It Doesn't Matter Where EA Lives — So Let's Stop Arguing About It

George Colony, our CEO, just released a post on his blog about enterprise architecture, aptly enough named “Enterprise Architects For Dummies (CEOs).” I retweeted the post to my followers and received a flood of responses, most of which were violently disagreeing with George’s assertion that EA works for the CIO. I think this is a pointless argument, but underscores a very important change that most are missing.

Here’s what I mean:

  • The objection to putting EA under the CIO is based on an old-school notion.That notion is that CIOs are chief technology infrastructure managers. Our data shows that the role of CIO is changing, fueled by cloud and other as-a-service technology. CTOs or VPs of IT are increasingly taking on the job we used to think of as the CIO, while progressive CIOs are evolving to something else. Locating EA under the CTO is a bad idea, we all agree.
  • Every business is a digital business.If you don’t believe me, I’ll send you a pile of research. There is no such thing as a non-information-centric business anymore — or at least there won’t be for very long, because they are going out of business. Forrester has been using the term “business technology” (BT) for a while to indicate that there is no room for having separate business and IT — it simply won’t work much longer. Even in the most paper, analog verticals, we can give you example after example; check out Monsanto’s IFS (they are a seed company!).
  • CIOs are becoming chief business technology officers. We have even gone so far as to retitle our CIO to CBTO; but even if you don’t take it to that extreme, the notion still has value. Since your business is a digital business, CIOs/CBTOs are stepping into the business design role, and that design is increasingly focused on the use of digital information for competitive advantage. Every other competitive barrier, except goodwill, has been or will be erased before long.

The practice of EA supports the C-suite’s digital transformation and digital business design. When you think about the CBTO leading this effort, it all makes good sense. I discuss this in much more detail in my report “Build Trust And Agility With An EA Process Framework.” In this light, it doesn’t make much difference where you put enterprise architects or even what you title them, so long as the firm is doing the right things with the right people.

Does this make sense? Love to hear what you think.

Comments

It matters!

I can sense that the "six reasons.........." discussions seems to have shaken the entire industry.

Brian,

I have read your arguments and the blog post by George. You have taken a very limited view of EA. If with EA you actually mean EITA, then your arguments are reasonable and it does not matter where EITA reports to. However if EA is indeed EA then it matters, because this then defines and shapes the practice, discipline, and many other things. If CIOs are becoming CBTOs - good for them, but it still hasn't got anything to do with EA.

My question would be - why can't CIOs (now CBTOs) get EA out of their heads? Why are they so fixated on EA?

Really?

if you think what I'm saying about EA is 'very limited' then you haven't read what I've said. EA is about architecting the enterprise. Since every enterprise is digital, that means EA = digital business design. The CIO is assuming the role of chief digital business design officer, so where else would you put EA?
I'm being pragmatic as wel. Suggesting the creation of a CEAO won't fly with boards, so where do you put EA? Under the CIO makes sense.

Yes, it matters

Strong agree with Pallab here, on all points.

Even when rebadged as 'business technology', the IT/business linkage is still only one often-minor subset of business-information and/or business-technology: and an organisation - and, even more, an enterprise - is a lot more than just its information and its technology. So I'm sorry, but I'll have to say this: Just about all that you and George have done in these two posts is tweak the badging a bit, whilst retaining intact almost all of the known fundamental anti-patterns for EA.

You've elsewhere dismissed my views as 'outdated': may I ask you to look perhaps a bit more closely at whose views actually _are_ 'outdated' here? After all, even the Open Group has moved on from there. I'll admit that the type of enterprise-architecture I and Pallab and others promote - the literal 'architecture of the enterprise' - would be uncomfortable for many in business, because it necessarily demands a true whole-of-enterprise and whole-of-organisation view that must bridge across every silo and fiefdom. Yet the evidence from HBR and elsewhere is clear that that's the only way that works; whereas the approach that you and George have promoted here is guaranteed only to cause a fallback to rampant IT-centrism and a field-day for IT-vendors, to the certain detriment of the organisation itself. Not wise...

