Thinking Globally - Where Should a CIO Begin?

Share on LinkedIn1Tweet about this on TwitterShare on Google+0Share on Facebook0

A few weeks ago, I participated in a workshop with a global financial services CIO and a few of his leadership team to discuss what the coming 3-5 years would mean for the IT function and its role in the business.

The most important driver of this kind of conversation has to be where the business is headed.  To serve as a proxy for this, we spent a good deal of time talking about questions facing the business:

  1. Geographic expansion?
  2. Product range diversification?
  3. Distribution channel expansion?
  4. New target customer segments?
  5. M&A activity?

These kinds of discussions often tailspin into how to organize the IT function, with the concept of a “federated” IT model taking center stage.

However, my advice is to think about the business operating model first, and let that drive the IT discussions.  The framework we used for this conversation was one developed by Jeanne Ross at MIT’s Center for Information Systems Research.  Along with her colleagues Peter Weill and David Robertson in their excellent book Enterprise Architecture as Strategy,  she describes a business operating model as:

The desired level of business process integration and business process standardization for delivering goods and services to customers.

By looking at each business (or business unit) along these two dimensions, with a good idea of the answers to the five questions above, you can start to develop characteristics of both the business design and technology systems for each of the four resulting quadrants.  MIT’s version looks like this:

Business Operating Model - Ross, Weill, Robertson (MIT)

Business Operating Model - Ross, Weill, Robertson (MIT)

Does Your Technology Match Your Business Model?

When looking at a chart like this, the obvious question is “where are we?”  When answering this, it’s important to first look at where the business is today.  For example, a wireless telecommunication company I spent many years with was very entrepreneurial and based its business on creating scores of local offices across North America.  To do this, they created a cookie cutter business and technology model that could be applied in any market - high business process standardization but low business process integration, or a “Replication” operating model.  Their IT organization was basically a centralized IT factory that developed and maintained the core cookie cutters and support systems for the local businesses to operate.

In the case of the financial services organization, things are a bit different.  Because they have been growing through acquisition of late, they have a collection of businesses that do the same things, but in much different ways - low business process standardization but a high(er) level of process coordination, or a “Coordination” operating model.

Coordination is where they are today, but not where they want to be.  For the vast majority of their business, they hope to move to the Unification model to bring their core products into a single set of business processes.  Even across Europe, the UK and North America, their products and processes share much more than they deviate.

Addressing the Federated IT Model

With an idea of the target operating model in mind, the target IT operating model is next.  Here are some thoughts on how the two relate.

Operating Model IT Strategy, Planning & Management Application Development & Maintenance Infrastructure
Diversification

Distributed, with some basic infrastructure and vendor planning centrally

Distributed

Centralized network & office infrastructure utility

Replication

Centralized infrastructure planning; Shared planning for core app suite

Core Business Apps - centralized; Local apps and innvoations - distributed

Provisioning of “replicated” infrastructure is centralized

Coordination

Centralized infrastructure, data and integration planning & management; Business specific planning is distributed

Centralized for Enterprise apps,  data and middleware systems; Enterprise apps

Centralized infrastructure provisioning for shared apps, data and middleware

Unification

Centralized

Centralized

Centralized

Two final thoughts. First, this kind of thinking can be applied to companies with several distinct business units; a global footprint is not required. Second, this is a much more complicated topic than is outlined here - especially when culture and language enters the discussion.  However, I think its an important starting point to start with the business direction, think about the resulting business operating models and only then, start working with the resulting target IT operating designs.  Love to hear your thoughts.

Share on LinkedIn1Tweet about this on TwitterShare on Google+0Share on Facebook0
  • Pingback: Tweets that mention Thinking Globally – Where Should a CIO Begin? — CIO Dashboard -- Topsy.com()

  • Chris,

    Good advice! So often departments in organizations get lost in the ‘details’ and forget the real reason they exist - to help the company make money. If your department is not doing it directly, make sure you are supporting one that is. As organizations grow, this point is often forgotten.

    Good stuff!

    Glenn Whitfield

  • Chris,

    Good advice! So often departments in organizations get lost in the ‘details’ and forget the real reason they exist - to help the company make money. If your department is not doing it directly, make sure you are supporting one that is. As organizations grow, this point is often forgotten.

    Good stuff!

    Glenn Whitfield

  • John O’Gorman

    Hi Chris;

    I’m working on an enterprise-level ‘theory of everything’ model that has been very successful adressing some of the issues you address here, and while I think the representation (especially the matrix) is OK on one level it misses a huge short-coming in most current organizations.

    First off, getting the business to hammer out what it wants before engaging IT is absolutely the right way to go. Unfortunately, I think the matrix used to describe IT’s response has a built-in bias by suggesting that ‘centralized’ is the ultimate answer.

    Like many emerging issues in Enterprise Architecture, this one is related to one very elementary issue in computer-based (as opposed to human-based) information processing: identity.

    While the pendulum has historically swung from centralized to distributed (or diversified) the fundamental fact that the word ‘Service’ for example looks identical to a computer whether it refers to church, tennis, customer or animal husbandry. This rigidity in digital systems - and the countervailing requirement for the business to be diametrically opposed to that limitation - is amplified a thousand-fold when companies attempt to rigidly control processes and applications.

    This is especially and almost tragically evident when companies go global, or acquire or merge with another ‘centrally controlled’ entity. Each entity has its own rules and/or architecture and is unwilling to let it be ”centralized’ by another.

