Building an Enterprise Information Architecture: Top-Down or Bottom-Up, Which Is Better?

metaphacts is a Gold Sponsor of the Knowledge Graph Conference 2025. Get your ticket today or learn more about sponsorship opportunities.

***

Executing any type of enterprise-wide change can seem intimidating due to the processes involved and the need for alignment among various stakeholders and departments. 

In a recent blog post, we discussed how an enterprise information architecture (EIA) with a semantic model helps drive communication and understanding between all stakeholders and create transparency for change. However, introducing an EIA itself has this same challenge as each department has its own pre-established priorities, influencing its decision to invest in such a new project, given the effort and paradigm shift involved when introducing new technology or processes. 

But with the right preparation, introducing an EIA to your organization can be more manageable and achievable than you think (as we’ve seen happen with previous customers).

In this blog series on enterprise information architecture (EIA), the metaphacts team has already covered:

• What is enterprise information architecture
• How EIA supports the creation of digital twins
• How semantic models work to elevate the EIA

But there is still a fundamental question to address: How to get started.

In this article, provided by the team at metaphacts, we’ll explore the two typical approaches for implementing an EIA and offer recommendations for how you can ensure a smooth transformation of your data environment.

While there’s no one-size-fits-all approach, we’ll cover what strategies you can use to bring an EIA into your organization as well as any considerations you’ll need to keep in mind for implementation to be a smooth success. Let’s dive in!

The Top-Down Approach to EIA: Driving Change from the Top

The ideal scenario for integrating an enterprise information architecture in your organization is when it’s driven from the top. 

This top-down approach is when senior management is the initial driver for making the EIA happen. They decide that the organization needs an EIA and determine that it would be beneficial for everyone working with data to own and embrace the EIA. 

Executive leadership then designates a specific team to support or govern the project – most often Enterprise Architecture. However, this could also be the IT, Information Architecture or Data Governance teams.

This is the ideal case because having C-level executive support opens the doors to more resources, priority and investment. It can also make the project a priority for the necessary stakeholders involved, essentially giving you a good foundation that guarantees this will be a long-lasting and sustainable change. With EIA essentially being a process and culture shift towards a data documentation culture, it will affect anyone in the company who touches data.

Motivations for senior management to drive such a major architectural transformation often includes the desire to:

1. 1. Bridge the semantic gap: Establishing a common language that all teams in the organization can understand streamlines communication and ensures alignment on the same goals, targets, plans, and activities.
   2. Bridge business and IT: Bridging the gap between business and IT empowers users to analyze new business initiatives for IT impact through the EIA without burdening IT representatives who likely have other existing priorities. Likewise, an EIA understands the business impact of IT changes (including unplanned changes such as outages) without requiring the involvement of business representatives – both groups then spend more time delivering their core services.
   3. Address the disconnect between data catalogs and business processes: Business critical information that describes business processes, business objects (the input and output of business processes), roles and responsibilities, IT systems, and data products are explicitly defined, connected and mapped into a unified, coherent business object framework (a.k.a., the semantic model). 
   4. Easily find answers to questions about enterprise systems and operations, such as:
      • How will a new CRM system in the U.S. impact invoicing? 
      • Where can we find customer data for an API to enhance customer experience? 
      • What IT changes are needed for business process modifications?
   5. Map success: Align KPIs and OKRs across the business while simultaneously monitoring the impact on both business and IT, understanding who needs to be involved and determining how initiatives should be driven.
   6. Accelerate change: Accelerate innovation and improve operations effectively by enabling everyone to communicate about the business (and its IT systems) in the same way.

Case Study: A Large German Automotive Manufacturer

One example of this top-down EIA approach in action comes from a large German multinational manufacturer of premium vehicles and motorcycles. 

This enterprise organization benefited from having a senior executive lead the enterprise information architecture project from the start, effectively shaping how the technology would be adopted across the company. Since they determined it would be a priority, a plan for governance, implementation and onboarding was set from the beginning and established within the Enterprise Architecture team, then executed within 12 months to a global rollout. 

While this rollout was built on prior experience in enterprise architecture and information architecture, it still required many changes in the responsibilities and approach to these topics. Having these critical discussions early on made it possible to be flexible and allowed the time to make these decisions. 

