Before adding a sensor, I check what information the existing system can provide. An amplifier might report its state. A control processor might know which source is selected. A schedule might tell us when a room is expected to be in use.
Those signals can be useful, but they need interpretation. An amplifier being on does not tell us how many people are in the room. A scheduled event does not prove that it happened.
Start with a specific question
“Is the system ready for the next class?” is a different question from “Is anyone moving in this room?” The first might be answered with equipment status and a signal check. The second may need sensing hardware and a separate experiment.
I list the decision the information will support, the available inputs, and the conditions that could make each input misleading. That makes it easier to decide whether an existing signal is enough.
Find out what is actually accessible
Useful information may be available through a documented API, a control protocol, or a device status page. Some equipment exposes very little. Network access, credentials, and permissions also affect what can be collected reliably.
Wi-Fi motion sensing needs compatible hardware and access to suitable radio measurements. It is not a feature I would assume an ordinary router provides. I wrote more about the constraints in Using Wi-Fi to detect movement.
Test the wrong answers
If an input is going to trigger something, I want to know what produces a false trigger. An HVAC fan, an unattended audio source, a delayed schedule update, or a device reconnecting can all confuse a simple rule.
I also check what happens when the input goes stale. The interface should show the age of the last reading and distinguish an unavailable measurement from a valid zero.
Collect only what is useful
Access to data is not a reason to keep all of it. I want a clear purpose for each measurement, a reasonable retention period, and an explanation of who can see it. Existing network information can still reveal activity, even when no camera is involved.
Sometimes the best answer is a new sensor. The existing signals are simply the first place I check.