News

9/4/2020

Next-Level Enterprise Architecture Plan: Dynamic tool for the companies of the future

Enterprise architecture is now playing a much more important role in the digital transformation by supporting innovation and serving as a source of valuable input and advice. At the same time, new instruments are needed in order to ensure the continued effectiveness of enterprise architecture approaches. The Next-Level Enterprise Architecture Plan presented by CBA Lab represents such a dynamic, integrated, and collaborative instrument that can be used by both business stakeholders and IT departments. The Next-Level Enterprise Architecture Plan offers a conceptual depiction of the context within which a business operates. In other words, it involves a lot more than just IT. It singles out the pivotal connections between operations, the business organization, IT, and data as the starting point for processes of change.  

 

A plan that can be used on different levels

“The purpose of the Next-Level Enterprise Architecture Plan is to offer all relevant stakeholders – for example the CIO, CEO, and the portfolio manager – a common knowledge, information, and design platform that reflects their respective points of view and can be clearly understood by everyone,” says Joachim Schmider, Coordinator of the Next-Level Enterprise Architecture Plan workstream at CBA Lab and also Head of Enterprise Architecture at the automotive supplier Schaeffler AG. Schmider uses the analogy of an architectural drawing for a residential building to illustrate what he means: The bricklayer looks at the plans in order to find out which bricks need to be laid where, the electrician needs to know the setup for the cable and wiring ducts, and the client looks at the drawing to decide which rooms he or she would like to use for which purposes. “Nevertheless, everyone looks at the same drawing, or at a certain section of the same drawing,” Schmider explains.

Looked at like this, an architectural drawing – or an enterprise architecture plan – no longer seems to be a static depiction of isolated elements. In the case of an enterprise architecture plan, the various business and IT stakeholders who contribute their expertise make the plan collaborative and dynamic. In other words, such an enterprise architecture plan can offer added value in different forms to the various stakeholders involved, while also helping to ensure:

• The effective management of complexity – by making clear the connections between various components of the “conceptual depiction of the company” and therefore revealing modes of action and dependencies as well.
• Harmonization and optimization – by pointing out interdependencies between the organization, processes, IT systems, and data in a manner that enables the development of fact-based arguments and decisions.
• An agile operating model and culture – by promoting cross-functional collaboration and improving communication between the business organization and IT units, thereby making EA a component of virtually every business discussion.
• The acceleration of the digital transformation – by highlighting changes and their effects, which in turn promotes the creation of new digital skills and establishes data-driven and/or data model-driven thinking.
• Cost and value optimization – by increasing transparency and development speed, which reduces process and technical debt and complexity, while at the same time increasing the value contribution achieved.
• Risk minimization – by creating a basis of facts regarding interconnections and modes of action, which helps make it possible to correctly prioritize changes and, for example, consider and incorporate security and compliance aspects from the early stages of conceptualization. With its greater transparency and broader cross-functional utilization, the Next-Level Enterprise Architecture Plan enables the implementation of proactive changes that increase business flexibility.

EA needs to develop additional skills

In order to generate the aforementioned added value, progress has to be made in the area of visualization technology. However, a new cross-functional EA data model also needs to be created that combines and establishes connections between elements from business, data, and IT architectures (strategy and objectives, processes, business objects, IT applications, technologies, etc.). According to the workstream, such an EA data model must display the following attributes:

• Integrity – Connections between all EA data objects from their sources (PPM, CMDB, SAM, ERP…);
• Data centricity – Business data (objects) are a central element of the EA metamodel, and EA models utilize available operational data wherever possible.
• Context sensitivity – The visual structure of the enterprise architecture plan is target-group focused and based on the requirements of the respective target groups.
• Accessibility – Barrier-free access to EA information for all interested stakeholders; ideally with self-service systems. Everyone can use EA knowledge, and also contribute to knowledge growth.
• Dynamism – The enterprise architecture plan reflects changes and interdependencies. It also allows for the simulation of future target states in accordance with various contexts.
• Intelligence – The creation and refinement of the enterprise architecture plan is supported by recommendations, forecasts, and context-based viewpoints that are made possible by new IT capabilities (AI, graphs, visualizations, etc.).

In order to implement the Next-Level Enterprise Architecture Plan, enterprise architecture as a whole, but also artifact architecture plans, must develop new skills. According to the workstream, EA should be used not only in operational IT applications but also as a strategic and cross-functional planning and management tool. To this end, next-level enterprise architecture plans need to include expanded tools that can process EA data from a persona point of view, integrate operational data into EA models, allow for simulations, and use AI to expand EA models and make them more simplified and automated.

