Enterprise Architecture Strategy

Introduction

Enterprise architecture (EA) is a discipline for proactively and holistically leading enterprise responses to disruptive forces by identifying and analyzing the execution of change toward desired business vision and outcomes.

To present and deliver ongoing value to the business and IT leaders, enterprise architects require nextgen enterprise architecture strategy and best practices while avoiding the most common challenges and distractions.

📚 2023 Gartner® Magic Quadrant™ for Enterprise Architecture Tools

Sell the value of nextgen EA to stakeholders

The value of EA might not be immediately obvious to a large part of your organization, sometimes not even your CIO. You need to be able to demonstrate its value – some even say that you should sell first and architect later.

CIOs usually care about three things: saving money, increasing innovation levels, and decreasing risk. EA can help them achieve all three, you just need to show and sell its value when answering the most common questions of CIOs.

Figure 1: Enterprise architecture value

EA strategies

Reduce costs

According to Gartner, companies with a strong EA foundation have 25% lower IT costs. This is no surprise because EA can help to uncover previously untapped sources of savings, some of which we have listed below.

We are focusing on a few quick-win activities here that will help you to realize savings as fast as possible:

1. Rationalize your applications

Phase-out obsolete applications and consolidate applications. It is possible to reduce costs by 15% by eliminating redundant applications without affecting quality and business value. This will not only save you license costs but also decrease costs for support and maintenance.

Analyze your applications by Business Capability and User Groups. Use your Application Matrix with Business Capabilities and User Groups to look at your applications and configure it to view the functional fit.

You will be able to identify support gaps and redundancies in your application landscape. It may make sense to agree on global standards across your locations and phase out non-standard applications.

By rating the technical and functional fit of your applications you will be able to identify applications that are not up to the job and do not meet technical requirements.

Focus on divesting those while also keeping an eye on their business criticality to make sure you do not switch off applications critical to the business without a successor in place.

According to a case study by McKinsey, in an assessment of their application portfolio, one retail bank was able to find more than 50 unused applications to decommission, 150 redundant applications to consolidate, 800 point-to-point interfaces to put on an integration platform, and 400 applications to connect with a data integration platform.

Application rationalization can be the basis for other cost savings activities, such as software license optimization, application retirement, or server consolidation (see Figure 2) with various levels of complexity and duration.

Figure 2: IT Saving Potentials from Application Rationalization (Deloitte)

2. Consolidate your vendors

A global study of the Everest Research Institute with a focus on a company’s external IT services found that fewer suppliers bring down the total cost of ownership (TCO) of applications annually between 22 and 28 %. While Gartner found out that it is possible to save up to 30% on software licensing costs alone, by optimizing them.

Enterprise architecture can provide important input on user groups and user numbers for IT use across the organization. Use application inventory to analyze who your vendors are and how much you are spending with them and pass this information on to your procurement (see Figure 3).

You will easily be able to identify opportunities where vendors can be consolidated or where procurement can negotiate better rates, as you can see how much you spend with them in total.

Also, as an enterprise architect, you are ideally positioned to be able to analyze the impact of any changes to the vendor landscape. Provide consolidation scenarios that address functionality issues from swapping IT components and reconciling any interdependencies.

Work closely with your procurement department to ensure an effective vendor selection going forward and do not forget about the changes in support and training that come with a change in vendor and product.

Figure 3: LeanIX report showing provider costs

3. Save time on documentation

It is harder to put a figure on how much is saved by saving time otherwise spent on searching for IT-related information.

But studies estimate that by using an EA tool a significant reduction is possible and even more time is saved on documentation and preparation of reports by using state-of-the-art reports that are always up-to-date.

Another benefit of having an EA tool in place is that everyone can be given access to relevant IT information, so time is saved in distributing the information. Staff can also collaborate straight in the tool, directly in the relevant context.

According to McKinsey, organizations can raise the productivity of workers by 20-25 % using social technologies. A searchable record of knowledge can reduce by as much as 35 % of the time staff spends searching for information.

You will be able to start projects faster because all the relevant information is already in place, which means significant savings in staff and consulting time and costs. Considering that a single IT consultant will cost a minimum of €20.000 per month, this adds up very quickly.

