Hi Anna,

Something seems amiss with this method, and I can't quite put my finger on it. It looks like you're building communities around a quota of buildings/services instead of creating buildings/services to suit the community's population.

But I could be misinterpreting. So I'll disclaim both knowledge of your campaign setting and historical accuracy. My experience is with low-fantasy/quasi-historical settings, and while the values I use are based on real-world research, I'm not an archeologist or historian. That said, the numbers work well for my campaign; YMMV, so please just take this post as friendly advice. I hope it's helpful, but I won't be offended if you discard it.

Based on my research and taking a cue from S. John Ross' work, here's an alternative approach:
  1. Classify the settlement as rural or urban
  2. Determine population
  3. Determine social roles (noble, freeman (artisans/composers/shopkeepers, etc.), peasant, serf)
  4. Determine number of households
  5. Allot buildings as appropriate

This takes into account the number and types of buildings you've broken out, but ensures that they exist only if the total population numbers can support them. For instance, instead of saying that 3% of all village buildings are shops, this method says the number of village shops is based on the number of village shopkeepers, the number of which is based on the total population.

It also avoids redundancy. For example, in a rural settlement, a majority of residents will live on their farm, so one farming household gets one building (which counts as a farm and residence). Many shopkeepers will live in the same building as their shop, so one shopkeeper household gets one building, which counts as a shop and a residence. Conversely, an industrial type probably has a residence separate from his industry, so one industrial household gets two buildings: a residence and a "factory" (could be a smithy, a tannery, etc.).

Determining social roles is the most important part of this process. To do so, I use a modified version of S. John Ross' Merchants and Services chart. Without recreating it here, I'll say that it takes into account the number of:
  • Nobles - number of families and relatives, servants, and houseguards in each
  • Officials - Reeve, Messor, Woodward, Constable, and town watch
  • Clergy - Clerics and Priests
  • Freeholders - Adventurers, Apothecaries, Armourers, Artists, Butchers, Chandlers, Charcoalers, Cobblers, Entertainers, Foresters, Furriers, Glassworkers, Innkeepers, Jewelers/Moneylenders, Litigants, Locksmiths, Masons, Metalsmiths, Millers/Bakers, Ostlers, Outfitters, Physicians, Potters, Roofers, Ropemakers, Sages/Heralds, Salters, Scribes, Shipwrights, Tailors, Tanners, Taverns, Teamsters, Timberwrights, Tinkers, Vintners, Weaponcrafters, Weavers, Woodcrafters, and Yeomen
  • Citizens - everyone else (basically serfs engaged in some labour industry)

I'll make some additional assumptions (again, these are based on my own campaign world):
  • Mediaeval households numbered about 5 before the black plague and less than 4 after; cf. http://www.bbc.co.uk/dna/h2g2/A6952188; I tend to low-ball the figure at an average of 3.75 people per household
  • Population dictates settlement classification: 10-100: thorp; 100-1,000: village; 1,000-6,000; town: 6,000-11,000; city: 11,000-20,000; metropolis: 20,000+
  • Settlements of less than 3,000 people are considered rural. Until a given area can attract more people, it isn't sufficiently developed to support urban subsistence
  • In rural areas, one acre can support 1.25 people; in urban areas, one acre supports 60 people

With all this in mind, consider:

Thorpton is a rural community with a population of 238 (village-sized) occupying an area of 190.4 acres (0.3 square miles). The population breaks down as follows:
  • 15 Nobles - 1 ruling house (8 relatives and 1 servant) and 1 guildmaster house (6 relatives)
  • 18 Officers - 1 reeve, 1 messor, 1 constable, 2 town watch (includes family)
  • 2 Clergy - 2 clerics
  • 63 Freeholders - 17 businesses/services (2 cobblers, 1 entertainer, 1 furrier, 1 innkeeper, 1 jeweler/moneylender, 1 metalsmith, 1 miller/baker, 1 ostler, 1 potter, 1 roofer, 1 salter, 1 tailor, 1 tavern, 1 timberwright, 1 woodcrafter, 1 yeoman)
  • 140 Citizens - everyone else, in 37.3 households

Based on the above, then we break out buildings as follows:
  • 2 noble households
  • 5 officer households
  • 1 shrine
  • 22 businesses (17 businesses, plus 1 smithy, 1 mill, 1 stable and grazing commons, 1 food stores, 1 lumberyard)
  • 37 residences/farms
  • 67 total buildings

Again, YMMV, but I've found this method to be highly useful and it's provided very workable results. It takes a bit more work, but the math can easily be automated with an Excel spreadsheet or online tool.

Hope this is of value to the group. Cheers,
-Erin