These motivations not only resulted in technological improvements in the organization but also contributed to a broader cultural shift in how the organization views and uses data. Rather than operating as an organization made of individual departments, processes, and systems, an EIA introduced a more holistic and unified approach to information. 

Today, the EIA surfaces crucial metadata that can be found within the data so that it can be shared widely and easily across the organization and wielded for decision making in all areas of the business. 

What to Consider with a Top-Down EIA Strategy

If you’re adopting a top-down strategy for your enterprise information architecture implementation, here are a few things to consider.

1. Get Crystal Clear with Your Vision

Do you have a vision of what you want the EIA to accomplish for your organization in the long term? While the value of an enterprise information architecture might already be obvious to you, some may still be unaware of all that it can offer your organization.

For example, an EIA enables you to make informed architectural decisions so that you can optimize processes for speed and efficiency, plan for future growth, expand into new markets, incorporate new systems and technologies, and manage risks effectively by understanding dependencies between business objects (BO), business processes, and IT infrastructure. 

This vision is central to your EIA success as it’s why you are initiating such a massive change in the first place. So you must have a wider, explainable vision because you’ll need to present and revisit it repeatedly during the course of the project. The motivations listed above are simply part of the vision, but you need to have a clear picture of how these goals come together, and how they will shape the future of your organization. 

This vision is meant to inspire, motivate, and encourage team members across the organization to embrace the new technology as well as the cultural shift that comes with it. A good vision also ties back to your company’s P&L statement and highlights how costs will be saved or how time-to-value gets shortened.

2. Management in the Driver’s Seat

Since you’re imposing a new change from the top down that directly impacts current processes – most of which have already been in place for many years – it will naturally prompt hesitation and skepticism from employees. And even more so for EIA, because you’ll be actively addressing how each department tends to guard the data it produces and processes.

Generally, any change will impact processes – which is often the root fear behind change – but with an EIA, processes are captured and understood from start to finish. This process mapping makes it possible to actually change these processes smoothly because there is a comprehensive understanding of how the enterprise operates in its entirety. An EIA could even serve as the central tool to analyze impact and conduct actual vs. target comparisons. 

Making the EIA a priority for the entire organization early on, and being able to communicate the vision behind the change will help ease any worries that individuals may have about the new approach. Demonstrating the short- and long-term benefits of an EIA, for individuals and the wider company, will serve as a compelling motivator to encourage support from all parties involved. Continuously reiterate the vision and allow for open communication throughout the process so that all stakeholders have the opportunity to share their input and expertise. 

3. Collab & Conquer

Building an EIA requires collaboration with multiple stakeholders and departments. For example, you’ll need to get consensus on many items early on, such as:

• How to select the most relevant business objects for your organization
• How specific BOs are named and defined
• Who is responsible for each BO

Business objects represent the input and output (or intermediate results) of business processes, and they must be named by a term that is perspicuous, explicit, and harmonized within the organization. As a result, relevant input for decisions can come from several departments and levels of seniority. 

Logistically, it may be difficult to get everyone in the same room and reach an agreement, especially with individuals who might not yet understand the vision. Organization, communication, proper tooling, and again, promoting your vision are key to earning support and smooth collaboration from stakeholders. 

In a future blog post, we’ll propose how you can best define things like roles, responsibilities and dependencies through collaboration with multiple stakeholders. 

4. Trust in Your Blueprint

Although your EIA won’t be built overnight, you’ll already be able to see immediate improvements to your organization and the value being generated as you progress in its development. 

For instance, many organizations have utilized the linking of business objects to physical objects to automate the provisioning of new API services or support developers in finding the physical data sources required to build a new service, which was defined based on the BOs. These might seem like small wins, affecting only a specific team or part of a process, but they can serve as evidence of what could be accomplished down the road. 

It’s important to encourage your teams to identify which of these improvements can be attributed to the EIA. At least two to three such benefits should be apparent within the first three months of building your EIA.

Trust in your vision and continue to share it widely – especially including these initial successes – and you’ll see that your organization will become one that no longer perpetuates silos, but fosters more open collaboration and empowers its employees to seek out and interact with data independently. 

Benefits don’t appear only once the EIA is complete, but from the moment you decide to make the change. 

The Bottom-Up Approach to EIA: Start Small, Move Fast

If you’re not part of the executive team, you may find yourself starting with a bottom-up strategy instead. 

