Archive for June, 2010

Unbundling the bank

Just over 2 years ago I first wrote the post entitled “unbundling the bank” and hear I am finding myself planning a talk at CloudCamp next week entitled the same! The title, I thought, came to me earlier this week when I had chance to visit a team of Enterprise Architects at a large UK bank. I was sure the title was not “original” and came from somewhere, so you can imagine my surprise when I did the necessary search and my blog came out on top – lol!

What I really love though – is that the thoughts are just the same, the evidence however continues to mount, pointing towards the further differentiation of IT, but increasingly of the business itself, from “centralised vertical models to decentralised models made up of “single value specialists”.

Unlike my talk to bankers back in 2007, this time, the EAs are thinking about the implications of cloud and software+Services more seriously, and while most would say never, to some the light bulb is on and the opportunity for disruption is imminent!

One thing I really liked and had forgotten about the talk by Craig Heimark back in 2007 is the drive to “create increased value to the consumer at the expense of the margins; the higher the volumes the lower the margin” which to my mind reads “commoditisation”.

In the old world view of innovation, it would make sense to keep this closed or hidden, to maintain ones market lead at the expense of competition. Thereby keeping margins high.However, in an open innovation model the race to commoditise is high, Keeping margins low to avoid competition and drive to mass market appeal and therefore scale quickly. If an innovation does not scale in terms of consumer appeal then it’s not an innovation worth pursuing.

Although, it takes time for shift to happen, there are examples all over of this taking place. Below is a list of common occurrences that may take place in isolation or combination but are enablers in moving towards a more “composite enterprise”. I’ve covered some of these previously so some repetition might arise.

Rise of Multi-sourcing

As discussed previously. Enterprises that have outsourced are actively re-insourcing, but in so doing they are differentiating between what comes in and what stays out. The latter are often non-differentiating or commodity, but by going through this process the results seldom remain with the encumbent outsourcer but move to new specialists.

Service Provider Convergence

The traditional models of software suppliers like system integrators, outsourcers and ISVs are converging. The SI is moving away from delivery of bespoke software to the delivery of bespoke services, software no longer lands inside the organisation’s datacentre but is automatically outsourced by the integrator. The integrator has a shared risk-reward with the customer – they are innovation partners with a desire to grow the market of the service itself, not just the value it itself fulfils.

Delivering Innovation as a business

Many enterprises are project-driven, but this is a mis-understanding, it is more that they are project-organised, they are in fact innovation or opportunity driven. The problem is they don’t realise this due to the project-based mentality that results. By taking an innovation-driven approach it is easier to see this as a business investment and to consider creating a structure that mimics more of a business than a project. this has dramatic and profound results both in terms of outcomes, but also to the people involved, the processes they create and the resultant capability that they generate.

Innovation Hubs

Closely related is the rise in innovation hubs or R&D centres, that incubate or provide support to new pilot innovations. This extends outside the enterprise too, to the service provider communities where a joint risk reward reduces up front costs of innovation in favour of longer term commitment and profit, leading to the model of collaborative innovation between partners.

Rise of the new COTS Service

A great example of the re-rise of COTS (Service) is the growth, adoption and subsequent mass-customisation of the large ERP systems over the late 90s and early 00s. For those that invested heavily, there is an increased growth in EA projects looking to detangle the enterprise from these investments in favour of the new COTS Services often delivered over the web as SaaS. The simple question being why customise what doesn’t differentiate? Furthermore, by taking a simplified ‘industry standard or accepted’ approach to a business problem you start to create the opportunity to chose who delivers the service over time.


The Cloud, the Enterprise Architect and the CIO

It was great to finally get a chance to speak at the IRM Enterprise Architect Conference Europe the other day. I’ve attended, submitted topics many times and this time around they gave me my big chance! I even got to speak in the keynote room as I’d had the most votes for that slot, so there were plenty of people to talk to which was great!

The presentation focused on four main areas:

  • Talking clouds
    Build a cloud taxonomy and an approach to using it with key stakeholders from business to IT.
  • Business Capabilities
    Discuss how looking above process and implementation at a business’s capabilities enables EAs to engage in different discussions about the business.
  • Future of IT
    The future IT department in terms of new responsibilities and roles and understand the key architectural considerations of entering into a world of hybrid architectures.
  • Lessons from Agile
    Finally, while EAs yearn to be heard by the business, it is too easy to isolate ourselves from the rest of IT along on the journey. We’ll look at key lessons from agile development and how these can be applied at the architectural tier and in so doing learn about “technical debt” and how in the right hands, it is a good thing!

