'Scientific' canvassing is hardly revolutionary. Ian Mikardo was at it in the 1940s.
you might note that I said evolutionary. They put together a lot of new and some comparatively disused techniques to produce a more effective system of campaigning.
You mention Mikardo. Yes, politicians have been doing these sorts of things since forever, but, there are 150 million voters and civic participation has been declining for years. How do you take your limited resources for contacting and identifying who they are and use them to the best effect? You target, and the refined targeting allows you to target all the better and focus those resources on having a better idea of who to talk to to find IDs and then on election day a better idea of who needs an extra push to the polls.
Again nothing revolutionary, but evolutionary. It's using superior research and organization to inform the use of timeless campaign methods.
Wow. You could be one of these consultants yourself. That's a big ass paragraph and it basically says nothing.
You might be on to something...