The first question, of course, is “output, outcome and service- is there a difference?” – and at first glance it may appear that there is not. Measuring output is the less complex of the two. An agreed scope for sizing and measuring the work is arrived at, and in the IT world this is often achieved by way of a function point count or similar estimating method. Once the size of the job is agreed, the vendor undertakes to perform a fixed measure of output for a fixed cost – even, in extreme cases, with an element of adjustment so that the customer benefits if the work is less than estimated (delivery at a lower price) but the supplier is penalised if the estimate is inaccurate (the price remains as per the original estimate). Such an arrangement can work exceedingly well to mutual benefit, but requires close attention to detail from the customer's point of view. Measuring outcomes is more complex, and requires that the customer think at a higher level. No longer concerned with how – that's the supplier's prerogative – the customer must be able to clearly express what is to be achieved and by when. Outcomes are most often expressed in business terms or impacts (e.g. improve productivity by X, or reduce cost by Y), and are the most difficult to measure. A service contract falls somewhere bewteen the two, and covers the provision of a service to a defined service level agreement.