Next-Generation Architecture

Next-Generation Architecture

Digital transformations are fostering disruptive business models, and organisations expect that their IT solutions can support business changes at an ever-increasing frequency. This is the market recognition that now business is about technology, and technology is about business. It's almost impossible to find an industry that hasn't already been disrupted by digital in some form. Meanwhile, Chief Information Officers (CIOs) executing a digital transformation must ensure that their investments achieve the desired outcomes, considering the implications of implementing new IT solutions and technologies within the context of their legacy systems.

But changes can result in complexity, risk and cost. So as one part of an organisation focuses on being stable and ensuring that daily activities are supported, another part focuses on process improvement and product innovation, trying to lower costs and stay competitive.

Enterprise architecture (EA) as a capability is an ideal way to meet these challenges by providing guidance and analysing the impact of emerging technologies, such as Artificial Intelligence, Robotic Automation, API, Cloud, Internet of Things, 5G, Virtual Reality (VR)/Augmented Reality (AR) and Blockchain. Besides that, EA provides guardrails to clarify where an organisation has flexibility and where they need to comply with tighter principles (for example, security and privacy).

In general, these are the 3 main stages of a traditional EA project:

?1. Current state mapping

Architectural representations of the main components of an organisation and how they work together. These components include business capabilities, value streams, IT systems, data, and technologies.

?2. Future state ideation

Guidance in the most appropriate business model changes and technology choices to achieve strategic objectives. An EA team helps to understand the reasons for a change (the “why” and “what” questions) and focuses on managing complexity to reduce risk and cost. Usually, they will rely on disruptive technologies and business trends to ideate future state scenarios with stakeholders of the organisation.

3. Transformation roadmap

Putting together a roadmap of initiatives necessary to deliver the future state in a prioritised manner.

A modern approach to Enterprise Architecture

We no longer have the privilege of executing an enterprise architecture project for months. Markets are changing fast, and the business must adapt and move on. So the EA's actual challenge is knowing when to stop – when it is good enough. Today, the level of detail of an EA project should be just enough to decide to move forward in a highly iterative way.

As digital business becomes essential, enterprise architects must understand innovative technologies, anchor them in disruptive business models, and use rapid prototyping and system development approaches.

This brings us to a modern approach some influential people have been talking about in recent years:

No alt text provided for this image

Figure 1. Thoughtworks - Three mindsets of product development

No alt text provided for this image

Figure 2. Gartner - Enterprise Architects Combine Design Thinking, Lean Startup, and Agile to Drive Digital Innovation

In short, combining these pieces (Design Thinking, Lean Startup, and Agile Methods) is quite different from what we used to do in traditional enterprise architecture projects. We used to gather high-level requirements and make current and future state diagrams for several months (without experimentation and adjustment cycles). Ultimately, we used to release a roadmap of multi-year implementation projects that commonly treat changing requirements as a sign of failure.

Design Thinking

Design thinking is a way to come up with new ideas using proven techniques with the right people. When you skip this step and go straight to "problem-solving," there is a serious risk that you will not understand the root cause of the problem and will invest in wrong solutions to problems you do not understand properly.

More than that, design thinking is no longer just to make new products, it's increasingly being applied to devise business strategies and to understand intangible issues such as how a customer experiences a service. With design thinking, we can do this by using tools that everyone understands (like post-its and sketches) rather than just enterprise architecture spreadsheets and diagrams.

Design Thinking techniques get around the human biases (rootedness in the status quo) or attachments to behavioural norms (“that’s how we do things here”) that block the exercise of the imagination.

Lean Startup

Lean Startup is a method that helps us to validate new ideas, turning them into Minimum Viable Products (MVP). These MVPs are quickly implemented and tested with real users to answer questions like:

Do people want this?

Does it really work?

Do people want to pay for it?

