In response to both posters... a problem with selling positions on any national type site is that Google generally will rank only two pages from a domain for a given query. So you couldn't offer a company a page of its own if you expected to rank more than two pages for each "placename widgets" search.
To take on more than two clients per placename, you have to turn the site into a directory, and Google, as we've seen in many verticals, has for some years now been resistant to most directory models. Some sites like Yelp do well, because of user-generated content and because of the essential differences among city businesses at a very granular level, but many don't.
The problem with this strategy is that I'm wondering how difficult it will be to rank for a multitude of keywords in each city, when only one page is optimized specifically for that city.
This gets into repetitious content pretty quickly, where the only thing that changes is the placename and maybe some material about the location lifted from Wikipedia. Essentially, you're building doorway pages... and I do think that Google doesn't want these, even though some such pages, in noncompetitive areas, can hang on for many months.
In the Google SEO forum in our supporters section, we observed some doorways that ranked for about 8 months. The site in question targeted maybe a half-dozen or so keyword combinations per city/ small-town in its list.
I should add that there are many reasons why I've advised clients not to go with
SEO companies offering either model described in the first post. In general, I think that SEO in limited niches gets inbred pretty quickly... you either run out of ideas for content, or you run out of linking approaches. With niche SEO companies, you ultimately end up with cookie-cutter content that has a dreadful sameness to it.
A company running a franchise is somewhat different from an SEO company, but you're still faced with some of the same tasks. A national site offers some advantages if dealers have exclusivity of territory and/or are cooperating rather than competing.
If you're the exclusive e-commerce retailer, I would think you'd have to have a back end that would send customers to local physical stores based on some kind of internal search (distance or zip-code based), posting information about who has what in stock... and that you might have an arrangement something like Target's. But those stores really don't have separate identities. Business model inevitably enters into it.
I once optimized a chain of something like 16 hotels under common sales management but otherwise separately owned and highly individual, in the same metro area, and at a certain point I insisted the we draw up a spreadsheet and give each hotel only one or two preferences... because it was really impossible, with roughly the same approach, to do an even-handed job for all of them.
Lots of good, well-edited user-generated content, as I noted above, with unique things to say about each place (or about businesses within a location) is probably the only way around the sameness you'd get in either the directory or the separate site approach.... but those approaches really don't handle e-commerce.
