Tuesday, January 17, 2012

Capgemini's CTO on Why Cloud Computing Exposes the Duality Between IT and Business

Transcript of a BriefingsDirect podcast in conjunction with latest The Open Group Conference in San Francisco. Capgemini CTO Andy Mulholland discusses the transformed enterprise.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: The Open Group.

Register for The Open Group Conference
Jan. 30 - Feb. 3 in San Francisco.

Dana Gardner: Hello, and welcome to a special BriefingsDirect thought leadership interview series coming to you in conjunction with The Open Group Conference this January in San Francisco. I'm Dana Gardner, Principal Analyst at Interarbor Solutions and I will be your host throughout these discussions.

The conference will focus on how IT and enterprise architecture support enterprise transformation. Speakers in conference events will also explore the latest in service oriented architecture (SOA), cloud computing, and security.

We’re here now with one of the main speakers of the conference, Andy Mulholland, the Global Chief Technology Officer and Corporate Vice President at Capgemini. In 2009, Andy was voted one of the top 25 most influential CTOs in the world by InfoWorld. And in 2010, his CTO Blog was voted best blog for business managers and CIOs for the third year running by Computer Weekly.

As a lead-in to his Open Group conference presentation on the transformed enterprise, Andy and I drill down on one of the year’s hottest technology and business trends: cloud computing.

Capgemini has published a white paper on cloud computing. It draws distinctions between what cloud means to IT, and what it means to business -- while examining the complex dual relationship between the two.

To find out more about these two cloud imperatives, please join me now in welcoming Andy Mulholland, Global Chief Technology Officer at Capgemini. Welcome back to BriefingsDirect, Andy. [Disclosure: The Open Group is a sponsor of BriefingsDirect podcasts.]

Andy Mulholland: Hi, and thank you very much for inviting me.

Gardner: My pleasure. I really enjoyed reading a preview of this white paper. I read it with great interest, and what jumps out at me this duality. Why do business people think they have a revolution on their hands, and yet IT people look at as an evolution, something about efficiency of infrastructure?

Mulholland: Well, that’s because we define the role of IT and give it the responsibility and the accountability in the business in a way that is quite strongly related to internal practice. It’s all about how we manage the company’s transactions, how we reduce the cost, how we automate business process, and generally try to make our company a more efficient internal operator.

When you look at cloud computing through that set of lenses, what you’re going to see is how I can use it in that model. Some of the technologies from cloud computing, principally virtualization, give you ways to improve how you deliver the current cloud server-centric, application-centric environment.

So you see there’s an evolution, and you start asking questions about how far can we go: Would I go outside and put enterprise applications on the cloud? Would I maybe run a private cloud internally? Etc.

However, when the business people want to talk about that subject, they tend to talk about it and reflect on it in terms of the change in society and the business world, which we all ought to recognize because that is our world, around the way we choose what we buy, how we choose to do business with people, how we search more, and how we’ve even changed that attitude.

Changed our ways

There's a whole list of things that we simply just don’t do anymore because we’ve changed the way we choose to buy a book, the way we choose and listen to music and lots of other things.

So we see this as a revolution in the market or, more particularly, a revolution in how cloud can serve in the market, because everybody uses some form of technology.

So then the question is not the role of the IT department and the enterprise -- it’s the role technology should be playing in their extended enterprise in doing business.

Gardner: In the paper, it describes the IT view of cloud as "inside-out" -- so their IT-centric world, their legacy, their requirements, what they view as their mission, and how to project that outward. And then the term it uses for the business side that you just described is "outside-in." That is to say, beyond the perimeter of IT, beyond the perimeter of the business.

How is it these seemingly disjointed and confusing approaches can come together? Is there a need for them to somehow mesh and be aligned?

Mulholland: Most businesses ought to be aligned in their operations or it becomes quickly quite a problem. But if we just pick up the terms, which we used as ways to define the first word, the function inside is the primary function, and out is the secondary function, and the other way around.

Most businesses ought to be aligned in their operations or it becomes quickly quite a problem.



IT is clearly internally focused. When we look at what we do outside the firewall, we define it on the governance, security, and the risk structure of inside IT. In other words, we're worried about the exported information. We're worried about who comes through the firewall and under what circumstance. And we’re worried that when you go out with your corporate machine in the big wide world, someone might steal it with data on it.