It is a way to reduce the risk of investing large amounts of money and time in new ideas that have not yet proven their worth. With Lean Startup, we test the business value of a new idea through short cycles of experimentation, learning and adaptation. By the way, not producing MVPs has been a major reason for the failure of large digital transformation programs. Companies are making huge leaps of faith in new digital platforms without evidence that they will deliver the promised outcomes. If you are exploring uncertainties (and there are many uncertainties in today's complex and dynamic business environment), making decisions through short continuous improvement cycles is the best way to achieve the desired outcomes.

Agile Methods

Agile methods (such as Scrum, Kanban, SAFe, among others) and Lean Startup have many similarities, no wonder they are both derivations of the Toyota Production System (TPS). But while Lean Startup focuses on validating if an MVP makes sense in all its aspects (value proposition, customer relationship, key partners, key resources, cost structure, etc), agile methods focus on developing and evolving (in an iterative way) the software solutions supporting an MVP. In other words, an agile method can help us continually deliver new software functionalities (or changes) with quality and business value.

Three pitfalls to avoid

  1. Don't create isolated teams for each piece of the puzzle in some "idea factory" that generates a requirements backlog for an "MVP factory" and so on. Ideally, EA people should be part of a value stream from ideation to building and validating an MVP. After that, you could delegate the MVP evolution to a product team dedicated to implementing new requirements.

No alt text provided for this image

Figure 3. MVP + Product Evolution

2. Don't design the future state architecture in a single, broad round of design thinking and lean startup. Instead, we need to slice the scope of a digital transformation into sizes that fit the concept of an MVP. For example, we can treat as MVPs the independent steps of a large workflow or the most valuable functionality of a complex business model. And don't forget to validate these MVPs with real users before you make significant investments in evolving an MVP into a broader product.

No alt text provided for this image

Figure 4. Sliced digital transformation with design thinking, lean startup and agile.

3. Sometimes, the result of an EA project will advise not only building new solutions but also procuring off-the-shelf software packages for recurring problems. But we don't need to do that without any validation. In this case, the concept of MVP remains applicable. We can run demos with real users as MVPs.

References

?Gunnar Menzel, https://www.capgemini.com/2019/02/shifting-sands-the-change-of-architecture-in-an-agile-world/

Randy Potter, https://www.capgemini.com/2018/11/ea-reboot-an-executive-point-of-view/

Peter Classon, https://www.capgemini.com/2019/07/viewpoint-digital-architecture/

SAFe Portfolio, https://www.scaledagileframework.com/portfolio-canvas/

Jonny Schneider, https://www.thoughtworks.com/insights/blog/understanding-how-design-thinking-lean-and-agile-work-together

Saul Brand , Marcus Blosch , Neil Osmond, https://www.gartner.com/en/documents/3941917/enterprise-architects-combine-design-thinking-lean-start

Edzo Botjes, https://labs.sogeti.com/three-steps-to-successful-innovation/

Jeanne Liedtka, https://hbr.org/2018/09/why-design-thinking-works

Nelson Avelino (.

Business Development Manager | Account Director | Sales Account Manager | Helping people understand that their finished goal is the starting point for their next achievement.

5 年

Good answer Marcelo!

Gunnar Menzel, FBCS

Master Architect at Capgemini, BCS Fellow

5 年

Great post Marcelo - question is how to speed up / how to avoid an EA that takes 3 month and develops something no-one uses; speed is of the essence in today's digital world!??

Amanda Rothwell

Senior Engagement Manager at Capgemini | Board Trustee at Pilotlight

5 年

Great article Marcelo. Love the idea of combining Design Thinking, Lean Startup and Agile.? Agile’s focus on collaboration, iteration, feedback and speed means that teams are less likely to waste time building things no one wants, and Lean helps adapt traditional practices to match the pace of faster development teams. Design Thinking then gives people from all kinds of disciplines an approach to explore problems and focus on users.? If everyone has a shared vision, they measure their success, set their teams up for success, & maintain users focus, I think everyone would be on to a winner with what they set out to accomplish.?

要查看或添加评论,请登录

社区洞察

其他会员也浏览了