Microsoft Dynamics CRM 4 : Scribe Integration - Integration Options

Generally speaking, companies implement three distinct levels of integration:

• Data replication: Moving data “one way,” into or out of CRM

• Data synchronization: Moving data “two ways,” into and out of CRM

• Process integration: Facilitating business processes that rely on consistent data across a broad range of business applications

Each category has its own characteristics, implementation requirements, benefits, risks, and costs.

With regard to costs (we cover the benefits, risks, and requirements in each section), there are varying levels, both during the initial project and on an ongoing basis. As a general rule, the complexity and resulting costs associated with an integration project will be exponentially greater for two-way or multidirectional integration projects versus one-way integrations. Complexity also rises dramatically as more and more dependencies between applications (and their respective user bases) are expanded. Costs tend to break down into two major categories: the technical implementation and support costs, and the organizational disruption costs (with the latter in many cases greatly underestimated).

Figure 1 illustrates the relationship between value versus cost/disruption for companies across the three levels of integration.

The following list summarizes the costs you should consider when embarking on a CRM integration project:

• Technical implementation and support

  • Integration requirements analysis and design

  • Integration software (including maintenance, support, training, and upgrades), whether purchased from a vendor or developed internally

  • Application “adapters,” particularly when applications are targets in the integration (another reason why one-way integrations can be simpler and less costly)

  • The costs of ensuring data integrity across systems (especially costly with synchronization and process integration because of the need to maintain updatable data in multiple systems)

• Organization disruption costs

  • User training and orientation around the use of the newly integrated systems (sometimes referred to as reengineering).

  • Risks and costs of creating dependence between mission-critical transaction systems and CRM

  • The political barriers of data ownership and coordinating activities across functional areas of the business (one of the most common being CFOs who don’t want salespeople messing with their back-office data)

The following table maps the three levels of CRM integration against the preceding criteria to represent the relative cost of each option.

^[*] Assumes the purchase of a full-function CRM integration platform.

^[**] Although two-way, synchronization involves much simpler application touch points.

It becomes apparent that as we move from one level to the next, the cost and disruption of each grow exponentially.

So now that we understand the relative costs of the three different types of integration, how do we make sense of the relative benefit of each? To answer this question, it is important to understand how your company markets and sells its products. To illustrate how a company’s sales and marketing process impacts the value side of the equation, let’s define two types of companies at opposite ends of the spectrum. We refer to one category of company as Type 1 and the other as Type 2:

• Type 1: These companies have intensive, relationship-focused sales processes. They generally need to educate their buyers about their products, have longer sales cycles, and have direct and indirect sales teams skilled in the “art” of relationship, value-based selling. Industries that tend to fall in this category include financial services, professional services, health care, capital goods manufacturing, and much of high technology. These types of companies generally benefit the most when you provide their knowledge workers with information that enables them to target customers better and to more effectively manage customer relationships. Given that the transaction side of the customer relationship is generally an occasional event in the sales process that does not dominate the lion’s share of the sales team’s efforts, they tend to gain diminishing value from additional levels of integration. This is particularly true for process integration.

• Type 2: In these companies, the majority of the sales process is centered on transactions. Their customers generally require less information about the features and benefits of products and are more concerned about things such as quantity on hand, price, and availability. Industries that fall in this category include consumer goods, distribution, process manufacturing, and commoditized high technology. These types of companies generally benefit the most from integrated, coordinated, and efficient management of customer transactions. Type 2 companies still benefit from data replication and data synchronization, but ultimately realize the greatest strategic advantage through process integration.

After you’ve decided on the appropriate level of value versus cost, integration can be expanded to include additional data replication, data synchronization, or for Type 2 companies, expanded to include process integration.

With the right long-term plan implemented on a technology platform that can expand with your changing business, your CRM investment will be well positioned to deliver meaningful and sustainable competitive advantage.

Figure 2 summarizes the key capabilities required to support the integration requirements previously outlined.

As shown in this figure, you need five major capabilities to perform these data aggregations.

• Data extraction: You must have direct access to your source applications via a database or a proprietary application programming interface (API) from enterprise applications such as Microsoft Dynamics GP, JD Edwards, SAP, MAS 90/200/500, and Siebel. You must also be able to capture net changes either through source queries or via published messages from the source where available.

• Data translation: The semantics and format of many fields of your source data will likely differ from those in Microsoft Dynamics CRM. Important capabilities include parsing and concatenating text fields, performing date and numeric calculations, executing conditional logic, and performing lookups to resolve synonym values. You also need to maintain a cross reference of primary key values, to apply updates from one record in your source to the corresponding record in Dynamics CRM.