Alternatively, you might be unfortunate enough on the Internet to pick up some nasty malware that could have some bad effects. For all of those reasons, we put very heavy governance and restrictions on PCs. They usually lock down more than only one control, but when we go outside, we’d like to hear there’s a virtual private network (VPN) to reincorporate inside the firewall for safety.

By definition, we have a way around outside-in. People are saying, well no, actually, I require very little about the internal world. I'm very focused on the external world. Obviously, that serves people, service engineers, but actually it’s quite a lot of people.

The small joke there is that people don’t buy iPads in order to get better use of enterprise IT. They buy iPads to escape the limitations of enterprise IT, because that's fundamentally working on the web outside, with very limited internal links.

When I’m out on the road, there are four services I use from Capgemini. All of them are web-mounted. I simply don’t use a VPN connection, and I don’t use a lock-down machine during the day. I tend to do it in the evenings or in the early mornings, when I have to do things that are about the more sensitive side of our operations.

The rest of the day, web-based, push email works really well on my iPad. Not to give particularly advert for that, but I also use a Windows phone. I also use social networking system, and I use a knowledge management system, and I use time and expenses recording systems.

The outside world

A
ll of them enable me to function in the outside world without a number of restrictions. Why? Because the primary task I have is to work with industry partners, clients, and various teams based on other people’s sites. All of that is about how I function in the outside world.

Now when we take an interesting example of that: customer relationship management (CRM). You can see very clearly CRM has meant how the company keeps its sales funnel, its clients, and other things inside its IT, transacts it, secures it, and grinds it into business information so it knows where it is.

Today, we talk about social CRM and when we talk about social CRM, we mean it's outside-in. We mean it's sales people using packages that can look at the person they’re selling to, find all the information about them by looking at various social sites. They can exchange through collaboration and knowledge, and share in social networks with their colleagues, any information about the account that's known or whatever is happening. In other words, it becomes an external task.

Now the two sides clearly exist together because you must keep your funnel up. You must know what’s happening. You must keep the internal clients. The other way around, the sales people want to exploit insights of what’s happening that they can gain from very different directions than classic internal structured information.

Gardner: So there are some significant advantages to users like yourself to pursue outside services, recognizing that they can get process innovation, data sharing or transference. There's an opportunity to engage with partners. At the same time, IT still needs to be mindful of its mission around security and protection.

They slowly, but surely, destroyed the business integrity of the data by all having different versions.



So I'm wondering again, not so much alignment, but somebody has to bend. Does IT need to go thinking more "outside," or do the folks who are doing these outside activities need to think more about IT? How can they meet up in the middle somewhere?

Mulholland: Now we’re back to your point about enterprise transformation and what that really means. I'm always very conscious of the fact that the phrase has been used for a long time in a variety of ways, as have many of the other buzzwords that go with it.

But this time, what we actually mean is that -- as with the last wave of where the big technology changed in late 80s and early '90s, when we brought in the PC, there is almost a direct correlation between the two [trends]. Business people brought in PCs, because they could use spreadsheet and could be more insightful in their use of information, such as it was at that time.

But what happened was that they slowly, but surely, destroyed the business integrity of the data by all having different versions. Where we went to with that was two things. We went to enterprise resource planning (ERP), which was one version of the truth. But the really important point was that we started to redesign around business process reengineering to flow all the process across the organization. Not that we had separate isolated departments, but the question was how do we flow across.

That was quite difficult at that time because it presented a lot of command and control problems. In fact, email was brought in as the answer, because you needed the names of the people along the process and you could do command and communication along the process, even if in the previous structure, department organization hadn’t fit.

Business transformation

T
hat was a business transformation at that time. It was a transformation around the way we organized our business to do business. From that, we organized our business model to be based on that.

So we use phrases like "do more with less," "concentrate on one or two or three lines where you're the number one or two in the market," etc. That was a very clear business transformation in the way we do business, the way we organize our business, and our business models.

Two of the most popular books recently, include Seizing the White Space, which argues that in the past, it was difficult to transfer your business model too far. I use an example, Amazon. If they sold books, they could sell DVDs, because fundamentally the same business process supported both. But in Seizing the White Space, a popular book on Harvard Business Press, it defines for a lot of people 19 new business models that their enterprise could adopt.