And do not underestimate the benefits of documentation – losing knowledge can be a costly business. Employees can leave your company taking their knowledge of systems with them.

If you have no proper documentation, costs can quickly spiral out of control. €11 billion are estimated to be lost every year due to knowledge drain.

Reduce risks

1. Avoid IT security incidents

An up-to-date EA inventory gives you information on all your applications including the technologies they are based on. This helps you assess which applications might be at risk because underlying IT components are no longer supported and lets you keep track of your technology standards. Incidents that happen because of unsupported technology components on average will cost companies around €600.000.

"A shocking example of how a runaway IT risk incident can have a catastrophic impact is what happened to the airline Comair, a subsidiary of Delta Air Lines. One busy December, Comair’s crew-scheduling system failed because it was only capable of handling a certain number of changes a month.

The system abruptly stopped functioning, leaving nearly 200,000 passengers stranded throughout the US in the run-up to Christmas. Revenue losses as a direct result of this incident are estimated at about US$20 million."

An EA solution offers the required support for determining the responsibilities and roles relevant to IT security. Clear responsibilities for Applications, Processes, and IT Components can be set up and maintained. Quality mechanisms ensure that the data stays current and will be available in the right context in case of a security audit or incident.

Your EA tool can also help you to classify the criticality of the data objects used by your company’s applications. With the help of EA, you can base your technology decisions on reliable, up-to-date facts.

2. Avoid IT outages due to obsolete technologies

Most companies are much better at introducing new technologies than at retiring them. Even government agencies are not immune to this.

"US Government auditors blasted the Internal Revenue Service (IRS) in 2015 for missing deadlines to upgrade Windows XP PCs and data center servers running Windows Server 2003, both of which have been retired by Microsoft.

Nine months after Windows XP fell off Microsoft's support list, the agency still could not account for 1,300 PCs, about 1% of its total, and so could not say whether they had been purged of the ancient OS. The IRS also had to pay Microsoft for post-retirement support contracts to be provided with critical security updates."

Thanks to your EA Inventory it is easy to keep on top of IT components that are no longer supported and thus signify a risk. In LeanIX you can report on those applications that are at risk because the underlying IT components are out of their lifecycles.

View of the technology risk of an application landscape:

Figure 4: Business impact of technology obsolescence.

3. Do not get caught out by compliance issues

Businesses need to comply with many regulations from HIPAA to PCI and FISMA. While compliance does cost money and in terms of technology requires an accurate view of applications and technology, the cost of non-compliance is usually higher. As a rule of thumb, experts say that the cost of non-compliance is 2.5 times higher than the cost of compliance.

An up-to-date EA Inventory does not only provide you with reliable data that you can use to document your compliance with regulations. The LeanIX Survey Add-on can also help you to create ad-hoc or regular surveys for the appropriate staff to maintain accurate information about, for example, the use of sensitive data by applications.

Become more agile

Enterprise Architecture can become a driver for innovation across the organization. Free up your mind for innovation: By simplifying efforts for documentation, governance, and reporting the business can focus on driving change and making Cloud, Big Data, and Digitization a success. Modern development methods like DevOps rely on access to information and the reuse of services.

1. Empower your developers

You can lower the barriers to efficiency by simplifying efforts for documentation, governance, and reporting, as you make all information relevant to your organization’s IT landscape available in your EA solution. By automating, where possible, the documentation in IT inventories of all new services, their lifecycles, and interfaces, you can achieve faster compliance with regulatory requirements. New employees also benefit as they have all the information about their IT environment necessary for their onboarding easily available.

2. Collaborate easily

Break down monolithic structures and engage your employees inside and outside of your IT organization to use the collaboration features of your EA tool. Develop a common language with all stakeholders to avoid misunderstandings and pick up speed.

According to McKinsey, nearly 80% of the senior executives surveyed in a study said that effective coordination across product, functional, and geographic lines was crucial for growth. Yet only 25% of the respondents described their organizations as "effective" at sharing knowledge across boundaries.

They also describe the case of an engineering company. An analysis of one of the company's high-performing groups showed that a small number of construction managers and engineers single-handedly accounted for 35 % of all the collaboration occurring within it.