• Data update: This capability is the most crucial, yet complex area of your integration task. Capabilities you should look for include the following:

  • Avoiding duplicates using fuzzy logic (like comparing elements of the company name and ZIP code to look for an account match) for record lookup

  • Performing insert and update operations against multiple objects within Microsoft Dynamics CRM when processing a single source record

  • Performing all target processing against the Microsoft Dynamics CRM integration API to ensure that all data imported has been validated by Microsoft Dynamics CRM’s application rules

• Automation: This is where using a customizable template model that incorporates a one-step process from source to target proves very useful. After you have designed your business process, you need to implement an automated event-detection mechanism to initiate an update to Microsoft Dynamics CRM. Look for a solution that supports both batch and message-based processing; each approach is appropriate in different integration scenarios.

• Monitoring and management: After you have developed and implemented your data aggregation solution, you need to consider the ongoing management of the solution. Look for a technology that

  • Enables you to remotely support the solution (including start and stop processes, diagnose errors, and so on) via your web browser

  • Automatically alerts an administrator via email when processes fail, error, or produce abnormal data conditions

  • Can scale across multiple processors to support high-volume data scenarios

Traditional approaches to data aggregation for CRM cover a broad spectrum from custom development to the use of sophisticated technologies such as Microsoft’s BizTalk Server. Unfortunately, these choices represent approaches that are either way too little or way too much. In the case of custom development, someone has to code all the functionality outlined previously to deliver a workable solution. More often than not, these custom solutions are lacking in functionality, unreliable, or difficult to manage. In addition, they are inflexible to changes in your business.

BizTalk Server, on the other hand, may include some of the functionality required but is designed as more of an infrastructure backbone to support a wide range of integration scenarios. This poses two challenges. First, BizTalk Server lacks quite a bit of the specific CRM-focused functionality needed, forcing you to fill in the blanks with custom coding (with all the challenges of custom coding mentioned earlier). Second, it tends to be very complex to install, configure, and manage, and generally requires significant additional hardware and software infrastructure investments. You can quickly lose track of the fact that you just want to get customer data to the sales team.

Data Replication

Data replication is by far the simplest and least interdependent type of integration. With replication, a copy of certain customer data that resides in one system is added to the customer records in the CRM application, with data moving in only one direction. Typically, the replicated data is “view-only” in CRM; that is, it cannot be modified by the user but provides more complete customer data to increase the effectiveness of CRM.

Figure 3 illustrates data replication.

The following table outlines common replication scenarios and the benefits of each.

Bear in mind that implementing a replication scenario may involve extending the data model of a packaged CRM application if the key data elements you want to share do not exist in the base configuration of CRM.

By providing this additional information about customers within the CRM system, sales users can improve the quality of customer interactions and more effectively target customer opportunities.

Replication has another benefit that is not so obvious: It dramatically improves the adoption of CRM by users. CRM is one of those odd business applications where adoption by its “users” is difficult to mandate in most cases. It is rare to find a case where sales reps who were 200% of quota lost their job because they didn’t put their sales activities in a CRM system. So how do you get these individuals to adopt your CRM system? You provide them information in CRM that will help them sell more—information they couldn’t otherwise get. In many cases, this type of quid pro quo has formed the basis for successful CRM implementations.

Data Synchronization

The objective of synchronization is to maintain the same set of customer information in multiple systems, reflecting changes made in one system across the others. Synchronization typically focuses on the more basic demographic customer information that is common to multiple systems, such as company contacts, addresses, phone numbers, and so forth. Figure 4 depicts a typical customer synchronization scenario between a company’s ERP and CRM system.

As you can see in Figure 16.4, only certain subsets of data within each system are being synchronized. Given the varying structures of these systems, it is not uncommon to see multiple integration “touch points” between systems to support a synchronization scenario.

Because changes made in one customer database are reflected across all customer databases, data entry effort is dramatically reduced, errors are eliminated, and your entire organization is working from the same information.

Process Integration

With process integration, data is shared from one system to the next based on each system’s role in an integrated customer process. The most commonly discussed customer process related to CRM systems is the “quote-to-order” process.

The following table outlines the integration steps required to support quote-to-order activities between a CRM system and a back-office system.

One thing to consider in process integration is that the order of the steps is extremely important. In the example shown here, if the quote is created from an invalid or out-of-date item from the product catalogue, the ERP system will not be able to support later steps of providing product availability dates or processing the order. In addition, in most cases, data replication and data synchronization are prerequisite integration requirements for implementing process integration.

Typically, process integration relates to those activities in the sales cycle that involve an event or transaction, such as a sales quote, an order, an invoice, a credit verification, a contract renewal, a product return, and so on. By coordinating these activities more efficiently across the users and systems involved in these processes, companies can accelerate revenue and cash flow, eliminate redundant effort, and provide a better experience to their customers.