Schmider acknowledges that it will definitely take some time until companies implement and actively utilize the ideas and concepts described in the Next-Level Enterprise Architecture Plan. The process has already begun, however and “we’ve taken a major step forward with our precise description of the objective we’re trying to achieve, the strategic added value it will result in, and the roadmap that will help us get to where we want to go.“

What companies can do to reach the next EA level

The Next-Level Enterprise Architecture Plan is not going to be implemented any time soon – but companies can already take specific steps today to ensure all requirements are met tomorrow. In the areas of tool and mindset in particular, steps can be taken to increase the degree of sophistication of enterprise architecture – and the entire organization. The goal in both areas should be to improve a company’s ability to deal with complexity, according to Schmider.

Many EAM tools are primarily designed to create transparency with regard to existing structures and systems. These tools are used mainly to collect data on the IT landscape, and they either do not take into account, or insufficiently take into account, data and data models from other areas of a company, which means they cannot reveal important connections and relationships. If the latter are not identified, systems usually cannot be changed successfully. That’s why companies that are preparing to implement changes should use EAM as a data-driven discipline and innovation-support mechanism. Schmider believes it is crucial that the introduction and utilization of EA tools be combined with the gradual further development of the EA operating model, as this will automatically lead to a fundamental change in thinking and an increase in complexity. EA tools can more effectively support this process if they can identify complex connections and relationships and present them in a simple and intuitive manner. Good EA tools are already capable today of combining and intuitively consuming several data models. Here, the innovation trend is clearly moving toward the use of graph-based semantic EA models that can be expanded as needed and converted into traditional data models and views. Such models show users results as knowledge graphs, for example, which makes it much easier to identify connections and relationships.

Tools aren’t perfect yet, but they’re good enough to start out with

Schmider believes that today’s state-of-the-art tools, some of which already use innovative analytics systems and artificial intelligence, have not yet developed into the perfect tools that will be needed as we move toward the Next-Level Enterprise Architecture Plan. “Still, there’s no reason to wait for the perfect tool,” he says. “The tools we have are good enough to start out with, especially when you consider the level of EA maturity at most companies today.” This doesn’t mean that CBA Lab is content with the status quo, however. A new working group has been formed whose tasks are to identify the shortcomings of the EAM tools that are now being used and to motivate providers to specifically address these deficiencies in their innovation and development roadmaps.

Schmider says EAM tools need to have a type of multidimensionality that on the one hand allows users to analyze a specific objective on the basis of the capabilities, processes, IT systems, and data needed to achieve it, and on the other hand enables users to work with everyone involved to develop a change scenario that takes the multiple dimensions into account.

According to Schmider, tool self-service capabilities are also important as we move toward the Next-Level Enterprise Architecture Plan. If EAM tools are to be used to improve modular building blocks and organizational units, then they can no longer be used exclusively by architects. Other IT and business stakeholders also need to use them and be able to understand the results they produce.

When selecting tools, companies should therefore make sure the tool in question can perform three tasks:

• Combine various data models and present connections and relationships on different levels in a clear and understandable manner.
• Collect sufficient amounts of various types of data in order to ensure the existence of a sufficiently broad database that can be used to support decisions relating to changes.
• Self-service capacities: The tools need to be so easy to use and so simple in terms of their presentation of results that even non-experts can use them. The greater the number of employees who can use the information, the more effective will be the entire organization.

Recognizing complexity and working with experts

In terms of mindset, the EA discipline seeks to recognize complexity and internalize the realization that architects don’t necessarily have more detailed knowledge than experts in departments and units, and must therefore work with the latter and incorporate their expertise. The entire company needs to learn that EAM is not a pure IT discipline. Instead, EAM structures information and complexity from various areas and can therefore help with the establishment of a basis for decision making during change processes. To this end, it is also important to transform EAM from a push discipline into a pull discipline that can be used by all departments and units planning to implement changes.

This cannot be done from one day to the next. Schmider recommends helping top management recognize the value of EAM by demonstrating how EAM creates value in the short term: “Then everyone will have enough time, and also enough confidence, to pursue long-term objectives such as the Next-Level Enterprise Architecture Plan.” As an example, Schmider cites the development of an integrated enterprise architecture for a digitally expanded business model, which Schaeffler is pursuing with a system of warehouses equipped with sensors. Another example involves flexible and modular solutions for integrating business acquisitions or, conversely, selling off divisions or business units. “Here, EAM can demonstrate its value in a short-term situation and will then be given more time to achieve a long-term objective like the Next-Level Enterprise Architecture Plan,” Schmider explains.

Detailed work results are available exclusively to CBA Lab members.

This article appeared in a similar form in “Computerwoche” (23-24/20).

Zurück