Intumescent coatings (also called reactive coatings) are widely used to protect structural steel from fire. These thin coatings swell on heating to form a highly insulating char, protecting steel members and preventing them from reaching critical temperatures that could cause them to fail. As is the case for most structural materials and assemblies, intumescent coatings for use in buildings are typically developed and certified solely according to the standard cellulosic fire resistance test by exposure within a fire testing furnace. Reliance on furnace testing is expensive, non-representative of realistic fire conditions, and insufficiently versatile to gather detailed performance information on the response of reactive coatings under the full range of design fires which might be considered during a rational, performance-based design assessment. This paper presents a novel testing methodology for studying the performance of reactive coatings when subjected to non-standard heating regimes. The new approach is calibrated and validated using furnace test data, and is shown to offer considerable advantages over furnace testing in terms of reliability, repeatability, versatility, speed and cost. An investigation is then presented to study the effective variable thermal conductivity of a commercially available reactive coating when subjected to various timehistories of heat flux. It is shown that the heating rate and dry film thickness of the coating do not drastically affect the development of effective thermal conductivity with substrate temperature, leading to a proposal for a simplified method for specifying coating requirements and/or performing heat transfer design calculations when designing to non-standard heating regimes.