It defined the idea that actually they could do something like Amazon Web Services, where how they service the market was distinctly different from how they ran their business process and created an invoice.

A more popular book more recently has been The Power of Pull, and in all of these, the idea is that we’re really seeing a decentralization of the front office in order to respond to and follow the market and the opportunities and the events in very different ways.

That was a very clear business transformation in the way we do business, the way we organize our business, and our business models.



The Power of Pull says that I do what my market is asking me and I design business process or capabilities to be rapidly orchestrated through the front office around where things want to go, and I have linkage points, application programming interface (API) points, where I take anything significant and transfer it back.

Most of the major technology players in the software industry are pretty advanced with this in the way that they're supporting their current application-centric IT environment, developing a new environment in front of that, and offering middleware and mix the two together.

But the real challenge is -- and it was put to me today in a client discussion -- that their business was designed around 1970 computer systems, augmented slowly around that, and they still felt that. Today, their market and their expectations of the industry that they're in were that they would be designed around the way people were using their products and services and the events and that they had to make that change.

To do that, they're transformed in the organization, and that's where we start to spot the difference. We start to spot the idea that your own staff, your customers, and other suppliers are all working externally in information, process, and services accessible to all on an Internet market or architecture.

So when we talk about business architecture, it’s as relevant today as it ever was in terms of interpreting a business.

Set of methodologies

But when we start talking about architecture, The Open Group Architectural Framework (TOGAF) is a set of methodologies on the IT side -- the closed-coupled state for a designed set of principles to client-server type systems. In this new model, when we talk about clouds, mobility, and people traveling around and connecting by wireless, etc., we have a stateless loosely coupled environment.

The whole purpose of The Open Group is, in fact, to help devise new ways for being able to architect methods to deliver that. That's what stands behind the phrase, "a transformed enterprise."

Gardner: All right. So we certainly have a strong case for transformation being necessary and pressing, especially as organizations try to react to their very dynamic markets, accommodate them, and then to try to tool the means of orchestrating the processes and supporting those new market requirements.

At the same time, Andy, there's some added complexity in that, the external landscape has shifted when we think about things like mobility, which means any connection, any device, any service. Also, when we think about cloud, which is compute and development resources, as well as past and present IT resources on demand, and then we think about big data -- so real-time information and intelligence as well as greatly improved efficiencies around storage and search.

Then, I suppose, the last big variable to consider in this mix is the external economic environment. The timing is that most organizations are still facing reduced spending. They have also expectations from the customers that are more demanding.

Most organizations are still facing reduced spending. They have also expectations from the customers that are more demanding.



So, given the fact that we’ve identified the need, how can we leverage these changes in the market -- things like mobility, cloud, big data, and the requirements around efficiency and productivity -- to spur the enterprise forward?

What do we need to start doing differently that was not the same as in the early 90s with business process reengineering?

Mulholland: Let’s go back again to the conversation this morning with a client. It’s always interesting to touch reality. This particular client is looking at the front end of a complex ecosystem around travel, and was asked this standard question by our account director: Do you have a business case for the work we’re discussing?

The reply from the CEO is very interesting. He fixed him with a very cold glare and he said, "If you were able to have 20 percent more billable hours without increasing your cost structure, would you be bothered to even think about the business case?"

The answer in that particular case was they were talking about 10,000 more travel instances or more a year -- with no increase in their cost structure. In other words, their whole idea was there was nothing to do with cost in it. Their argument was in revenue increase, market share increase, and they thought that they would make better margins, because it would actually decrease their cost base or spread it more widely.

That's the whole purpose of this revolution and that's the purpose the business schools are always pushing, when they talk about innovative business models. It means innovate your business model to look at the market again from the perspective of getting into new markets, getting increased revenue, and maybe designing things that make more money.

Using technology externally

We're always hooked on this idea that we’ve used technology very successfully internally, but now we should be asking the question about how we’re using technology externally when the population as a whole uses that as their primary method of deciding what they’re going to buy, how they’re going to buy it, when they’re going to buy it, and lots of other questions.

If we go back to the basic mission of The Open Group, which is boundarylessness of this information flow, the boundary has previously been defined by a computer system updating another computer system in another company around traditional IT type procedural business flow.