On 'positioning', I'd recommend Chris Potts' view on the CIO as Chief Investment Officer, and the CEO role as ultimate Chief Enterprise Architect: see his books 'fruITion' and 'recrEAtion' at http://www.dominicbarrow.com/writing.html . Although I still disagree with Chris on some EA themes, he's right on the money here in terms of how EA actually works within large organisations.

Of course it matters

Because where you put it is a direct reflection of the organization's maturity and expectation of value. On the other hand, one of the most mature EA practices I know doesn't have an EA team at all. They have taken TOGAF, distributed responsibilities among appropriate execs and execute EA virtually.

As to your statement - 'architecting the enterprise' > that's exactly right and the term that you and I used when we visited and a term I've used with clients. My and George's post in no way conflicts with this idea. That's what you and Palab, two EAs I have much respect for, are missing. Sorry to call it out so directly.

Please follow my logic -> there is no such thing a non-digital business (excluding the very small ones that don't need scale.) Every business of any scale relies on digitized information as its lifeblood in today's environment. See my Integrated Farming System example above. This is just one of many I can discuss with you.

The logical conclusion is that if every business depends on digital information, which is 1's and 0's, then they depend on information technology because you can't count bits on an abacus. To wit, there is no difference between a business strategy and an information technology strategy in a digital business world

Next point - we find over an over firms still trying to align an IT strategy to a previous stated business strategy and failing. This is old school, aligned IT behavior. To get beyond this, we are urging out clients to start with rethinking their IT organization. Change starts with lexicon, so we are meeting our clients where they are (IT/CIO/EA) and pushing them to a new concept - (BT/CBTO/EA).
EAs in an ideal CBTO organization are architecting the enterprise, and the CBTO is not a Chief Infrastructure Manager retagged as you suggest. Rather they are the executive digital business architect and they lead digital business design.

Brian, Change starts with

Brian,

Change starts with lexicon - absolutely. That's precisely why me and Tom are saying that EAs must be EAs.

In an entirely digital business the Chief (Digital) Enterprise Architect must be placed higher than and outside of the CBTO. Digitization is lot more than BT, just as automation is lot more than IT.

(Digital) Enterprise Architecture > Business Technology Architecture.

See that's where we disagree

Regarding - "In an entirely digital business the Chief (Digital) Enterprise Architect must be placed higher than and outside of the CBTO. "

You keep implying that the CBTO is a glorified CIO who is till mostly being Chief Technology Infrastructure Officer in a keep the lights on or trusted utility mode of IT - I guess this works for your argument, but it is not what we are saying.

The CBTO is ultimately the Executive Chief Architect, interfacing with other CXOs and the board to help (or even lead) the digital business design. As much as we can opine about architecting the enterprise as some kind of a gestalt thing we just kind of do; the reality is that CEOs need help leading their organization through digital business transformation. Who is going to take this leadership role?

This CXO leader, call her what you will, must have a VP level leader to do the next level down strategy collaboration needed to turn vision into reality. That person can have any appropriate title, but we have chosen Chief Enterprise Architect.
Its perhaps illustrative that in several conversations I've had w/ CIO's about this vision of EA and the skills needed to pull it off, the comment back to them is "that person sounds like me." To which I said, "yes, you are the chief digital business architect!"

Then we disagree as well

I am sorry Brian, but I have to disagree with you as well. You're view on business is too limited. Much of today's business may be digital or depend on digitization, but that is really only a part of it. I agree that information technology has become more mature and more important and can be considered as a production factor in its own right and therefore could be used as a design perspective for the business, but its is only one of the factors (like building materials, labor and money). There is and will necessarily be a lot of analogue and even unformalised business as well. Think of goods, building and most importantly people. So yes, there is lot of non-digital business. As I said, your're view is too limited. The digital business is still to a large extent an image of the physical business, helping us to understand what is actually going on in that complex System we call the real world. An Enterprise Architect should engage in all those aspects.

Considering the fact that the Enterprise Architect is dealing with the largest scope possible in an enterprise, the position of the Enterprise Architect function does matter, significantly. We should not be part of the technology-oriented people nor should we take technology as our only perspective on business. It is one of the perspectives to consider. In order to take that holistic view the Enterprise Architect should be positioned as high as possible in the organisation, avoiding a technology-oriented label.

At the same time we should acknowledge that the EA is not making the decisions. The EA provides management with the best possible insight to make well-informed decisions, through an Enterprise Architecture.