This kind of collaboration dramatically enhanced the group's ability to deliver expertise. Identifying and building connectivity between specialists in other groups helped the company to raise its revenue from $80 million to $275 million in a single year.

3. Lower the “complexity barrier”

Organizations comprise people, organizations, things, processes, goals, policies, rules, events, locations, and so on. For large organizations, it is impossible for people to retain and work with so many variables and bring about meaningful change unless information about them is documented through EA.

If information is locked in people’s heads it is very hard for other staff to make changes, add new product lines, or to increase the capacity of systems. When the system is changed to meet a new requirement, errors will often occur in unexpected places as the full information had not been available before the change.

This not only increases the time to make changes but also makes people reluctant to initiate change. An A.T. Kearney study reveals that companies can increase their EBIT by 3 to 5 percentage points on average with the help of systematic complexity management.

By providing all relevant information in an accessible format in your EA inventory employees can access data about the enterprise and its technology landscape whenever the need arises. Enterprise Architecture can make an important contribution to decreasing the complexity within the enterprise and therefore allowing it to become more agile.

The first 30-day EA plan

A month may not seem like such a long time but you can achieve a lot in 30 days. You want to focus on activities that deliver some early results to keep the enthusiasm of your management for your EA initiative going, while also giving you a sound basis to build on.

Concentrate on these five steps that will pave the way to an immediate clean-up of your portfolio information and thus jump-start your Enterprise Architecture efforts.

Here is how to build an enterprise architecture step by step.

1. Collect your IT portfolio data

Having reliable information about your IT landscape is the basis for all your future activities. Inspect your existing data sources and formats and clean up their content. Remove any data that is outdated or irrelevant. Assign content owners and quality check your data with them.

Once your data is refined, you can upload a sample to your EA inventory to test it. If it works out, you can complete the data migration.

For LeanIX you have the option to migrate your data yourself using Excel spreadsheets based on a template, mass import using the REST API or to update the inventory manually.

Leveraging data extraction features of existing tools can make it easier to create your inventory. Involve your content owners to improve your data quality and do not forget to test a sample.

Once you have your basic inventory, it is easy to invite more users to get involved in maintaining the information.

2. Analyze your portfolio information

Assess your portfolio focusing primarily on business criticality, functional, and technical fit. In LeanIX EAM you can rate each of these categories on 4 levels of importance.

Business criticality assessment identifies the appropriate service level and disaster recovery requirements and can be rated as:

• Administrative Service – failures are undesirable but do not affect customers,
• Business Operational – contributing to an efficient business operation but out of the direct line of service to the customer,
• Business Critical – requires continuous availability, short breaks in service are not catastrophic, and
• Mission Critical – breaks in service are intolerable and immediately significantly damaging. Availability is required at almost any price.

Functional fit shows how well applications meet business requirements today and in the near future. Ratings range from:

• Unreasonable – not enough or wrong functionality available,
• Insufficient – rudimentary functional support available,
• Appropriate – all major functions available,
• Perfect – high number of functions available.

Technical fit focuses on whether there is a need to replace services, software, or hardware with respect to business requirements today and in the near future. Ratings range from:

• Inappropriate – replacement mandatory to satisfy the business requirements,
• Unreasonable – replacement recommended to satisfy the business requirements,
• Adequate – some parts could be optimized, and
• Fully appropriate – no change needed apart from regular maintenance.

Figure 6: Functional Fit of applications within LeanIX Application Matrix

This will already give you enough data for the first analysis of their relevancy to the business and hence which applications should be invested in and which should be divested.

3. Identify and communicate possible measures

Invite everybody to keep track of planned measures in your EA solution and encourage debate about them using the collaboration features. Everybody is in the know all the time.

The high-level assessment of business criticality, functional fit, and technical fit you have implemented earlier gives you an invaluable communication instrument for the business and IT to spot improvement opportunities at a glance.

If you added lifecycle information to your applications and technologies, you can easily produce roadmaps of what will happen in your IT landscape that is available to every user.

