In last week's Cortex newsletter, I introduced the following diagram of a customer-centric digital architecture, where digital architecture is shorthand for enterprise architecture that's laser focused on driving digital transformation.

Customer-Centric Digital Architecture

In the diagram above, the traditional architectural layers (represented here as concentric bands) have been grayed out in favor of customer journeys that cut across user interface, process, technology infrastructure, and data concerns, instead focusing on the preferences and behavior of individual customers as they conduct all their interactions with the company in question.

The notion of customer journeys, of course, are central to the digital marketer's playbook. Clearly, enterprises should focus their digital efforts on such journeys, as they represent customer interactions over time. But making customer journeys the centerpiece of the enterprise architecture, however, leaves more questions than answers.

The challenge arises when EAs consider the context of the customer journey in the overall architecture as well as the architectural elements that make up each customer journey. After all, dividing up the world into familiar layers like process, data, and technology is relatively straightforward. There are overlaps and ambiguities at the edges to be sure, but everyone approaches such layers with a relatively common understanding of the overall scope of each layer and what belongs within each one.

Approaching the enterprise architecture as a customer-centric digital architecture, however, requires more than drawing new boxes and sorting things into them. Rather, such architecture must take a different approach than EA has been done before, or the end result won't meet the business need - which in this case, centers on addressing always-changing customer preferences and behavior. Customers are simply too fickle in their desires to expect traditional architecture to rise to the challenge.

The Three Threads of Digital Architecture
In order to flesh out this new approach, we need to pull together three threads in the tangled weave we call digital transformation: traditional EA, Agile, and scaled self-organization.

From traditional EA we take a reductionist approach to solving problems: breaking them up into well-defined chunks that have well-defined relationships with each other. This ‘boxes and lines' notion of architecture is familiar to every architect, of course, and hearkens back to the Zachman Framework, TOGAF, and just about any other EA framework you care to mention.

Reductionist approaches, however, only take us so far, because they assume a level of understanding and stability that the real world rarely if ever offers.

If only we had a clear, stable understanding of our requirements that everybody agreed on, and only if our ‘final state' was similarly clearly defined, then our architecture would be straightforward.

We all realize by now, however, that such ‘waterfall' thinking almost always leads us astray, because rarely if ever do the assumptions listed above hold true. Instead, we should always assume that we don't understand the requirements, that there's no such thing as a ‘final state,' and furthermore, we must also take for granted that everything is always in a state of flux.

Fortunately, we have an established set of principles for dealing with situations of unclear, continually shifting requirements - the principles we loosely group under the Agile banner: iterative, customer-focused approaches that focus on solving the problem at hand instead of other, less important tasks.

So far so good, but while combining Agile principles with EA - what people predictably call ‘Agile EA' - is an important step in the right direction, it doesn't take us far enough to solve the challenges we have with Digital Architecture.

The problem: Agile principles focus primarily at the local, team level. And while there are several approaches to scaling Agile, like the Scaled Agile Framework (SAFe), for example, they all suffer the same fate: as they rise from the team level to the enterprise level, they lose the inherent agility that is necessary to satisfy the preferences and behavior of customers that form the basis of digital architecture.

Solving this problem requires us to pull on the third thread of digital architecture: self-organization. Not merely the team-level self-organization that has been a part of Agile since day one, but rather self-organization scaled across the enterprise.

The Missing Piece of the Digital Architecture Puzzle
Circling back to our original questions about customer journeys at the architectural level - the context of each journey in the architecture as well as the relationships among journeys - the answer lies in scaled self-organization, within the context of Agile EA.

Let's take an example. Say you're an executive at a large consumer bank, and you realize your interactions with customers are sorely lacking. It would be easy to say you need a better mobile banking app - and perhaps you do - but focusing on a single app rather than the entire customer journey won't solve the larger problem.

In reality, the customer journey involves not just the mobile app and your web site, but also your call center, your outbound marketing, your branches, and of course, all the back end systems that make everything work. And then you also realize that even this list is likely to be dangerously incomplete.

So, who do you turn to in order to put together all the people and other resources across numerous departments to get the customer journey right? And how do you place this particular customer journey initiative into the context of all the other digital initiatives that are clamoring for attention and resources?

The answer is to call upon the people who know best what your bank does, and furthermore, what it can do - your own employees. Follow the basics of self-organization (see my previous Cortex newsletters, Conway's Law and the Emergence of Business Agility and Fixing Slow the Agile Digital Transformation Way for more details), and then get out of their way. (Subscribe to the Cortex newsletter here.)

The Intellyx Take
What, then, is the role of the EAs in this vision of scalable self-organization? First and foremost, they are on the self-organizing teams themselves, including both the teams that form around specific business challenges as well as the Center for Digital Empowerment that helps to coordinate and empower all the digital efforts across the enterprise.

As for the enterprise architecture itself, its focus is on facilitating the transition to scalable self-organization, within the context of Agile EA. There may still be boxes and lines, perhaps, but the EAs must move each element, each box and line, up a level of abstraction.

Instead of a box representing a piece of technology, the box represents a self-organizing team's dynamic choice of technology. Instead of a box representing a process, that box represents once again the result of a self-organizing team's choice about a process.

The overall digital architecture, therefore, represents self-organizing teams' approach to answering the basic questions about the context and content of customer journeys, rather than the context and content of those journeys themselves.

Only by working at this ‘meta' level will the architecture ever account for the inherent agility necessary to address ever-changing customer preferences and behavior in the context of a digitally transforming organization.

The goal of EA, of course, is building business agility as a core competency across the enterprise. It's not good enough to change, your organization must get better at change. After all, you cannot eliminate change. You cannot even simplify it. Digital architecture must embrace change.

Intellyx advises companies on their digital transformation initiatives and helps vendors communicate their agility stories. As of the time of writing, none of the organizations mentioned in this article are Intellyx customers. Image credits: Willi Heidelbach and Intellyx.