This paper describes a postal survey of UK Quantity Surveyors to relate human factors, such as experience and personality, to conceptual estimating expertise. Composite variables were derived by factor analysis and examined against estimates of average national prices for several types of building. It is shown that expertise is very much of a project specific nature and does not extend in a simplistic way to projects outside the defined domain and that estimators must exercise of great caution when undertaking work even slightly outside their regular activities. Different building types demand different emphasis and special attention is drawn to the complexity of the project, the degree of services content, and particular sub-market conditions. The easiest projects to estimate appear to be industrial (factories) and residential (houses) with offices being the hardest, probably due to the wider variety of design and quality options in the latter. 'Knowledge and 'care' are identified as the most crucial attributes of good estimators. A few myths are also dispelled. Geographical location, for instance, was found not to be a major issue. Similarly, there was no evidence of any 'X' factor whereby individuals can claim any mystical in-born talent. The indications are that good estimators have exactly the same attributes as good gamblers - they do their research selectively and thoroughly, think carefully and concentrate on what they know best.