[Background: We have a guide to estimating IP camera cabling costs.]
Unfortunately, in some cases that's what you have to do. To me it depends on what type of facility it is, and what plans you actually have available. I found that in most cases 200-250' was the sweet spot between accuracy and covering things so we weren't way underestimated. I also generally rounded up to the next 1000' or at least 500' of total cable, and used that as my basis for material and labor.
In some cases, I guesstimated the location of the head end and used Google Maps to measure some average cable runs (cameras on the outside of a building, for example). That can help to hone in on an average, too.
It depends. If it's a small place then I use 150' per run. If it's larger than for sure 225-250' per run. If it's bad then I used 325' per drop.
If you are working from any old rough drawing or napkin, Google Earth is your friend. With accurate scale and a pretty good measuring ability, as long as you know where the cams are to go and where the headend will be, you are 90% home.
You can get a much better idea of cable runs and therefore make a better, more accurate, more efficient proposal where neither you nor the customer needs to suffer.
Other than that, I tend to not get into jobs with a lot of guesswork (unless I have a bailout clause
- or a cash inject clause)!
Google Earth works great I used a lot but additionaly having asite survey helps more than that. Depending on the environment you may need additional labor or parts. Firestopping is the big thing now that has to be heavily taken into consideration when estimating. Poor conditons, forklifts driving around all day in the way of lifts, additional conduit, etc all have to be taken into account. I would change my labor per foot up or down depending on the cable running conditions. Then of course you could just get a bad installer and then there's nothing you can do right...