Standard Data Models in CCM

How Interchangeable Parts Can Improve Your Capabilities  

Many of us learned in school about Eli Whitney, Marc Isambard Brunel, Henry Leland and others who pioneered the idea of “interchangeable parts”, a concept which accelerated the industrial revolution. Over 200 years ago, standard, identical parts drove explosive growth in manufacturing as assembly line production replaced hand-crafting for many goods. In recent decades the ability to customize within any given set of standards, through modular design & production techniques, has shown that the same principles of interchangeability still apply.  

By building parts that are standardized & effectively identical, each part can freely replace another of the same type. This allows for easy assembly and repair of a system, and minimizes the time and skill required. Another result is that the “division of labor” is more easily accomplished – people can focus on their core responsibilities while being assured that their work will integrate with everyone else’s without having to manage a large set of customized requirements. The use of interchangeable parts unlocks potential that is otherwise lost to managing unnecessary complexity.  

These same concepts can be applied to your CCM framework. Some of the more obvious examples are the standards that govern images (EPS, TIFF, JPG, PNG, etc.) and documents (PDF, AFP, HTML5, etc.) where organizations often choose a limited set of input or output standards to be used. These are important but simple standards to establish, and most organizations have adopted standards in this area.  

However, a common gap we often experience in our CCM solution development at BelWo is that organizations miss the opportunity to establish a set of “standard data models”, or standard specifications for the technical layout of certain data and templates, to enable broad, modular reuse of components across their CCM framework. Whether it’s because of poor governance or poor architecture design or just sheer scale, it is easy for organizations to lose control of the complexity of their CCM applications. Standard data models, applied appropriately, can help solve that problem.  

Let’s explore some of the concepts of standard data models and how they can be used effectively in CCM.  

What is a standard data model in CCM?
In a CCM solution, a standard data model should include a database model and a template model:
  • Database models are either relational or flat models that show the field definitions, parent/child arrays, etc. for all data sources in the framework, as well as their relationships to each other. These are typically visualized in a hierarchical or tree format, and typically implemented as XSD (XML schema definition) files that define certain interchange points in an application.
  • Template models show how data elements from the database model are represented in the variable composition from your CCM system. These are blocks of layout logic that might range from simple & common uses like address blocks and key lines, to more complex business blocks like transactional tables, targeted messaging areas and content insertion logic.

In combination, these models define how data should be “shaped” when it comes in or out of the application, and how that data is represented when it is composed and printed or emailed or sent to any other channel.  

These become our interchangeable parts. A standard data model in CCM ultimately results in an inventory of interchangeable parts that, within the appropriate context, can be re-used across various applications.

When should they be used (and when should they not be used)?  
Let’s start with some discussion of how standard models can be misapplied. While they are critical part of a successful CCM framework, standard data models aren’t a cure for all your CCM solution woes. Improperly applied, they can slow you down.  

For instance, a print service provider wouldn’t build a single, monolithic standard data model to serve all their customers. Trying to build a model that would support healthcare, insurance, financial, and retail customers would be impossible.  

Even an enterprise with a narrow business focus might not build a single, monolithic standard to drive their customer communications. When there are significant data structure or template layout approaches between communication streams, they often warrant different models. And in almost every enterprise there will be some small amount of customer communication output that just doesn’t happen consistently enough to warrant the development of a model.  

Over-standardizing can create confusion, lengthen development times, and decrease flexibility – exactly the opposite of what standard models are meant to achieve. Choosing the boundaries for a standard model is not a trivial task and requires an understanding of the bottom-up realities (existing data structures and how they operate today), the top-down goals (what the business wants to accomplish), and the capabilities of your CCM framework.  

So no, the service provider wouldn’t build a single standard model for their entire business. But they would certainly develop a model for address blocks and key lines. And they would probably develop a standard model for healthcare ID cards, for instance. For auto policy quotes. For daily confirms. For quickly deploying emergency letters. In this way, they can address those products quickly, using well-established data and layout standards. They might create a layout variation from the standard for a specific customer, but the existence of the standard cut the development time enormously, and the layout variation is linked to the standard for easier maintenance and management. In addition, the standard model components are well-tested and robust, lowering testing effort and the risk of untested conditions.  

Our enterprise customers use standard models to drive lines of business or product lines, and to help govern standards across their organization. A property & casualty customer might have their own auto policy quote model with layout variations for different regional markets, along with similar models and regional variations for a range of personal and commercial P&C products. When standard models are in place, onboarding a new customer or even a brand-new product becomes the assembly of some standard components, some (much lower than normal) customization work, and a faster, lower-risk deployment. The result is a communication that the enterprise knows will be accurate and consistent in content and style.  

What are the impacts of using standard data models?  
As we’ve noted, the biggest impact of adopting standard models is an increase in speed-to-market, communications accuracy, and improved governance of standards (data and layout). Onboarding new customers is faster and developing new communication variations or completely new communication products becomes more efficient. As with the interchangeable parts that helped launch an industrial revolution, standard models in your CCM framework can help launch a revolution in capabilities.  

Will every new document be assembled solely from pre-configured standard blocks and use only the standard data model within the application? No, there will be some document streams that require variations from the standard. These are easily addressed by “extensibility”, the ability to carry non-standard data structures and template layouts through the application while not abandoning the core standards. A customer has a special transaction array that’s not part of the standard? No problem, within that customer’s application we can join that array to the standard data model and use it for composition. A customer has a very special transactional table that’s not at all like our standard? No problem, we can build that custom table layout and use it as an exception in our customer’s templates, effectively creating a customer-level standard that is a variation from the core standard. And who knows, perhaps that very special transactional table will become a new core standard over time.  

This extensible approach to standard + custom development means that we get the huge benefit of standard models while not losing our flexibility for customized products.


Enter Captcha Here : *

Reload Image