I think it might be a lot of work doc!
I decided to do something similar. When I began researching my tree I was very surprised how common my surname - Wilding - was in Lancashire, especially around Longton where my gggg grandfather lived. I have a LFHHS fiche of burials from Longton St Andrew where Wilding is the most common surname.
And like you I wondered whether everyone was related, and so I jumped in and tried to build a tree of the Longton Wildings.
Four years later and with more than 3,000 people on my tree, I haven't finished yet, although I do now think that the end may now be in sight. In your case, it will be more difficult because Cross is a much more common surname. Even with my less common surname I've chosen to ignore "outlying" colonies of Wildings around Woodplumpton and around Standish. Otherwise the work would become too much. So I'd recommend putting some realistic limits on what you want to achieve.
As for how to approach the task, I've mainly worked with marriage and census info. Full marriage info is vital because of the parent, occupation and address information. The LAN-OPC website is invaluable for marriage info, and for parishes and periods it doesn't cover I simply did the transcription work myself at the Lancs Record Office using either their fiche and film, or the original church registers. So for churches not covered by LAN-OPC I've now transcribed nearly all the Wilding marriages from all the churches in Preston and south to Leyland and Chorley.
I do think that eventually I will have completed something that was worth doing, and it's great when different strands begin to connect up. But I can't stress enough how much work it could be.
Good luck if you decide to go ahead. BTW The Cross name features quite prominently on my own tree, but they're all around Longton and Leyland, rather than the area you mention.
Tim
