There seem to be surges of spam on the IS stack exchange, most of these bizarrely seem to involve some sort of face cream and are normally from first time posters. Obviously, proactively filtering all first posts is not viable, but first time posts that contain specific keywords could be held from publishing until they are approved for publish.
Is this a viable approach, is it better that retrospective spam deletion?
Spam on Information Security is actually holding at a pretty steady rate. This is a graph of confirmed spam posts per day over the last month:
A lot of these do seem to be centered around face creams or whatever creams they promote these days. It would be possible to keyword-block these, but getting together a comprehensive list of keywords to do so is... not easy. It also has the downside that doing so will not feed Stack Exchange's native spam-blocking systems.
There is also, as far as I'm aware, no mechanism to hold posts for review before they're posted.
I'd advise just continuing to flag spam when you see it to get rid of it quickly. Spam isn't going to stop, unfortunately, so removing it quickly is the next best option.
