This is how to do it. These two monographs from Oxford Archaeology, which I will call Winchester and Southampton for short, are both impressive presentations of a large amount of archaeological and documentary information from recent excavations in two of the most important English towns. Both have appeared with commendable speed.

The full-colour format is now normal for reports. If anything the plans are too colourful, with an immense amount of information packed into the larger ones. Hundreds of individual pits and walls are numbered on the plans, a consequence of the increased use of computerised planning systems on site and GIS analysis thereafter. Please pity the archaeologist now writing up a similar site from the 1980s who is staring at a ziggurat of uncomputerised single-context planning sheets.

There is much to praise. In the Winchester report, the initial site assessments are put on the CD. All periods from prehistory to modern are reported, though the main part is medieval and post-medieval to 1550. Disappointingly, coins and tokens were absent, so the dating framework is broad (though refined by archaeomagnetic dates largely taken from hearths, claimed to be on a scale so far unique in British urban excavations). The regular layout of properties implies a central authority by 900. Local corrections to the published survey of properties in 1148 by Keene are made; the library site was in a peripheral, north-west corner of the Roman and medieval town, but was within Oram's Arbour, the Iron Age enclosure. Previous excavations on the site by Cunliffe are reassessed. Since all unaffected archaeological deposits were to be preserved, the research design ‘removed the need for subjective choices to be made about where to excavate’ (p 21); hence the importance of the post-excavation assessment, which structures the final report. The pottery analysis is particularly fine. The proposed reconstruction of fourteen medieval properties is largely on archaeological grounds, which is then compared to the documentary evidence. This is available from the eleventh century; both projects have historical records which start earlier than the majority in London. After redevelopment of the street in the thirteenth century, this peripheral site in Winchester enters a long period of decline, perhaps representative of the rest of the town, from the late fourteenth century to the nineteenth century.

The Southampton report is of a large site in the French Quarter, in a more central location with one side against the High Street, and it overlaps with earlier excavations of 1953 by Wacher. It is the largest excavation in the town to be reported so far. Overall, the exposition with graphics is clearer than Winchester. Supporting data is available on a website (which works). The pottery (557kg) includes a large post-medieval assemblage which complements previous work on earlier periods of the town's ceramic sequence. Study of the bird bones reveals that the residents of Southampton probably ate the seagulls which scavenged around their houses, and one cesspit contained bones of a great auk, now extinct. One wonders what it tasted like. A French raid of 1338 was a turning point in Southampton's history, and the properties are examined for evidence of it (some, but not much). In comparison with Winchester, there are better standing fragments of buildings, some within later ones, from the thirteenth century, and a treatment of the buildings standing on some of the twenty-three medieval properties until the present. A courtyard house, ‘Polymond's Hall’, is reconstructed through several phases from the late twelfth century to the late eighteenth. Not only are tiled roofs a sign of status in the twelfth century, but it seems that reused Roman tiles in walls were also a prerogative of the rich. There is a variety of archaeological evidence from which richer households are inferred, including perhaps the presence of bread ovens in their houses. The documentary evidence is presented up to the twentieth century (the known owners of one tenement being traced from 1225 to 1940); the occupiers of one property supplied fruit, vegetables and flowers to the Titanic. Archaeological evidence included signs of fire from bombing in the Second World War. Here ‘perhaps more than fulfilling any specific research issue, the investigation has supplied a wealth of historic and archaeological data that anchors the town to a sense of place’ (p 259). Research designs, it seems, are of limited use in urban situations. Or perhaps we should say, adapting a phrase from recent military history, that in deep town sites the research design rarely survives contact with the stratigraphy.

New suggestions about the history of the two towns are made. At Winchester, the grid of streets previously ascribed to Alfred or his immediate successors may now be earlier, even from the 840s, or developed enough by 860 to make the town an attraction to Viking raiders. There is little to link Winchester with Hamwih, except that fish came from the coast. At Southampton, Holdsworth's 1984 suggestion of a tenth-century enclosure is now doubted, in favour of straggling, organic growth. From large excavations such as these, the history of the town is gradually rewritten.

The two reviewed volumes share a fault. The plans in both are so dense that finding individual features is sometimes difficult; at Winchester, walls and cart ruts, for instance, have colours which are too close together, so finding the latter is almost impossible. We need to stand back from the possibilities offered by illustration software. In both cases, the confusion caused by scores of context numbers could be lessened by making them smaller in size. At present, the features are suffering.

Both publications are a product, in part, of the adoption of word-processing and database software in archaeology in the last two decades. This produces a unified approach among all archaeologists (at least those in the larger units) working in historic towns, which is of the greatest importance. We can now compare pottery, animal bones, plant remains or insects between towns, something hardly done or possible twenty years ago. These two reports join other blocks of data about large areas of medieval towns which have been published since the 1980s, on Hartlepool, Hull, London and Norwich, to name only a few. Techniques of presentation vary between towns, but by exploiting documentary evidence when available, these reports are more informative, inch by inch of ground, than reports on similar-sized areas in Roman towns. The volumes are large and require sympathy with, if not knowledge of, many specialisms. But I am convinced this is the way to go, and I shall be using both these reports for a long time.