Consolidation, Over-consolidation and Simplicity Part 1
Over the last few months I have been asked to develop a number of “Instance Strategies” for customers. The brief is normally around consolidation. These typically involve looking at a customer’s application portfolio and suggesting ways to simplify or consolidate. Often times simplification is defined as a principle but this manifests as consolidation. People seem to think these two are the same things. This has led me to thinking about the different types of consolidation as well as some of the risks.
In this blog post I’m going to talk about the different kinds of consolidation that exist and what you could expect to get out of them. In the next post I’m going to be looking specifically at this idea of equating consolidation with simplification and the some of the problems I’ve encountered when people do this.
The best explanation of consolidation I can think of is to eliminate redundancy. It is the combining of multiple components into a single entity or thing. In IT, there are different types of consolidation, or levels as I like to think of them. Different part of the business or technology domains are being consolidated in each of them.
Business Operating Models: How does the customers’ business choose to operate? Are there standalone or integrated processes between business units? Are there shared processes across business units or individual business unit autonomy? What levels of integration and commonality exist across the different business units? This level often deals with combining different business units together, possibly through an acquisition. A company acquires another business (a competitor perhaps) and wants to merge the two to gain some benefits.
Business Process / Capability: Within a business unit operating model what consolidation and standardisation of processes are possible? Where is the organisation doing the same thing twice? This commonly manifests as shared service centres providing a single capability to multiple business units or groups.
Application: Application consolidation involves reducing the number of production application instances by moving functions from one application into another. Combining or merging applications can result in reduced application management and integration costs as less systems need to be managed and integrated. A classic example of this is ERP, combining Finance, Logistics and Operations into a single ERP consolidates what was previously multiple systems in a single one.
Data / Information Models: This involves consolidation of data models and/or shared master data. I often see this as analytics consolidation. Combining data warehouses or data marts is quite a common problem given how easy it is for data marts to appear and how difficult they are to remove once they appear.
Technology / Infrastructure: What I mean here is shared and consolidated infrastructure. Applications and data reside on a common platform. This could be storage consolidation through combining SANs or similar technologies, server consolidation using virtualisation tools or establishing some kind of shared database, application server / middleware platform.
This list is a hierarchy, cascading down. When business units consolidate, the opportunity exists to consolidate capabilities, applications, technology etc. When applications consolidate, data models converge (in the consolidated application) and also the technology underneath these applications consolidate.
In the above list I see the complexity and benefits go down in each item of the hierarchy. The most difficult to implement and the most beneficial is operating model consolidation. The main reason for this, as I see it, is the linkage to the real world. When businesses consolidate they can implement the same supply chain, with the same trucks, drivers and warehouses etc. By sharing the real world, the IT world can then be easily shared as well. Its easier to implement a single supply chain system for a single supply chain.
As I go down the list the benefits reduce but so does the difficulty to implement. The easiest of this list is infrastructure consolidation – server consolidation using virtualisation and similar technologies are the simplest to implement of all the above but have least value. IT can execute this largely without business change at all. I am still running the same number of applications, the same integrations and processes but I have less physical servers or less data centres.
This is how I see IT Consolidation. It’s a concern for nearly all customers I have interacted with. The larger ones have an inordinately large share of the problem. When we think about how to simplify landscapes and remove complexity it’s an excellent tool. Thinking about it at different levels informs us what level of benefits we are shooting for, but also informs us of the level of organisation involvement needed in each (e.g. Business unit consolidation is not led by IT).
Next post I’m going to look more closely at the link between consolidation of IT and simplification, when are they the same and when are they opposites?