Now, we’re talking about the idea that the information flow is around an ecosystem in an unstructured way. Not a structured file-to-file type transfer, not a structured architecture of who does what, when, and how, but the whole change model in this is unstructured.

It’s a model around big data, saying that there is information everywhere. How do I get the insight I want from it? And when I’ve got the insight I want from it, which is more driven by search than ever was driven by queries in the old landscape, how and where do I use it? In other words, how do I start to evoke a process between different companies?

Let’s just reiterate this whole theme about clouds, mobility, and so on, in a very simple way. It is actually the fourth generation of the Internet. Some people will talk about it being the third because they will miss out one of the stages. I would say it’s the fourth for the following reason.

Web 2 is quite important, because it showed us that actually we are focused upon people making insightful decisions, as much or more than we've ever been focused previously around the computer.



The first generation was universal connectivity. That’s what underpins mobility. The second generation was universal shared content. We could read and look at content, the beginnings of the big data model that we know today, the beginnings of the shift to the search engine model, and the way we used the big data model of the web.

The third one is sometimes not included by one or two other people. One or two of my colleagues, friends, and companies don’t always include Web 2. I think Web 2 is quite important, because it showed us that actually we are focused upon people making insightful decisions, as much or more than we've ever been focused previously around the computer.

The fourth one is that if I can connect to you, if I can see the content, if I can interact to find out that, that really is what I want to do. I ought to be able to trigger shared process. I ought to be able to trigger something that the process is from the various parties in that model. Travel, as I just said this morning, are actually able to come together to give me my version of what I want, and that includes other comments people hear about open data, etc.

If you want to see a classic example of this it's from Apple. I appreciate that I'm using Apple a lot, but I'm using it because this is relatively mature at the moment and it's pretty easy to demonstrate. Go to the Apple App Store and load iFly. If you're a frequent airline flyer, you're going to thank me a lot for this.

It takes the information which is published all the time in an open data format by various airports, airlines, etc., and consolidates it to give it a polarized view for you of the travel you're about to do. It tells you about the airport you're going to go through, you can find out what restaurants are by the gate you're going to travel from. It tells you whether the aircraft is on time/off time, how it synchronizes with the next flight you're going to make, etc.

Register for The Open Group Conference
Jan. 30 - Feb. 3 in San Francisco.

Transformation model

That is a transformation. When we talk about these elements, if we recombine them around this loose structured coupling to give different polarizations to the person, the situation, or the event, by combining those four factors, that’s what’s leading to the business transformation model.

Gardner: I guess it's important to point out here, Andy, that the stakes are relatively high. If you look at these issues and you think that it's a perfect storm that these are things that are too complicated, difficult to manage, you're just going to hunker down, reinforce your firewall, then this could be an existential decision.

On the other hand, like the CEO that you mentioned this morning, if you look at this as a game changing opportunity, 20 percent improvement in revenue and share, but at no additional cost, well, then this could be a game-changing beneficial approach.

How do organizations make sure they're the latter, and not the former? Who in the organization can be the change agent that can make that leap between the duality view of cloud that IT has, and these business opportunists?

Mulholland: Frankly, it's happening in most organizations already in much the same way as I said earlier. There's a direct correlation with what happened with the PC. If you go into many organizations today, and the advice I usually offer is go through the corporate credit cards and find out who is spending money with places like Amazon or Google or something like that, the answer is usually pretty shocking. It's much more than people realize.

CEOs are quite noticeably reading the right articles, hearing the right information from business schools, etc., and they're getting this picture that they're going to have new business models and new capabilities.



My point about that exercise is that the business managers on these systems -- which are relatively easy to do something quick around, like a quick spreadsheet was -- are actually already implementing and getting good results.

The other way around, the CEOs are quite noticeably reading the right articles, hearing the right information from business schools, etc., and they're getting this picture that they're going to have new business models and new capabilities. So the drive end is not hard. The problem that is usually encountered is that the IT department’s definition and role interferes with them being able to play the role they want.

What we're actually looking for is the idea that IT, as we define it today, is some place else. You have to accept that it exists, it will exist, and it’s hugely important. So please don’t take those principles and try to apply them outside.

The real question here is when you find those people who are doing the work outside -- and I've yet to find any company where it hasn’t been the case -- and the question should be how can we actually encourage and manage that innovation sensibly and successfully?

