The accurate assessment of the thermal conductivity of concretes is an important part of building design in terms of thermal efficiency and thermal performance of materials at various temperatures. We present an experimental assessment of the thermal conductivity of five thermally insulated concrete specimens made using lightweight aggregates and glass bubbles in place of normal aggregates. Four different measurement methods are used to assess the reliability of the thermal data and to evaluate the effects of the various sensor types. The concrete specimens are also assessed at every 100 °C during heating to ~800 °C. Normal concrete is shown to have a thermal conductivity of ~2.25 W m(-1) K(-1). The surrogate aggregates effectively reduce the conductivity to ~1.25 W m(-1) K(-1) at room temperature. The aggregate size is shown not to affect thermal conduction: fine and coarse aggregates each lead to similar results. Surface contact methods of assessment tend to underestimate thermal conductivity, presumably owing to high thermal resistance between the transducers and the specimens. Thermogravimetric analysis shows that the stages of mass loss of the cement paste correspond to the evolution of thermal conductivity upon heating.