4. Implement quick-hit initiatives

When you have collected some key data, a lot of improvement ideas will jump out at you, for example rationalizing your applications, filling gaps in IT support, or consolidating your application hosting.

Assess all your ideas by their impact and ability to implement them quickly. Focus only on the high-impact/high-feasibility projects for now.

Define your first projects in your EA tool tracking value, risk, budgets, and status. Projects should not last longer than three months.

A good place to start is to identify and divest your redundant applications using the portfolio analysis described above. Others you will want to invest in, as they are critical to the success of the business.

5. Develop your business capability maps

Business capabilities encapsulate what a business is doing right now and what it needs to be doing to implement its strategy.

Business capabilities can help you to identify redundancies, spot risks, and develop solutions, which is why it is crucial for you to develop your Business Capability Map. Start thinking about the major capabilities that your business needs to operate.

The first level should only be the critical ones, most companies have around 7-10 capabilities on the highest level. Once you have completed the first level, it can make sense to drill down one level.

On a deeper level, all capabilities should completely describe their parent capability and be without overlap. About 70% of organizations have capability models that have only one or two levels. If in doubt, go for breadth rather than depth.

6. Know your data and your interfaces

When you have identified the major applications, the most urgent improvements, and managed to get ahead of operational worries, it is time to look at data and how it drives your business.

Focus on the key data objects that drive the business and the interfaces they use. Typically, you only need 10-20 data objects to get started.

With such a basic set you can already answer questions about which applications have access to certain data and which do not; which information is classified and which can easily be moved into the cloud or who will be affected by an API change.

Look at how the information flows across your application portfolio and whether there are any applications with an increased risk of failure due to their high number of interfaces.

What happens further along the road depends on your use cases. Ask yourself what your questions are and what analyses you need to answer them. You can derive your necessary data from that.

We recommend that you focus on one or two use cases at a time, implement them, and then iterate again.

How to ensure the ongoing success of EA?

1. Keep the data quality up
   Make sure you keep your users engaged, so they will maintain the data in your EA inventory. You could consider doing regular surveys to keep the data quality up or use a quality check mechanism that prompts application owners to check the correctness of the information at regular intervals.
   
   Focus on data that creates value and helps you with your use cases. Integrate automated data sources where it makes sense and make sure your EA repository becomes a trusted and reliable information source that is the basis for decisions.
   
   
2. Make data available, get everyone involved
   Have you thought about making key reports available on your intranet? Agile businesses rely on open access to information, make sure everyone who needs IT data can easily access it.
   
   Use your data to help others understand complexity, business priorities, and budgets. Do not limit involvement to the IT department. Engage the whole organization and keep the conversation going. Invite more people across departments to your EA platform to keep them informed.
   
   Technology decisions influence the whole company, so it is important to get people on board. Make sure everyone understands that digitization is not just an IT issue but will affect every aspect of the business.
   
   
3. Speak in understandable and easy terms
   It is easy to slip into jargon when talking about EA. But while EA is not an easy concept to grasp, using vocabulary that people do not understand makes it even worse. Adjust your message to the target audience; for example, speak of quicker times to market with the marketing department.
   
   
4. Help teams solve real problems
   Make sure to provide fast and real benefits to win allies. For example, assist the team that executes the acquisition of a new company by providing them with information about the future IT landscape.
   
   Do not get caught up in complex models that will not mean anything to anyone but you, but focus on tangible benefits.
   
   
5. Gradually integrate EA into your company’s processes
   Initially, it makes sense to focus on only a few areas and use cases. But over time you should tie EA into all of your company’s processes. Capture all responsibilities from the business and technical side and introduce policies that include providing all IT-related information before a project is approved.
   
   Do not forget that establishing a successful EA program needs time and will go through stages. But once you have demonstrated first successes you will be able to use EA more actively rather than just reacting to requirements.
   
   Eventually, your EA will be so well established that you can focus more on identifying opportunities and disruptions and thus drive business transformation and innovation.

Conclusion

Always remember; EA is a constantly changing structure that changes with organic growth and developments in the company.

The more accurate the manually entered or - even smarter - imported data, the more conclusive the overviews, reports, and ultimately the company-wide optimization potential will become later.