For many C-level executives, the value of cost transparency is self-evident, but it nonetheless faces a huge obstacle to success in the form of the availability, reliability and timeliness of data. After all, logically speaking, the importance of obtaining and analysing the right data grows in direct proportion to the increasing complexity of the organisation’s IT environment, particularly when multiple as-a-service offerings are available.
With this in mind, if cost transparency is to be truly effective, it is vital to have not only a clear understanding of the cost drivers, but also the origin of the cost itself. The simple explanation as to what a cost driver is describes it as either the basis for allocating costs or as the allocation method. Put simply, cost drivers are the elements within an enterprise that create an overhead cost that is weighed against the goods or services provided by the organisation. Coupled with the matter of finding the best cost driver is the challenge created by the availability, reliability and timeliness – or lack thereof – of data.
So, when it comes to the issue of choosing an appropriate driver, what then are the key considerations an enterprise should ponder in its approach?
The first question that should be asked is ‘what data is available?’ It is vital to understand firstly whether the data you are seeking is readily available within your organisation and secondly, whether the source of the data is repeatable, accurate and complete.
A second consideration should be whether the basis of the driver is easy to comprehend and intuitive. After all, when end-consumers are not able to clearly understand how something is allocated, it can be easy for them to lose trust in the cost model. With this in mind, it is advisable to always follow the ‘Keep It Straight and Simple’ (KISS) principle, in order to ensure a foundation of simplicity and understandability.
A final thing to take into account in relation to this is to allow for optionality, giving consumers the ability to remove or substitute the service. Remember that cost allocations and drivers are most powerful when they allow users to opt out or change their consumption behaviour to something more appropriate.
Drive correct behaviour
It is worth noting that driving correct user behaviour needs to be focused on and needs to be looked at on a continuous basis. This is because even those drivers that are considered good at the start can end up causing incorrect behaviour in consumers, or lead them to attempt to ‘play the system’, in order to utilise the allocation model for their own benefit.
An example of an incorrect behaviour driven by a good cost driver is how many companies allocate their IT helpdesk based on tickets or usage, which causes consumers to stop calling the helpdesk, in order to avoid picking up a charge.
A similar issue is raised by those items affected by economies of scale, which may also wrongly affect the underlying behaviour one is seeking to drive. In these cases, early adopters of a new service often end up subsidising future consumers.
To this end, early adopters may not utilise the service as intended, because of this early adopter penalty. In such an instance, the enterprise should lower the rate or use a weighted driver to subsidise the new service.
Lastly, the method through which a business prices its Shared Services can also have a significant impact on the perception and usage of its services. In other words, treating Shared Services as little more than another corporate department may well mean that the services rendered end up being inefficient and costly.
However, when an organisation treats Shared Services as a true business unit, the services provided tend to be more innovative, more effective in meeting the needs of its buyers and more cost competitive in comparison to the external market.
What are the best drivers?
It is generally accepted that one of the best drivers is consumption or usage. If you think about it, the underlying consumption of the product or service is, in fact, what drives the cost associated with that service.
Of course, in such cases, it must be remembered that the cost driver is what causes the cost to allocate by ‘first consumer basis’ - meaning that the company should charge the initial consumer of the service being focused on, rather than billing the end beneficiary.
What makes this driver so effective is that it links consumption behaviour to the cost of a service. When this is combined with detailed granular usage information, such as server host name or desktop serial number, the consumer can easily understand the charge and – more crucially – can be given the option to influence and change the charge.
In those cases where there is an absence of readily available usage-based drivers, the next best option is to utilise whatever data is available on a loose-linked basis. An example here would be to allocate IT Security based on end-user devices, since this is ultimately how all IT Security is consumed.
If this type of driver is also unavailable, the recommendation is to instead use a high-level estimate – such as percentage IT spend per recipient – which can then be refined later, as the specific data around the driver matures.
Keep data up-to-date
Looking at the majority of modern cost models, the question must be asked ‘why do most fail?’ The most common reason is simply because the underlying driver data is either not kept up to date or the logic behind it is unfathomable.
So it is vital to continually enrich or update driver information. When doing so it is also important to strike the right balance as there is always a cost involved. Extracting and generating, or even changing, business processes to gather driver data comes at a price.
Therefore, one should always weigh the cost of resource effort, process change or systems to retrieve the data against the benefit gained. In other words, if there is a saving or benefit from enhancing a driver it must be greater than the cost of undertaking the enhancements.
Finally, when choosing a driver, simplicity is key, while the maintenance effort involved should also not be onerous. To achieve this, one should document the procedure of getting the driver data and set up a business process to ensure the driver remains current and that dependent parties follow the process. If this is not done, the driver and related data will soon become outdated, and outdated allocation logic is not only indefensible, but far more crucially, it could potentially lead to incorrect decision making, distrust and ultimately, the complete death of the costing model.
Blake Davidson, Head of Delivery, Magic Orange