Typically, as trenches are opened one first comes to a layer of large, often knapped, flints. Beneath this is a layer of roof tile. This roof tile is carefully removed and stored by context. Each piece of CBM is then classified as, tegula (flat tile), imbrex (convex tile), lydion (flat, rectangular brick),  or unidentifiable. The weight for each class is recorded against the context where it was found. Generally, at the end of each digging season, it is reinterred on the site.

At High Wood we are particularly lucky in being supported by Peter Warry FSA who is the leading expert in Roman CBM. At the end of each season he comes and examines what has been found and stored by context. Evidently, over the Roman period the size of tegula increased. This reduced the weight of the roof so needed less structure to support it. Consequently the cost of roofs reduced over the Roman period. This allows Peter to tell us that the majority of the tiles at High Wood are from the 4^th century. This contrasts with other finds that suggest the site was in use from the late Iron Age through to the end of the Roman period. This in turn suggest that the roofs may originally have been thatched or shingle.