The four basic types of business structures

In a tiered structure the activities in any one tier are directly connected to those of other tiers. In a managerial context this may be a branch structure at the lowest level, with a regional structure above it topped off by a head office and ultimately a holding company. In this scenario, a deployment of technology related to financial reporting for example would have to take account of a regulatory issue such as Sarbanes-Oxley, in a completely different way than in a layered, siloed or granular structure.

A technology deployment that leverages say, customer relationship management (CRM), would have to take account of the qualities of the layers as well as the qualities of the connections between them to be successful.

In a layered environment there are either no connections between the layers or, if there are, the connections are very weak or of no real business consequence. At its most obvious, this occurs with the two top layers of retail versus wholesale banking but there are many more examples, the most important example being market sector. The challenge for technology managers in these environments is that typically value is derived from technology when there are connections between things. In a layered environment, value is usually restricted to within the layer and the value is either not recognisable in the other layers or of no use.

So, for example hedge fund calculation and statistical analysis engines are invaluable within the hedge fund layer (market sector), but of virtually no interest or use elsewhere in the industry. This example highlights one of the general dangers in business, that of being blind-sided by the restricted use of any one of these models. I made a presumption in statin gthat there was no use for statistical engines used by hedge fund managers anywhere else in the industry, that is they are layer-specific technologies that must leverage their value from a very narrow usage base. The reality is that this may not be the case. Management, while deciding what technology to use, must be open minded enough to categorise what the available technology can do without restricting its freedom of movement.

From a technology management viewpoint this case study is important because it reminds us that each of the morphology elements has drawbacks and advantages. We can't assume that one model will work on its own, and in fact, doing so will create a very unstable structure that might work for a while but will eventually create risk. I use this anecdote as a touch stone whenever I'm discussing technology management. There are many existing technologies already deployed in financial services. We often have no need to reinvent the wheel. Often we fail to consider these technologies because they are in different layers, tiers, siloes or granules where we can't easily access the ideas let alone the knowledge. One final point in this article, for managers: it may not just be that the technology exists - even where technology does exist and someone identifies it. One of the most dangerous things about technologists (as opposed to technology managers) is that they can often fall into the trap of deploying ‘new' technology simply because its new, rather than because it's the right technology to deploy.

I see this effect gathering great pace as the technology managers of the future are coming up (certainly in western cultures if not in middle or eastern ones) in environments based on ‘instant gratification' and the need for a bigger better ‘bang for the buck'. There are occasions where an existing technology that has not previously been adopted in a particular place, is a good solution. We drop what works for what is new and shiny, far too often.

In a siloed environment, the emphasis is on the vertical and not the horizontal. This is most widely seen in departmental structures within financial services. Trade, Clearing, Settlements and Payments, Back Office Processing are all examples in wholesale financial services, of a siloed environment. There are connections between these, more so in recent years, but historically, each of these activities has been a stand-alone or one with weak interactions. So, while there are experts in payments, experts in back office processing and experts in trading and each one is dependent to a certain extent on all the others for the whole to work, it is rare to find anyone, at the managerial level, with cross-silo experience or knowledge. It is notable, for example, that most of the board of the Society for Worldwide Interbank Financial Telecommunications (SWIFT) has its experience base predominantly in the payments area despite the fact that its current focus is on other areas of financial connectivity such as corporate actions or back office processing.

This will clearly change over time, but this does highlight one of the dangers inherent in managing technology. If senior management experience and focus are not closely aligned with the changes in the market and business, it can lead to projects which should be delivered but which aren't because management either has no understanding of the need or imperative or has no interest. Equally, it can result in projects which don't reflect market need. At this level, perfect alignment is often not possible. The alignment always lags behind the need and it is one of my benchmarks to calculate the differential between the market need and the alignment to the market. This tells me, at a glance, the likelihood that the business is moving in the right direction and that the management is at the right stage of development to both understand and implement any given technology for the benefit of the business.

So, you can see that management of technology in financial services is not just about who puts the specifications together and what technology is deployed. It is affected fundamentally at board level where the interaction of the environment (the market) and the skill set (the board) come together to create technology imperatives.

In a granular environment there are no vertical or horizontal connections between elements. This can often occur in wholesale banking at the brand level. I often come across businesses which share the same brand name, that is, I think I'm talking to someone in Firm A. Firm A is actually a business unit of Institution A, so they share the same name. It is dangerous to assume however, in a granular structure, that any one element is either connected to,or even knows of the existence of, any of the other elements. I've found this on a number of occasions. Discussions with Firm A reveals that there are Firms B through D, but, even though they share the same name, none has anything to do with any of the others and often don't even know they exist. The most common manifestation of this is geography-based. For many medium sized financial firms the cost of managing a connection between an office in Sydney and New York is just too great, so the Sydney office is given an initial directive and from then on responds directly to its customer base within its own region or market with little or no connection back to the parent except at a financial reporting level.

Clearly this has impacts on management of technology where a typical deployment would want to be or even be mandated to deliver some kind of connectivity at one of the three levels. As I said before, we must not assume that any one of these models represents the totality, for example, of any one organisation, department or market. These are tools to look at the environment in which technology is going to be managed and it is highly likely, in my experience, that when a deployment is being considered, more than one of these models will apply at any one given time to parts of the management discussions. I mentioned earlier that, tiers, levels, siloes and granules operate in three ways - physical, technological and managerial. Do not make the mistake of believing that this article relates only to the second of these three. At the point of developing a technology strategy, the issue for business managers is to consider the morphology of their business as well as their environment.

A typical wholesale provider for example may well have some parts of its business in a tiered structure for example, it may have relationship management based in branches so that it can be ‘close to its customers'. It may have parts of its business and branch structure so geographically remote that it essentially makes them independent in a layer - some parts of its operations, often the back office specialist functions, siloed so that it can bring together its expertise in an easily manageable, centralised way.

Finally, it may separate business units in a granular way, for example by market activity, direct custody,wealth management, alternative investments and prime brokerage and so on where each of the units has no real interaction with the others. In this scenario technology managers must be able to model their businesses from a technology strategy viewpoint, in a way that leverages the most value irrespective of the morphology of the business.

The four basic types of business structures

Legal Disclaimer

Related Articles