Ian,
It is quite possible that this would work. However, much will depend on the specific magnet chosen. Magnet size, polarization, and material will all impact how well the sensor will be able to detect the magnet.
The sensor itself will only be sensitive to the portion of the field perpendicular to the surface of the package
In this case we do expect to see a portion of the field oriented in the correct direction, but it becomes a question of how strong that field will be. Again, this is very dependent on the qualities of the magnet selected.
It may be useful to consider a field calculator tool such as the one available at the link below
https://www.kjmagnetics.com/fieldcalculator.asp
Additionally, there are open source simulation tools such as FEMM (Finite Element Method Magnetics) which can assist with determining more complicated configurations.
http://www.femm.info/wiki/HomePage
We have an application note regarding design around Inductive Sensors using this software which may be useful.
https://www.ti.com/lit/an/snoaa04/snoaa04.pdf
Thanks,
Scott
