The best way to build your data integration architecture
By Henning Lund - March 16, 2020
Most businesses use multiple systems for different purposes. Data integration makes these systems work together. With as many as 20 integrations needed between systems, the process of creating data integration architecture can be complex. How can this be done? The obvious answer might be to integrate all components with each other. This is a tempting approach, but it tends to prove unworkable over time.
That's why we've been working on creating a generic reference framework that offers a high-performance data integration set-up, at a cost that is affordable SMEs.
In this blog, we explore how we did this, the logic behind our approach and the advice on best practice we've gathered along the way. It also explains why we integrate systems such as Salesforce Marketing Cloud into a CRM rather than directly into the ERP system.
What is integration architecture?
Integration architecture is the integration of multiple IT elements and components, facilitated by software architecture. This architecture is changed or altered by any new digital development models and patterns.
System integration architecture brings together large and diverse operational platforms and applications, and ties together the mix of platforms and operating systems.
Now let's go on to exploring the difference between primary and secondary system integration architectures.
Primary systems vs secondary systems
When beginning an integration project, it's important to determine which systems are secondary and which are primary. Primary systems operate core business processes. Secondary systems feed information to the primary systems.
We can set up data integration architecture that links several different primary systems, such as a CRM and an accounting system. This streamlines data recording and use, avoiding double data entries, automating processes and ensuring transparency.
We can then also set up integrations between various secondary systems. This means that data channels are always consistent and easily maintained. The alternative is spidernet architecture, which means multiple data channels, many of which are redundant.
examples of system integration architectures
One of the most common integrations we set up is between Salesforce and Microsoft Dynamics ERP (AX, GP, NAV, 365). We're going to look in this example at an integration of the three primary systems:
Salesforce Marketing Cloud (Pardot)
Salesforce Sales Cloud (CRM)
Microsoft Dynamics ERP
Step 1: we integrate Salesforce Marketing Cloud with Salesforce Sales Cloud
This integration feeds qualified leads from Salesforce Marketing Cloud into the Sales Cloud.This is a simple integration, with data flowing just one direction. It is included with Salesforce as a native integration (the same is true of many other similar products).
Why aren't the leads fed directly into the ERP? Because they need to be developed - they're not yet customers. It might make sense to do so if the ERP you use includes a CRM, but not otherwise.
Step 2: We integrate Salesforce Sales Cloud with Microsoft Dynamics ERP
This is slightly more complex, as it is a two-way integration. Both systems hold details on customers, so they need to be synced.
We need to set up secondary integrations into the primary systems. Which these are depends on the business. In general, secondary integrations into a CRM system are sources of data that provide information about contacts, communicate with the target market or measure the effects of actions. Integrating these into the CRM simply enriches the data.
Secondary integrations into an ERP are generally about specialist systems that support particular business integrations. They include product information management systems, such as product configurations and third-party logistic partners. Integrating them means you won't usually need an EDI (Electronic data exchange).
As well as a CRM solution and marketing system, you might be using a service system (such as Salesforce Service Cloud).
This could be fully integrated with your ERP - but it often won't be worthwhile to do this. It might be that a partial integration is more useful. ERPs usually include service functionality, so it doesn't make sense to also integrate the same functionality from another system. It does often make sense to integrate some limited data, such as timesheets or invoice information. What is most appropriate for you depends on your business.
If you feel like you are ready to find out about CRM integration architecture and strategy, then please get in touch and a member of our team will be happy to help and advise you. We can help you work out of the best approach for your business.
With over 25 years’ experience in strategically propelling businesses forward, Henning is considered a business development entrepreneur with a passion for transforming businesses, sales and marketing operations through out-of-the-box thinking, concepts building and process automation to improve overall performance and scalability.