It has just occurred to me that if you can make or get hold of a
switched attenuator to stick in between the antenna and the input socket
of your remote receiver, you can make very accurate measurements indeed.
I am thinking of the type with 8 or 10 switches. The first switch gives
1dB of attenuation, the next 2dB, 4dB, 8dB, and so on. So long as you
have some sort of signal strength meter you can monitor on the receiver,
you just switch in enough attenuation to give the same meter reading at
each test location and record how much attenuation you have switched in
at that point. The more attenuation, the better the received signal.
That will allow you to determine relative signal strength to within 1dB
which is going to be good enough for your purposes. The attenuator can
be used for all kinds of projects, so it might be worth taking the time
to build one irrespective of what you end up using for a signal source.
The usual Google search will turn up construction details, just
resistors and switches in a screened box with some PCB offcuts or copper
foil to provide internal screening between each section.

This is a bit of a challenge to make accurate to 1 dB at 144 or 440 MHz.
The leakage around the switches, etc, is difficult to deal with.

You're probably better off scrounging for a decent surplus/used step
attenuator made by someone like Weinschel or HP.