Business that get we are going digital will win

Thanks for chiming in. I think what your missing is that the information age has led to extensive commoditization. With only a few exceptions, like the Coca Cola recipe, there aren't many competitive barriers left and there will be fewer still in the future. So while businesses do have other things - people, money, plants, etc; all of them are becoming commodities and thus are not going to be the basis of competition except for those very few large enough to make a profit based on very small price differentials.
For all the rest of business, success and profitability are going to depend on 1) how well the company turns the knowledge that it has into information that it can use to drive profits, and 2) the loyalty of customers and partners, or "good will".
While the first point's relationship to digital information is obvious, the second may not be - so here it is. We are seeing more and more firm's turn to customer intelligence. Gathering as much information about their customers as possible (digital information), and using it to foster deeply loyal customers.
What about money? Firm's are turning to data science and advanced analytics to manage their financial risk. What about physical facilities? Firms are using digital information to optimize manufacturing, logistics and supply chain operations. The list goes on. I can't think of one aspect of business that isn't revolving around the use of digital information to make it better.
I'm writing about this now in my current research where I'm talking about the transformation of business from analog to digital. We are smack in the middle of that right now, and businesses that get that and get it right will be the winners.

Yet another 'it matters'

If you accept the literal interpretation of EA as 'architecture of the enterprise', as I do in company with Tom and others above, then it is critical that this function is able to invent or re-invent all enterprise capabilities and processes. This means that EA must be in a position to challenge any of the COO, CIO, CFO etc territories, without getting boxed-in by natural vested interests of such roles. So EA needs to be relatively independent, sponsored at the highest executive level that is able to drive generic change, which in practice means CEO sponsorship, and preferably a direct line of communication with the Board.

Perhaps ironically, I think we are also seeing a certain maturing towards a more literal interpretation of the CIO role, as being about information and systems of information in business (irrespective of any technology) rather than about information technology services management, which is a different focus, and perhaps best addressed by a subordinate role. The more mature CIO's architecture viewpoint shifts from EITA to a broader EIA (or perhaps EISA), rather than EA.

EA should not live in the Winchester Mystery House

Hello, Brian. This is E.G.Nadhan, Distinguished Technologist HP.

Great post. The last line in your post resonates with me: "In this light, it doesn’t make much difference where you put enterprise architects or even what you title them, so long as the firm is doing the right things with the right people."

It is extremely vital that the firm is doing the right things with the right people. The Winchester Mystery House in northern California is a classic example of how not to do Enterprise Architecture. Therefore, I agree it does not matter where you put enterprise architects -- as long as they don't live in an enterprise that is being built like the Winchester Mystery House.

Please check out my post http://bit.ly/zrKMHL at your convenience and let me know what you think.

Twitter: @NadhanAtHP

Six Reasons Why EA Should NOT be Assigned to the IT Department

6. Notwithstanding history and evolution, EA ≠ IT Architecture.

5. True EA leads to redistribution of authority and reallocation of accountability, both beyond CIO jurisdiction.

4. EA value proposition and benefits are solely business realizable.

3. The primary goal of EA is to build coherent enterprises, not better IT systems.

2. Organizations are complex adaptive systems, hence holistic synthesis takes precedence over fractional analysis.

1. EA failure is an organization failure, not an IT failure. Resistance to EA is a consequence of failure, not the cause for it.

Don't disagree

Everything you say makes sense, but it provides no actionable prescription that doesn't start with the board or CEO taking action. Since most CEO's and boards are not ready to move EA up to "the enterprise", what do EAs do in the mean time?
We try hard at Forrester to provide advice that is practical for the now while leading towards the future.

What's the Diagnosis?

Brian,

It is nice to know Forrester (I presume) agrees to these points. These points are a means to shift existing (and faulty) mental models (the most difficult and important aspect to change). Changing mindsets is where we start.

Before we get to "what", should we be spending sometime understanding "why" EA is where it is now - in the ditch. I fully understand that Forrester provides advice that is practical for now while leading towards the future. So being consistent with the same goal, what is Forrester's diagnosis of the current situation? It looks a bit strange to even ask about prescriptions (what) without trying to comprehend the "why".