What I mean by that is that if everybody goes off and does their own thing, once again, we'll end up with a broken company. Why? Because their whole purpose as an enterprises is to leverage success rapidly. If someone is very successful over there, you really need to know, and you need to leverage that again as rapidly as you can to run the rest of the organization. If it doesn’t work, you need to stop it quickly.

Changing roles

I
n models of the capabilities of that, the question is where is the government structure? So we hear titles like Chief Innovation Officer, again, slightly surprising how it may come up. But we see the model coming both ways. There are reforming CIOs for sure, who have recognized this and are changing their role and position accordingly, sometimes formally, sometimes informally.

The other way around, there are people coming from other parts of the business, taking the title and driving them. I’ve seen Chief Strategy Officers taking the role. I’ve seen the head of sales and marketing taking the role.

I recognize also that there are a lot of companies where they have actually formed a whole new business division to behave differently. Again, the real example is a global company in desking systems recognizing the number of people in offices at desks is finite at best, and possibly going down, starting a division around virtual offices and supporting their employees to work away from a fixed office.

It's the same clients they're dealing with, the same customers, the same core competences. They're just reinventing a new business model to get them new revenue as there are uncertainties about the other one.

Now the question behind that was that it's clearly a business strategic decision, but there was the possibility of recognizing that it could be done, the technology existed, and the customers were changing their mind.

They're just reinventing a new business model to get them new revenue as there are uncertainties about the other one.



Certainly, recognizing the technology possibilities should be coming from the direction of the technology capabilities within the current IT department. The capability of what that means might be coming differently. So it’s a very interesting balance at the moment, and we don’t know quite the right answer.

We had CIOs who were not sure what was the right answer. Some of them came in with the PCs themselves, and some of them were business managers who took over the role and started to look to see what they could do.

So right now, I don’t know that there is a single, fixed answer. What I do know is that it’s happening and the quick-witted CIOs are understanding that it’s a huge opportunity for them to fix their role and embrace a new area, and a new sense of value that they can bring to their organization.

Gardner: So perhaps it’s going to be some organic or combination of organic and structured approaches. It could be any number of people that are the drivers in these different companies and in different verticals. I suppose what’s really important then is identifying successes, and then making them repeatable.

How do the roles, the traditional roles of the enterprise architect and the business architect come to bear on this ability to recognize successes -- the inside-out, the outside-in successes, some combination? Make them repeatable and perhaps move toward this cloud opportunity, rather than cloud as a handicap, to your company’s success?

Issuing invoices

Mulholland: Well, this goes prominently about the new world and the transformed environment, but we should never forget that all sorts of business are actually about the issuing of an invoice and proving that it was a valid invoice to an auditor.

So, that puts us firmly back in the old world. What we're really talking about is how do you move through three different recognizable layers in an organization, while remaining compliant -- the world that says we have to able to show to an auditor procedures and processes and data and methods that are all clean and good.

Then if we look above that, we have our core competencies. What is the industry we're in, and, if I put it in business jargon, what is the value that the shareholders are buying from us.

Motorcars might be an example. We have factories, skilled staff, and every detail. But in that layer, we see a very rich set of applications that enable us to, if we stick to automotives, design CAD, do things with them, etc. All we're talking about is in front of that is a new layer that asks how we differentiate.

Classic differentiation has been around brand. There's Volkswagen, Audi, Fiat and Škoda. If we take a European respective of a very successful car company, each of those brands reaches a different marketplace, and that gives them more reach than if they only had one brand.

At the back, it's very focused on the procedure, application, and data. At the front, it's very focused on orchestration of clusters of different services to seek different environments.



But that differentiation is built on the same chassis in each one of those cars. So their core competency actually gives them a core base ordered to express differentiation. Beyond that, how do the people map to the layers?

If you start looking at the business that way, you actually start this top-down. You ask where we differentiate, how do we engage with a market in a different way, or is our new business model where you look bottom up? You ask how we make sure we're issuing valid invoices?

If you check that through, that use of thread in a process that runs through from the front to the back, always has to be. At the back, it's very focused on the procedure, application, and data. At the front, it's very focused on orchestration of clusters of different services to seek different environments.