Feedback was pretty good on twitter:

Chris Bird Old friends @taotwit, @Cybersal , @mattdeacon ,@aapthorp among others at #IRMEAC. New – too many to mention.Time with @alecsharp priceless
Chris Bird @mattdeacon me too. Liked your preso a lot, btw
Alec Sharp @mattdeacon Ah – I wondered if I’d got it right. Close, but no cigar. Thanks for taking my question on the cloud as an HR issue.
Alec Sharp  #IRMEAC @mattdeacon Looks at “technical debt” as a good thing when it helps deliver early, clarify needs, control investments, share resp
Alec Sharp  #IRMEAC @mattdeacon IT will physically contract therefore IT’s boundaries will expand.
Alec Sharp  #IRMEAC @mattdeacon So, it’s not the low-level utility stuff that is likely to move to the cloud, it’s the critical bits.
Alec Sharp  #IRMEAC @mattdeacon Nice bit of counterintuitive info – the more strategic the function, the more likely SaaS will be employed.
Kristof Dierckxsens  @alecsharp #IRMEAC @mattdeacon – What business does (its capabilities) is static – it’s the *how* (processes) that is more dynamic. << true
Alec Sharp  #IRMEAC @mattdeacon Some very good material here on what a capability is, and how it can be characterised in terms such as value.
Alec Sharp   #IRMEAC @mattdeacon – What a business does (its capabilities) is fairly static – it’s the *how* (processes) that is more dynamic.
Alec Sharp  #IRMEAC @mattdeacon Arrgghhh – my tweets about this promising presentation were lost because of a Network Timeout Error. A cloud problem?

So here’s the deck if you’re interested – thoughts as always welcomed:)!

Speaking at EAC Europe tomorrow

Looking forward to speaking at the Enterprise Architect Conference Europe tomorrow in London. Talking about implications of cloud – what else ;)! But also throwing in some tips from Agile too!

Service Centricity

There is a shift happening but it’s been happening longer than many think and its effects may be larger and more significant than we at first expect.

This shift is business driven, as it strives to maintain its innovation edge ensuring that costs are reduced and ensuring it moves to a more lean and efficient model of operation. In truth , we can look back as far as Adam Smith and the division of labour in the 1700s to the origins of this shift and may therefore suggest that the shift is not really new at all. However, the old model and granularity of operations is under major pressure to change once more.

The shift is to one of Service Centricity and although it has been a long time in coming, the advancement in technology and the role of technology as an agent of change and business differentiator in the modern enterprise is rapidly moving us over the tipping point that causes us to re-evaluate the very foundations of “the enterprise” itself.

Hype and hot air!

Cloud is undoubtedly centre stage and the hype is reaching tsunami proportions. Many blame this on the vendors but while the idea has got the technology market excited I think this view deflects us from the real question of cause and effect. Is this an example of technology creating new markets or is it an example of technology reflecting the needs of existing markets? In essence it’s an element of both and as is often the case with new innovations it often causes us to view things differently which results in new ways of solving existing problems.

As such ‘Cloud’ in all its forms is set to a have a significant impact on the enterprise, but it is in delivering direct quantifiable business value that the cloud will deliver its most significant impact. If you think of cloud simply in terms of utility computing or dynamic provisioning of infrastructure then the point of cloud and its significance may pass you by.

I want your solution not your software

The impact of cloud is already visible across the enterprise today in the endemic use of Software as a Service (SaaS) Solutions at least somewhere within the enterprise and it is not that this is something that just happened overnight but has certainly been growing over the past 5 or so years. Many Independent Software Vendors (ISVs) cite that customers want but don’t want to burden IT. They ask if they can prototype the solution hosted by the ISV. The solution seldom ends up coming in house. One ISV recalled a customer saying “we want your solution, but we don’t want your software”. The issue for the ISV is that they are now a hoster and need infrastructure expertise, something they have little in-house skill or budget for. Cloud platforms provide these ISVs with an instant, scalable, reliable solution that they themselves would be hard press to achieve and all at a price that scales with their customers.

The dilemma for Internal IT

For Internal IT this creates a problem if done without their knowledge and its incumbent on IT to be proactive in the establishment of external services rather than representing a obstacle which will only be seen as holding the enterprise back. Taking a lead on these fronts is where IT can achieve the best results, providing vendor selection services to streamline the process and actually becoming the agent of adoption promoting well governed use of services across the enterprise.