    The answer is to centralize the abstraction that glues all business and IT together then allow diversification based on the rules that created the abstraction in the first place.

  • John O’Gorman

    Hi Chris;

    I’m working on an enterprise-level ‘theory of everything’ model that has been very successful adressing some of the issues you address here, and while I think the representation (especially the matrix) is OK on one level it misses a huge short-coming in most current organizations.

    First off, getting the business to hammer out what it wants before engaging IT is absolutely the right way to go. Unfortunately, I think the matrix used to describe IT’s response has a built-in bias by suggesting that ‘centralized’ is the ultimate answer.

    Like many emerging issues in Enterprise Architecture, this one is related to one very elementary issue in computer-based (as opposed to human-based) information processing: identity.

    While the pendulum has historically swung from centralized to distributed (or diversified) the fundamental fact that the word ‘Service’ for example looks identical to a computer whether it refers to church, tennis, customer or animal husbandry. This rigidity in digital systems - and the countervailing requirement for the business to be diametrically opposed to that limitation - is amplified a thousand-fold when companies attempt to rigidly control processes and applications.

    This is especially and almost tragically evident when companies go global, or acquire or merge with another ‘centrally controlled’ entity. Each entity has its own rules and/or architecture and is unwilling to let it be ”centralized’ by another.

    The answer is to centralize the abstraction that glues all business and IT together then allow diversification based on the rules that created the abstraction in the first place.

  • Chris,

    Another interesting topic on the CIO’s doorstep. We are faced with this connundrum as a SSC on where to draw the line in supporting the business versus helping them truly transform. The concepts you discuss are more complicated when trying to serve the needs of 100’s of brand customers. We are tackling this piece by piece with an eye on business drivers in a Coordinated fashion. But as I stated, the real challenge is drawing a line on what services the SSC should provide vs. IT in the businesses since our mission is to be more than a back office shop.

  • Chris,

    Another interesting topic on the CIO’s doorstep. We are faced with this connundrum as a SSC on where to draw the line in supporting the business versus helping them truly transform. The concepts you discuss are more complicated when trying to serve the needs of 100’s of brand customers. We are tackling this piece by piece with an eye on business drivers in a Coordinated fashion. But as I stated, the real challenge is drawing a line on what services the SSC should provide vs. IT in the businesses since our mission is to be more than a back office shop.

  • Christian Brandt

    I think it is crucial to have a business centric approach. The market is constantly changing with an increasing speed. The winners will be those who can adopt the market and fulfill their customers changing needs quicker than their competitors. To be able to adopt the business will need sufficient support. Most often the business development flexibility is restricted by current IT-systems and solutions. To be able to compete these restrictions must be eliminated. Today the restricted situation goes for most companies but some companies has found a way to be even more flexible and they will be hard to compete with in the future.

  • Christian Brandt

    I think it is crucial to have a business centric approach. The market is constantly changing with an increasing speed. The winners will be those who can adopt the market and fulfill their customers changing needs quicker than their competitors. To be able to adopt the business will need sufficient support. Most often the business development flexibility is restricted by current IT-systems and solutions. To be able to compete these restrictions must be eliminated. Today the restricted situation goes for most companies but some companies has found a way to be even more flexible and they will be hard to compete with in the future.

  • Chris Curran

    Thanks for the thoughtful comments.

    John, to your point about the table implying that “centralized” is the right answer, I have two thoughts. First, if you buy MIT’s model, only some organizations would be a fit for the Unification model. In one study they did, only 57% of the firms responded that Unification was their target. Second, based on my experience, a firm’s culture and bureaucracy often gets in the way of a “pure” centralized service or function, so this is really meant as a discussion model rather than one that is perfect in practice.

    Thanks again,

    -Chris

  • Thank you for sharing this post with me… 😉

  • Thank you for sharing this post with me… 😉

  • Shayne Edmondson

    I agree with your assertion that Enterprise Architecture as Strategy is a great book. While the 4 operating model grids is a great reference; it is not appropriate to apply a single operating model for a global organization. The financial services sector for example has local legislation/regulation which prevents unification of certain functions (examples include: BASEL II and SOX). That said, there is benefit from applying these models to individual business units (examples include: Payments, Finance, HR, and CRM) this enables not only local legislation/regulation to be considered but also local market drivers; these are highly competitive local markets (e.g. mobile banking in South America) are probably not best suited to unification whereas Finance, Procurement and HR may be ideal candidates. Although you do allude to this in your article, ‘business (or business unit)’ I felt it deserved another mention having seen broad-brush approach in the past creating a world of pain for delivery teams.

  • Shayne Edmondson

    I agree with your assertion that Enterprise Architecture as Strategy is a great book. While the 4 operating model grids is a great reference; it is not appropriate to apply a single operating model for a global organization. The financial services sector for example has local legislation/regulation which prevents unification of certain functions (examples include: BASEL II and SOX). That said, there is benefit from applying these models to individual business units (examples include: Payments, Finance, HR, and CRM) this enables not only local legislation/regulation to be considered but also local market drivers; these are highly competitive local markets (e.g. mobile banking in South America) are probably not best suited to unification whereas Finance, Procurement and HR may be ideal candidates. Although you do allude to this in your article, ‘business (or business unit)’ I felt it deserved another mention having seen broad-brush approach in the past creating a world of pain for delivery teams.