In any walk of life, great reliance is placed on accurate metrics.  Metrics bring the ability to decide on a course of action before we commit to it; metrics chart progress and metrics form a guide if remedial action is needed.  Metrics, however, can only deliver their true worth if the right value is being measured in the first instance and in the ever changing IT service arena, this can ask some interesting questions of a business.

Historically, service organisations' remuneration has been decided on a quantitative basis - the greater the number of service calls, the greater the charge.  But recent years have seen a shift in the way companies consider their IT and consequently their IT service requirements.  With the reliance that companies place on their IT infrastructure and its importance to the functioning of the business, of greater significance is the length of time that that investment in IT is available to the business  – uptime or availability.  That has changed the service equation in that the fewer number of service calls that need to be made provides for greater value to the clients' business.  What this shift has meant is that service organisations have had to evolve alongside their customers' requirements.

‘There has been a significant shift in the role of the service provider,’ acknowledges George Williams, head of communications at business availability specialist and HP subsidiary, Synstar.  ‘Customers no longer want a reactive service organisation but a partner that delivers a real value add to that customer's business.  Customers want their systems working all the time - they want business availability,’ Williams insists.  ‘This means that instead of looking at responding to hardware failures, service providers are now looking at IT systems and the entire infrastructure, because failure in any one element can result in the system becoming unavailable and impacting the business.’

The service level agreement remains the benchmark against which service delivery is measures, but establishing the service level depends very much on the individual customer.  The value of an IT infrastructure is not the same for every organisation and so the service level, and the metric used to verify it, must be established on a case-by-case basis as Rob Lyons, head of business development for NCR across the Middle East, Eastern Europe and Africa explains.  ‘The shift towards availability has led to customers wanting to focus on individual devices not stepping out of line but determining the level of support that is needed must be undertaken by a discussion with the client and tailoring the support to that customer's business needs,’ Lyons insists.  ‘From time to time, it is not sufficient to determine the levels of support required on a device by device basis because an individual server may be more important to a business between 2.00 and 4.00 in the afternoon, but will not have the same impact on the business if it fails between 2.00 and 4.00 in the morning.  The question that needs to be answered is 'what level of support does a customer need?' and the answer to that question is reached through a dialogue with the customer.’

This occasional obsessive focus on availability and the magical 99.999 per cent uptime figure has meant that companies blindly seeking to achieve such goals, without stopping to consider which items of equipment matter most, have had to pay for the privilege in increased support costs.  This may be leading companies to be increasingly selective about service as Keith Moore, Getronics' UK service centre manager comments.  ‘In my opinion, people are moving away from thinking in terms of availability figures back towards response-based SLAs and this is primarily because of cost considerations as it is an expensive option keeping machines and engineers on stand by to meet such targets,’ he says.  ‘Whether customers are looking to measure SLAs based on response or availability, the key ethos is to work with the customer to deliver and develop the service that the customer needs and wants.  Too many service organisations have been guilty of selling the service that they provide and not necessarily tailoring the service adequately to the customer - of seeing themselves as a service provider rather than a service partner - and partnership is the key to a successful, long term relationship,’ Moore insisted, revealing that a flexible approach was needed to addressing SLA requirements.

‘Obviously there is a greater demand for comprehensive cover for high end products but on items such as desktop machines, the use of hot desking creates a flexibility in the system.  Problems can occur with hot desk machines since faults on those units may not be reported as the machine does not belong to any one individual and it is only when all the spare hot desk machines have failed that the faults are finally reported,’ he says.  ‘We are also seeing with certain customers the introduction of variable SLAs, where a specific machine must be available at a certain time of the day.  This means that the time we have available to fix that machine depends very much on when the fault is reported,’ Moore adds.

Moore's comment about problems being caused by operators not reporting faults on redundant equipment perhaps highlights the need for a tool to circumvent human behaviour.  Technology within hardware itself means that failures in equipment can be automatically and instantly reported, but a mechanism must exist to take advantage of such capabilities as Rob Lyons explains.  ‘Advances in the technology within the hardware itself have meant that many devices have the ability to issue alarms but to reap the maximum benefit from that depends on how the infrastructure is set up to monitor those alarms and how the service is set up to respond,’ Lyons points out.  ‘By having an effective solution in place, response times can be dramatically reduced, improving availability.  But by also looking at non-critical events and through the more intelligent scheduling of engineer visits, to reduce travelling times for example, will also deliver service cost reductions,’ he adds.

A metric that is often quoted in any number of service-based industries is customer satisfaction - or C-SAT as some practitioners have taken to calling it.  But with customer satisfaction based upon subjective customer responses to questions rather than empirical data, is it in danger of becoming a meaningless statistic rather than a useful data tool?

‘The true value of customer satisfaction depends very much on how that information is used,’ insists George Williams.  ‘It is useful to know levels of customer satisfaction on a day by day basis but by continually asking to assess these levels - after every service event for example - then it is quite possible to create upset and in fact lower customer satisfaction, regardless of service performance,’ he reveals.
‘We have developed a customer satisfaction survey that asks 10 questions on a monthly basis that gives customers the opportunity to show their levels of satisfaction with the service that we provide.  There is only one question regarding the SLA, the rest of the questions focus on issues such as the performance and ability of the service engineers.  By combining this data from over 100 customers, we get a true view of our performance and are able to identify where improvements can be made.  This means that customer satisfaction has proved to be an extremely useful measure for ourselves to use internally,’ Williams adds.

'Any measure of service levels will be an average figure and behind that figure will be a graph consisting of peaks and troughs and each trough will be a low point for customer satisfaction,’ explains Andy Powell, EMEA marketing director for NCR Worldwide Customer Services.  ‘Through root cause analysis and service improvements, a priority is to smooth out the SLA curve and so ultimately raise levels of customer satisfaction.  We will review service levels on a regular basis and with the information that we possess, we can make recommendations to customers that will lead to service improvements.  There will be some events that impact business availability that are beyond the control of the service provider and lie only within the customer's influence but it is important to have a helicopter view and to be able to identify where service improvements can be made,’ Powell says.

‘Customer satisfaction can be an extremely limited metric to apply and we do not use it,’ states Keith Moore.  ‘There are a multitude of things that can cause dissatisfaction that are wholly unconnected with our performance and customer satisfaction reports do not necessarily highlight trends or identify issues with engineers.  Of far greater value to us is looking towards call escalation and complaint figures,’ Moore insists.  ‘Getronics has in place a system based on ITIL (IT Infrastructure Library) and Six Sigma methodology that ensures that any breach of a SLA is investigated and corrective action is put in place so that our service is constantly developed.  An example of this is working closely with customer help desks and correcting scripts on a monthly basis to ensure that the information received from end users reporting faults is sufficiently comprehensive to ensure a first time fix.  Taking an extra five minutes at the call reporting stage, gathering the pertinent information, can make an enormous difference to ensuring that a timely, first time fix is achieved,’ Moore says.

While effective dialogue between the service organisation and the customer is crucial in establishing the key performance indicators that are measured that most impact an organisation's ability to operate, any discussion of service levels and service metrics now serves to highlight the key change that has occurred in the relationship between a company and its service vendor.  Remedial action has become almost incidental to the service partners' main role of ensuring that the maximum business benefit is derived from a companies' IT investment.  Service providers are now not solely judged by customers on how quickly they can respond to incidents, but on the value they add to that customer's business by creating a stable IT environment that meets today's business needs and can also adapt to accommodate its future requirements.