At a guess, inadequate unique relevant content.
In most areas, it's hard to create a "directory" site that really adds value to _the_ (open) directory. Our preferred approach is to mine the site for links and add those links to our category instead.
And even if you build one that's good enough to be added today, in six months we might have an active editor, who would then mine (and delete) it, unless you'd really kept running in the meantime.
Which comes back to ... in the long run, very few of them add value for our users. And you should basically expect a directory site to be rejected, unless it has something the ODP CAN'T do better.
There are things the ODP doesn't do, for example:
-- Slice data differently than the ODP does
-- Include entities that might not even have websites: Pentecostal Eastern Orthodox Churches in Athapaska County, Montana; Ethiopian restaurants in Podunk, New Jersey; Boy Scout Troops in Guyana; mail-order venders of decaffeinated pistachio nuts; etc.
-- For highly focused sites, deeplink substantially deeper than the ODP: MIDI file directories by composer, etc.
Real estate agents mostly have websites, the ODP indexes them down to city level, and there's no point in indexing deeper, so none of these apply. The marketroidish approach to generating content just isn't of interest. We tolerate sites built with that attitude if they also contain unique authoritative information about a business -- but we really appreciate it not at all. It takes a completely different mindset to generate informational content. I'm persuaded that many marketroids simply cannot do it -- any more than I can write adcopy.
In my opinion the only thing to do is save the URL; discard the website, hang the webmaster, shoot the content provider, take a sledge hammer to the hard disk from the server; and starting over from scratch. 99.99% of the time you really can't get here from there.
