it can be adjusted within Dynamics 365.
In the cases where users just need to be able to look at data, the integration solution is completely different.
In these scenarios, here at Advantage, we are able to implement an integration that will only allow data in the main system to be viewed on a read-only basis.
3) How do business processes compare in both systems?
Before work can start on an integration, there needs to be a thorough understanding of what exactly needs to be delivered by all parties. This is where Advantage can work with your business to outline each process so we can determine what needs to happen to be able to support it.
From a sales point of view, a lead is added and then converted to an account, contact and opportunity. After this, a quotation is given along with a price which involves certain products and units. Once the business has been won, an order with lines, products, units of measure is created to filter this information through to an ERP system to be delivered. However, the process doesn’t end here, as the order needs to be delivered and therefore an invoice needs to be added to CRM so that the sales team can see if the customer has processed the payment.
This just shows how complex an integration can be, furthermore, it's important to understand that complexity isn’t related to the number of tables, fields or records that are used. The complexity comes from combining a number of data models of two or more, systems to make sure that the data is accurately portrayed in a quick manner to help ensure that the business process is fully supported.
4) How to understand data & assess data quality
It doesn’t come as a surprise, but poor data is a common problem that should be resolved before any form of integration happens. However, that said, data management will always be the customer’s responsibility as no one will know the data as well as they will.
During the data review & checks part of the process, Advantage will review the following things:
Unique data keys – a unique property will need to be determined for each record. The best way of doing this is through a globally unique identifier (GUID), however any ‘primary key’ will work which can be formulated using a number of compound keys. Therefore, every time a record needs to be edited, the unique identifier will then be used to the define the target.
Duplication in the records – You will notice that these are often in the multiple versions of the ‘same’ contact or account. However, if both entries have different unique keys, which one is right? Again, this is another example of where the ownership of this task is down to the business itself.
Have all the mandatory fields been assigned a value? - Do you have any NULL fields in one system that will be required in a business process in another system? If you do, how will these be dealt with?
5) How often will the data need to be refreshed?
This is another significant factor in establishing how difficult the integration itself will be. If you require the data to be easily accessible in the other system in real-time, a form of orchestration will need to be created in order to allow the data to be pushed into the target system. If real-time integrations are required this poses more questions and with it comes a greater range of complexity particularly around error handling, retry logic as well as what is shown to the user.
On the whole, most integrated business processes can experience some form of delay in the system. To give an example, if a salesperson needs to add an order in CRM, does it need to be in the ERP instantly? To simplify this complexity within this process, will that second really have a detrimental impact on the whole operation? If it doesn’t, what will need to be established is what an acceptable time for this entry into another system will be? Just one minute or an hour?
How often an integration will be run largely depends on the business processes tolerance. To put this into content in the form of an example, new order integration processes may have been created to run every 15 mins, however the sync processes to add new products might only need to happen once a day.
The tolerance for this process will showcase which integration model will be used:
- A push model – moves data from source to target and gets a notification before progressing onto the transaction itself.
- A pull model – pulls the source system at certain times and relays the records to the target.
Next Steps?
If you are business that is looking to complete more integrations for your Dynamics 365 solution, looking to complete more data migrations or you have a CRM integration that is broken, please do contact our team of Dynamics 365 experts today to discuss your specific issues.
Articles like the above can be delivered straight into your inbox. How? Sign up to our mailing list.