Each of those services is a definable entity with a definable task. Success starts from SOA, which frankly we didn’t do very well as an industry. It starts from the idea that we know and define each web services properly, and we define the rules in terms of how the orchestration of those can work. That’s why there is a lot of interest at the moment in business process management.

Redesigning process

W
hat we’ve eventually done is say at the back we’ve bolted the clusters together in a monolithic application and how we integrate those together, whereas at the front, our task is actually to identify spectacular small business service elements in a very well-defined manner, so that they can be clicked together to give us the freedom to redesign process on the fly in order to adjust to this new market.

So the clarity of thinking about business, the transition of that into technology architecture has not decreased at all. In fact, if anything, it’s gotten more complicated and more interesting as we now add this new layer of business to technology architecture.

Gardner: Returning to the upcoming Capgemini white paper, it adds a sense of urgency at the end on how to get started. It suggests that you appoint a leader, but a leader first for the inside-out element of cloud and transformation and then a second leader, a separate leader perhaps, for that outside-in or reflecting the business transformation and the opportunity for what’s going on in the external business and markets. It also suggests a strategic road map that involves both business and technology, and then it suggests getting a pilot going.

We're about out of time Andy, but on this sense of urgency in getting started, as you say, a lot of these things are happening already. How does it become something that you can manage, something that you can measure that becomes something that is lower risk and more comfortable for the leadership in these organizations?

Mulholland: I usually reply to most challenges I'm given about the complexity of trying to keep everybody going in the same direction in Capgemini with one very simple answer. The question is do you know who is responsible. If you don’t, you'd better figure out how you're going to make someone responsible, because in any situation, someone has to be deciding what we're going to do and how we're going to do it.

No business can survive by going off in half-a-dozen directions at once. You won't have the money. You won't have the brand. You won't have anything you’d like.



Having defined that, there are very different business drivers, as well as different technology drivers, between the two. Clearly, whoever takes those roles will reflect a very different way that they will have to run that element. So a duality is recognized in that comment.

On the other hand, no business can survive by going off in half-a-dozen directions at once. You won't have the money. You won't have the brand. You won't have anything you’d like. It's simply not feasible.

So, the object of the strategic roadmap is to reaffirm the idea of what kind of business we're trying to be and do. That’s the glimpse of what we want to achieve. In other words, do we want to go from books into DVDs or do we want to go from DVDs into web services -- the example I gave earlier.

There has to be a strategy. Otherwise, you’ll end up with way too much decentralization and people making up their own version of the strategy, which they can fairly easily do and fairly easily mount from someone else’s cloud to go and do it today.

So the purpose of the duality is to make sure that the two roles, the two different groups of technology, the two different capabilities they reflect to the organization, are properly addressed, properly managed, and properly have a key authority figure in charge of them.

Enablement model

T
he business strategy is to make sure that the business knows how the enablement model that these two offer them is capable of being directed to where the shareholders will make money out of the business, because that is ultimately that success factor they're looking for to drive them forward.

Gardner: Very good. We’ve been talking with Andy Mulholland, the global chief technology officer at Capgemini. As a lead-in to his opening group presentation on the transformed enterprise, Andy and I have been exploring some of the major concepts from an upcoming Capgemini white paper on the intriguing dualities of cloud computing.

This special BriefingsDirect discussion comes to you in conjunction with The Open Group conference from January 30 to February 3 in San Francisco. You’ll hear more from Andy and many other global leaders on the ways that IT and enterprise architecture support enterprise transformation.

So thank you very much, Andy, for joining us. It's been a fascinating discussion.

Mulholland: Thank you, very much indeed. I’ve enjoyed it.

Gardner: And I look forward to your presentation in San Francisco. I also encourage our readers and listeners to register, explore, and attend the conference.

This is Dana Gardner, Principal Analyst and Interarbor Solutions, your host and moderator throughout these series of thought leadership interviews in association with the conference. Thanks to you for listening, and come back next time.

Register for The Open Group Conference
Jan. 30 - Feb. 3 in San Francisco.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: The Open Group.

Transcript of a BriefingsDirect podcast in conjunction with latest The Open Group Conference in San Francisco. Capgemini CTO Andy Mulholland discusses the transformed enterprise. Copyright Interarbor Solutions, LLC, 2005-2012. All rights reserved.

You may also be interested in:
Post a Comment