The cost/quality conundrum: Getting the balance right
Published in ComputerWorld UK
Many in the C suite start their forecasting from the wrong place.
The holy grail of most CIOs is how to minimise their running costs without endangering the organisation’s operational efficiency or the quality of customer services. This mandate to cut – expressed as ‘Save x%’ – is passed down to the various departments within an organisation where the great horse trade begins. In order to protect front line services this often means that disproportionate cuts are made in the back office.
That is why it is worth looking at at some of the IT issues that impact the cost/quality balance and how to find strategies that deliver savings, preserve productivity and boost the bottom line.
For the CIO making short term savings is relatively simple: just cut IT services (to both internal and external customers), reduce staff, reduce or don’t renew SLAs and delay capital projects. But in the long-term these decisions may be counterproductive. Because technology now underpins the lion’s share of enterprise functionality, cuts or non-investments made in this area have unforeseen results.
Take, for example, all the IT projects still pending following the 2008 crash. How about that new data security system? With cyber-attacks and corporate data raids on the rise, this is an investment that could be vital in combatting both financial and reputation risk management. And what of the modernisation programme calling for legacy replacements, platform standardisation and cloud migration? These may have seemed a good idea a few years ago. Today they may be critical to survival and yet may remain on the shelf because their long-term impact on the company’s health and profitability is simply not recognised.
The Goldilocks balance
Of course while corporate Britain’s economic future remains in the balance, we must all remain vigilant about its costs … while staying in business. So how do you get the ‘cost-efficiency vs. performance equation right? OK, the relationship between the number of staff and the quality of a company’s internal and customer support may be fairly obvious – although even this is often a process of trial and error involving expensive fires and rehires before the optimum balance is found. But in the world of IT there are other influencing factors that are far less straightforward. One of these is platform complexity.
Not so obvious on the cost/quality balance is the impact of specific factors within the IT environment like the percentage of legacy systems which, while they may be core business applications, are usually difficult to integrate with new applications and expensive to maintain. Another area is process methodologies. An organisation that has, for example, adopted ISO or other industry recognised best practice operating standards is typically more efficient and therefore achieves a better cost/quality ratio.
So how do you work out when and where to cut staff without causing damage? How do you pinpoint what components of the IT infrastructure or the outsourcer’s Service Level Agreement (SLA) are a drag factor on performance? How do you get the optimum efficiency balance between all the factors including staffing, complexity, process maturity and so on? Analysing all these parameters (each of which may have up to 100 attributes) is a bit like one of those perpetual motion pendulums where you tap on one sphere and it has a knock-on effect on all the others. Each IT service – the help desk, application server support, the data centre, networks & communication, printing, storage and all the rest – influences all the others and a change in one will have a knock-on effect across the IT universe.
Working it out virtually
Creating the algorithms needed to accurately forecast outcomes is a complex task – too complex to be done manually in real world conditions. This means that most attempts at achieving optimum value balances are done with a ‘finger in the wind’. Experienced management often gets it right but the trial and error approach can just as often lead to costly staff redundancies and customer service complaints. The remedy lies in getting the right data with which to create ‘what if’ scenarios that can accurately forecast the outcome of changing any given parameter.
The problem is that many in the C suite start their forecasting from the wrong place. If, for example, the IT department has already been weakened by cuts, then further reductions will simply lead to an extended underperformance. Before taking any action a detailed assessment should be made of the current state of the department’s health and how it is affecting the enterprise. Only once an accurate baseline has been established can the process of modelling efficiency, savings and optimisation forecasts be undertaken.
The smart way to forecast staffing, savings, efficiency improvements or change management programmes is not to experiment in the real world but experiment with a whole range of variables first, using virtual modelling tools to create ‘what if’ scenarios to forecast outcomes. Of course, this only works if the data that is fed in is accurate to begin with: ‘garbage-in/garbage-out’. This brings us back to the importance of first creating a baseline measurement of all areas, services and components of the IT operation (whether handled in-house or outsourced). The great advantage of virtual modelling is that it enables decision-makers to assess a whole range of diverse possibilities and options without having to interfere with actual operations and processes.
Creating ‘what if’ scenarios around the issue of the cost/quality balance would involve questions like ‘If we reduce staff by X what can we save and what levels of service quality could we realistically achieve?’ OR “If we reduce our SLA level from premium to standard, what additional inhouse staff would be needed to compensate and what would be the overall savings, if any?’ ‘If we rationalise our IT services by merging with another organisation or region, what are the restructuring costs and will it lead to overall savings and greater efficiencies?’
Having run all the options through the modelling process it may turn out that no further cuts are either necessary or advisable. Or it may be discovered that savings can be achieved in areas not previously considered. Take shared services. Rationalising IT support across various locations could save 20% or more of the collective operational budget. Or conversely the reconfiguration costs may turn out to be more than the return on investment, or savings may not be realised for several years. Either way, virtual modelling enables the outcome to be forecast upfront.
When new investment is needed
Another major consideration surrounding cost is capital expenditure. We talked earlier about IT projects clogging up the pipeline. What happens when they can no longer be avoided as opportunity costs mount? As organisations are forced more with less, there comes a point where the only solution is to bring in new, efficiency-creating technology.
But there is a Catch 22: a company may need new systems but not have the capital. Or do they? Again, virtual modelling can pinpoint how, where and when efficiencies budget can be tightened in one area that can be redeployed to investment. It can also work out the smartest procurement and sourcing options – licencing vs leasing, traditional outsourcers vs cloud-based providers, etc. Most importantly it can provide ROI projections and a business case that will either make sense (or not) to the stakeholders. Best of all, using virtual modelling these projections can be arrived at speedily and at a fraction of the cost it would take to do a feasibility study using a traditional management consultancy.
So back to our original question: How can organisations cut their operational costs without compromising in-house IT support and customer services? The real issue is about going beyond short term, ‘lowest-cost’ thinking to consider the kind of transformational frameworks that will deliver savings, efficiency optimisation, competitive edge and a better bottom line. Is this too big an ask in today’s climate? Perhaps the real question should be: What are the real costs of doing nothing?