Rise of multi-sourcing

The impact of cloud is not exclusive to the ISV or Internal IT but to the very fabric of the traditional software supply chain.Today we readily recognise different roles and functions across the supply chain from ISVs, to System Integrators, Outsourcers to Offshoring but what happens to these discrete entities if we look to the deliver of services? They start to converge. If I deliver a solution as a service, then I am no longer disengaged from the customer post-sale, we’re tied together for as long as the provision the service. What of the System Integrator? The solution is no longer deployed on-premises for running in-house, but as with the ISV, it’s hosted by the SI. The relationship between consumer and supplier are tied together once more. And so it goes, an ISV looks more like an SI, an SI more like and ISV, they both are outsourcing because the solution never in-sourced. The enterprise is now no longer exclusively in-house or outsourced, but multi-sourced.

Collaborative/Shared Innovation

similar to open innovation through multi-sourcing the enterprise is seeking partnerships for the provision of business services. This is fine at the commodity level, but as one delves ever closer to core business services the need to tailor and personalise and more importantly, protect IP becomes more significant. However, here too we see developments in service centricity through the development of innovation centres and incubation hubs often maintained by the enterprise itself. This works well but is limited and often expensive to nurture. By collaborating and sharing innovation more pro-actively with service partners it is possible to drive the innovation faster, cheaper through a shared risk/reward model, thus allowing the innovation the opportunity to expand much further beyond the horizons of the parental organisation as the market opportunity evolves.

Integration as a Service


Today we see see the commodity functions of email and collaboration moving out of the enterprise at a dramatic rate. So what next? Mission critical? Core Functions? Maybe not, but by using cloud to extend or embellish these core functions and to connect and integrate with other service providers, business partners or different parts of the enterprise itself then adoption rates here too can be seen to be gathering pace.

The Business as a Service

When I think of this I am always reminded of Amazon’s Mechanical Turk which takes it’s name from an 18th century chess-playing machine that really hid a human chess master inside.This service provides human workflow services that are accessible through a web service (programmable API). If we look at the business as a set of collaborating capabilities then we can effectively define boundaries or APIs and describe what the capability does but not how it does it or who does it. This is service centricity business style. This is the “composite” business or business as a service.

Small is the new big.

None of this for you? Maybe not. But if it’s not for you then be sure it certainly is for someone and as is the way with disruptive innovation, it may come from a source much nearer home than you think. Take one person with good business domain knowledge,  take one developer, take one cloud and what have you got? A business service that is reliable, scalable and available to a world of consumers at a cost that they can all afford.

The Economics of Cloud Integration

I met up with an integration partner the other day and we talked about how cloud could impact their business. He talked about the typical integration project and that it was quite an investment for the customer.

  • Phase 1: Partner builds out integration solution
  • Phase 2: Initiates technical training for customer IT staff
  • Phase 3: Handover management of integration solution
  • Phase 4: Work on joint second wave project
  • Phase 5: Full handover to customer.

In talking about cloud he suggested that it wouldn’t make sense when dealing with a simple EAI scenario of integrating two in-house systems. Why would you consider going outside the firewall to achieve this integration? The partner thought not, but I’m not so sure and suggested that if I offered to run the service for say a few £100 a month with strong guarantees I think the customer would have to think about it seriously. This is after all what the AppFabric in Azure is all about! At the time my mind was simply focused on the operational savings, but in going back to the phased delivery model above, there were also skilling and software licensing costs that needed to be factored into the equation.

So let’s imagine the cost of this simple EAI project from a traditional integration project and a cloud service perspective.

Traditional integration   Cloud Service  
Initial Development Project £45,000 Initial Development Project (50% up front remainder deferred over 5 years) £22,500
Training (x 3 people) £15,000   £0
Assisted Development Project £25,000 50% up front with remained deferred over 5 years £12,500
Software costs £25,000   £0
Support & Maintenance/year (10%) £7,000 Development cost recovery (divided 5 years)
Support & Maintenance/year (10%)
Hosting support (10%)
+ £7,000
+ £7,000
Internal operations expenditure (FTE @ 50%) £30,000   £0
Total Cost over 5 years £320,000   £140,000

In this scenario the customer saves around 60% costs over 5 years – why wouldn’t the customer be interested? Also, the partner earns an additional £35k over the 5 years too, plus they still retain a close relationship and ongoing revenue.

Ok, so the model is overly simplified and probably missing a 1,001 things but offers an indication of a more balanced economic and innovation based model that aligns much better to the customer’s needs than the old model of software delivery and maintenance. Besides the obvious savings what is far more powerful is the effect this has on cashflow over time …


Not only are the initial cost reduced and spread over the 5 years, but so too are the ongoing maintenance/management costs.

Time for a change …

Or new begining!

Having blogged first on Microsoft’s TechNet blog and then MSDN I thought it would be fun to go independent and see if I can attract an audience above the automatic search engines and web trawlers!

So if you’re into the IT architect space maybe give this blog a whirl!

don’t believe me then take a look at my previous blogs at

Categories: General Tags: , ,