This process may naturally begin in a couple of ways. For example, you may be an individual or part of a departmental team that recognizes the value of enterprise information architecture and wants to make a case to senior management about how it can benefit the organization as a whole. To do so, you’ll need to demonstrate its value by focusing on a specific use case and presenting its success as evidence of what EIA offers the entire company.

Another way EIA can emerge from the ground up is as an inadvertent result of tackling a specific use case with EIA and then seeing the greater benefit for the organization. As an example, you might be a data scientist tasked with providing insights that require integrating two or three data sources. Aware of knowledge graphs and the benefits of a semantic model-driven EIA, you determine that it is the best solution for the task. Since the amount of data sources and relevant stakeholders is smaller, the alignment process is simpler, the approval process is faster, and you can deploy your solution more quickly. After a fruitful launch, you might decide to pitch EIA to senior management as a company-wide solution.  

Take a look at one of metaphacts’ customers, an American multinational food corporation, who initially sought to reduce the time and costs spent analyzing its IT systems before each merger or acquisition as well as any system retirements, integrations or consolidations. Maintaining up-to-date documentation was impossible due to its existing wiki-based approach and missing critical company-wide knowledge led to the need for external consultants to support these assessments. While their goal was to build a documentation culture that empowers employees to document essential systems and processes, the outcome of our collaboration was the development of an enterprise information architecture based on a semantic layer.

Similar to this American food corporation, starting with a bottom-up approach can be motivated by:

1. Resource limitations
   • Budget: If you lack the resources or budget, beginning with a bottom-up approach might be the best way to present your case for the value of enterprise information architecture. Even without executive-level support, starting with a couple of smaller projects can offer concrete results that can motivate buy-in from key decision-makers to introduce this data and culture shift across the enterprise. 
   • Skill gaps: Being unsure if and where necessary skills exist in the organization might be a blocker for a global rollout. However, with a much smaller scope, it’s easier to build the skills within the smaller team that is driving this forward.
2. Existing solutions: There may be existing investments in solutions like data catalogs, and investing in a small EIA use case can be a controlled, manageable initiative that not only delivers value but also highlights how integration could look in that specific case.
3. Testing: Organizations might first observe how the EIA works in a contained environment, gather insights, and assess its impact. If successful, they may later consider adopting a top-down approach, where leadership drives its uptake across the entire organization.
4. Launching quickly: By starting small, you’ll be able to make progress quickly and show results from your efforts immediately with as few resources as possible. 

What to Consider with the Bottom-Up EIA Approach

Here are a few things to keep in mind if you’re adopting a bottom-up strategy for your enterprise information architecture project.

1. Start Small, Scale Smart

Starting with a singular, isolated use case can be an excellent way to begin your organization’s EIA journey. Although some initial modeling and governance decisions may need to be revisited as the approach scales across the company, this early work establishes a solid foundation for further discussions. 

For example, defining the required business objects for your use case helps set the stage for a broader embrace. Even if definitions and terms need to be refined later, having some initial agreements in place is a strong starting point. This is a clear step forward, as it’s easier to build on partial agreements than to start from scratch.

2. From Quick Wins to Big Buy-In

Justifying resource and time investment is often easier for a use-case-specific solution, as there is just one clear, measurable impact that will be measured during the project, whereas securing buy-in for company-wide utilization requires additional advocacy and pitching.

3. Move Fast & Smart

If you are thinking of starting with an EIA use case, these are some questions you might be asking yourself these questions: 

• Is my use case small enough to move fast, but relevant enough to make an impact? 
• How do I identify a good use case? 
• How do I minimize negative impacts or delays? 
• Which stakeholders do I need for this project? 

How you go about implementing EIA within your organization ultimately depends on different motivations and factors such as influence, timing, budget and resources. 

What matters most is having a clear overarching vision. Whether you are a senior executive with the authority to introduce EIA at an enterprise level or a data engineer with valuable technical insights to share across departments, your vision will be the driving force behind effective adoption. 

The metaphacts team has the experience from many such projects. We’re happy to share our expertise and help you find the right use case to start with. Contact us to speak to a friendly expert about your needs. 

What Do I Need to Guarantee Success?

To maximize your chances of success with an EIA project, keep these three key factors in mind:

1. A Clear Vision of What You Want to Accomplish

By now you should already know that you must have a clear vision of what the organization can achieve with an enterprise information architecture, and you’ll need to be able to communicate it in a compelling manner. 

