Forum Moderators: open
We have a database with 8 different types of content, with each between 500 to 5.000 items. All items are categorized, and can be grouped by author. Each items has only 4 fields, including the author, title, category and a small extra desciption (mostly keywords). The site is server-side scripted.
We have done basic SEO on the site, like SE friendly urls (the items can be found by www.example.com/type/category/number-artist-title.htm), semantic HTML and a bit of title/meta optimisation). One can reach the best-selling items in 1 or 2 clicks, the rest is on subpages, by category or so deep-down that it's only findable by the search-box.
The problem is that our site is basically a sort of big catalogue, with very limited textual content per-item. This also results in quite a lot of long lists of links, that possibly can be seen as link-farming (despite it is for the users convienience and every link is unique).
Does anyone have any advise on how we can improve our search results? Most SEO info I encounter is on sites that have a lot of text/info, which doesnt completely cover this case.
For instance:
Should we forget about the big bulk, and focus on the top sellers? Or maybe create a new site with only the best selling artists and write big blobs of semi-informative-but-attractive text?
Im thinking about grouping the content under (dynamically generated) subdomains, like [artistname.example.com,...] and then show all types of items for that artist, plus a small list of human-edited 'related artists'. Will this help us?
All your input is greatly appriciated: your random thoughts, sound advice, or theoretical insights on making catalogue items findable.
Then very important pages in lower levels can be linked from the homepage in body text, while controlling the amount of PR distributed and number of links going out from each of the levels.
-- Level A (Homepage)
----Level B Category 1
----Level B Category 2
---- etc.
Like that for one level down. Pyramid structure.
The problem is that our site is basically a sort of big catalogue, with very limited textual content per-item. This also results in quite a lot of long lists of links, that possibly can be seen as link-farming (despite it is for the users convienience and every link is unique).
That is probably a rather large problem that you'll need to address. Those pages are most likely going to be filtered to the Supplemental index due to lack of content. There's a threshold that can be crossed at which point the page appears to be a "stub" which is pretty much an empty page to the bot. Without that unique content on each page, most of those pages are duplicates from a bot's eyes.