Once we have this, we will move to the prognosis.

Pallab

Ten Factors Shaping Enterprise Architecture

1. Enterprise Architecture > Enterprise IT Architecture; the scope and footprint of EA is the enterprise in its entirety.

2. Every functioning enterprise has an architecture (formal or non-formal), if it didn’t, it wouldn’t be able to function.

3. Enterprises are complex adaptive systems, wherein the assumptions of reductionism, linearity, correlation and predictability are not valid.

4. Enterprises are inherently characterized by diversity, inter-dependence, ambiguity and flux.

5. The emphasis of EA is not on the things (domains) but the interrelationships between the domains.

6. Enterprises operate in a networked (heterarchical) mode, as opposed to hierarchical mode is used for administrative purposes.

7. Complex adaptive leadership wherein the power of influence exceeds the power of control, is the most effective architecture leadership style.

8. Improving alignment in enterprises makes them more rigid and less agile, because a complex system is approximated as a complicated system; and in doing so undesirable limitations and constraints are introduced into the system (i.e. the enterprise).

9. Enterprises (and EA) are characterized by holism, synthesis and systemic paradigms; and any assumption that the effectiveness of the whole will be achieved automatically, as long as the parts are optimal, does not hold true.

10. Enterprises are architected for flux, as opposed to being architected for stability.

Therefore, Enterprise Architecture is defined as the ongoing process of building the ability to manage complexity, with the pivotal goal of creating and sustaining coherent enterprises.

Out of curiousity, does the

Out of curiousity, does the EA team at your organization reports to the CEO? And how many percent of your customers have such similar arrangement?

'Since most CEO's and boards

'Since most CEO's and boards are not ready to move EA up to "the enterprise", what do EAs do in the mean time?'

Its not so much a matter of moving EA up Brian, as if its not up its not really EA (literal), which is what some of us have been saying for a while, that currently less than 5% of the positions being advertised as 'EA' are actually that!

The reality is though that, as I think we all agree, the majority calling themselves EAs are in fact practicing IT Architecture, focused on delivering effective technology services, to which this conversation probably matters little. So in the meantime they can just carry on.

For those of us who are passionate about expanding that 5%, we will in my view have the best chance by shifting the perception of EA as being primarily about business process architecture.

There are costs to moving architecture uptown

Reporting structure is never enough to determine behavior. But it should, at least, provide a path of least resistance to good behavior. I think it matters.

For example, does it matter to you who your boss is? Do you think a CIO may be incented differently if he or she reports to the CFO rather than the CEO or let’s say to the head of administrative services. Do you think that a CBTO or CIO reporting to the Chief Council (I know two) would be in the same formal and informal conversations that decide so much of the direction of the company? And might not there be a difference between what keeps a CEO up at night vs. a CFO; and wouldn’t they affect how they manage subordinates?

Ok, let’s say hypothetically, it does matter (though it’s not the only thing). Where should it report?

Chief of EA as a peer to the CIO or CBTO is a statement of clout and people look at these things. And his or her scope, if you include business architecture as a part of this, is beyond IT (it includes business processes, after all). This pushes us outside of IT and higher up.

However….

The chief architect’s boss needs to understand what architects do to measure their performance and occasionally help them fight battles (think what a CIO might do when an IT architect has to make a hard call about a system). On the other side, the chief architect’s political and communications skills will need to be in sync with others at this level – it’s dangerous flying this close to the sun, especially if you’re the one who’s trying to make a change. Furthermore, unless the chief architect is structurally separated from the rest of the architects, the entire group is separated from IT. Multi-headed IT shops nearly always have conflicts between the organizations (one three headed IT shop called its’ leadership “fluffy” after the Harry Potter dog). Finally, the CIO or CBTO has an organization that the entire company is completely dependent upon. That’s a very strong base of power that can be used to make a change. Is it better to divide that base of power between two people?

Finally, there are some practical limitations around span of control (shouldn’t security, QA, innovation and other functions also report outside of IT because of scope, independence and the need for clout) and this reporting and associated requirements will reduce the candidate pool for architects (how much harder will it be to find them).

So there are trade-offs but I shouldn’t be wishy washy. I think in most cases, architects should report within IT.