Your vision is the roadmap that will help you garner support and keep teams motivated to move forward. You must not solely focus on a technological transformation but also on a cultural and conceptual one too. 

But what exactly is that grand vision?

It might be rooted in a desire to transform your organization’s entire approach to data and cultivate a documentation culture that is proactive rather than reactive. Instead of having to constantly scramble to find information at the moment, setting up a culture of data documentation – where there is a clear and easy process for documenting changes about enterprise information in a digital manner – can change the way data and decisions are handled. 

Decisions can then be made more quickly and with more confidence, as considerations are based on meaningful context (i.e., metadata) and a full overview of your organization’s operations, which will have a lasting impact on your enterprise operations and its future. 

2. Real-Life Use Cases That Demonstrate the Benefits of an EIA

For both a top-down and bottom-up strategy, you’ll need to have real-life case studies that illustrate the tangible benefits of a semantic model-based EIA. See these metaphacts customer success stories for examples. 

3. Acceptance That EIA Is a Long-Term Investment

Initiating a change that will have both intangible and material impacts on your organization takes time. While an EIA offers significant, long-lasting benefits, you will begin to see a return on investment in the short term as well. 

With these three elements addressed at the outset, you’ll lay the conceptual foundation for building your EIA, which will set you up for whichever approach you choose.

Is There a “Right” Approach? 

There is no “right” approach to starting your EIA journey, but there is an ideal one. 

A bottom-up plan can be a good starting point, but the optimal strategy involves transitioning to a top-down approach. To achieve maximum impact, the EIA must become a company-wide initiative, with leadership driving its adoption and integration across the organization. This approach ensures that the technology aligns with the company’s overall goals and is fully embraced at all levels, improving the likelihood of adoption and sustainability over time.

The ultimate objective of launching an EIA for a specific use case should be to convince senior management of its broader value, demonstrating why it should be a strategic priority and ultimately, a top-down initiative to ensure smooth rollout across the organization. Therefore, even if you’re only beginning with a specific use case, it’s important to aim for top-down support from the outset.

Both approaches have their challenges and strengths, and you may even find yourself in a situation where there is only one path forward, but with metaphactory and the friendly experts on the metaphacts team, you’ll be able to accomplish your EIA goals with either strategy.

Building Your Enterprise Information Architecture with metaphactory

Our enterprise knowledge graph platform, metaphactory can help facilitate the creation and management of your EIA. 

metaphactory is a FAIR data platform that empowers you with semantic knowledge modeling and knowledge discovery capabilities. It enables you to capture and organize business-specific knowledge in explicit semantic models, extract insights from your data and democratize knowledge across the enterprise.

Semantic Knowledge Modeling

metaphactory’s modeling capabilities support modeling of your enterprise information in three layers: 

1. Conceptual: modeling of business objects and their attributes 
2. Logical: capturing a business object’s logical definition and its relations to other business-relevant concepts
3. Physical: capturing of connections to IT systems that store the physical representation of concepts

The technical features of metaphactory’s semantic knowledge modeling include: 

• A visual ontology editor
• Vocabulary and taxonomy management
• Data catalog integration
• Collaboration, versioning, and metadata curation
• User-friendly modeling interface for collaboration with non-tech experts
• System based on open standards to facilitate interoperability and reusability

Knowledge & Insights Discovery

With metaphactory, you can also dive deeper into your data and unlock insights and discoveries that help drive business decisions. Here are a few examples:

• Map and visualize all physical systems through a visual interface to determine which ones would be impacted by a new data protection law in a specific territory
• Easily parse insights with all relevant stakeholders through a user-friendly visual interface
• Identify early on the business processes affected by a system shutdown – such as CRM downtime – to mitigate the impact on processes like invoicing and marketing automation
• Semantic search and conversational AI interfaces can build on the semantic layer and enable any user to easily interact with the data, while receiving high-quality, trusted and verified results

Try It Out for Yourself

Speak with one of metaphacts’ experts to discuss your organization and specific use case, and learn how metaphactory can support your EIA journey. Ask for a demo or free 4-week trial of metaphactory.

 

Talk to an Expert at metaphacts

 

Come meet the metaphacts team at the Knowledge Graph Conference in New York City from 5-9 May 2025. Get your ticket today!