This is something I've given a lot of thought to, as I've been trying to create a rocket design to get a BT-20 rocket as high as possible. From my thought experiments, getting max altitude is the easy part; the hard part is trying to figure out how to get the rocket back.
GPS is an obvious option, but it's expensive. Oh, and it adds weight. If you're gunning for altitude, every gram counts.
A small electronic buzzer device and using a directional mic seems like a good option too, as has already been suggested by
@Charles_McG. I see the biggest problem being that you have to be within a few hundred feet to find it if it's lost. And if it's lost, you may have ZERO idea as to where it could be.
My thought is to add a very bright blinkie LED and launch at night at a location with little light pollution and a moonless night. I imagine this hypothetical device would be as light, if not lighter than the buzzer, and a hechuva lot cheaper than a GPS unit. It would also have the potential to have a further "spotting range" than the buzzer.
Lastly, I think there's the the possibility of hiring extra sets of eyes. Not sure how this might work, but maybe everyone gets $1 for watching your launch and trying to track your rocket. If you lose track of your rocket and have to rely on someone else, they get $5 if they see it and can show you were it is. And if the rocket is temporarily lost, a $20 reward to the first person to find it. I really think having an extra few sets of eyes during your launch can be a way of dramatically improving your ability to track the rocket without adding any weight to the rocket itself, and for relatively nominal cost. It might even help to coordinate the visual tracking, with 2 people assigned to a given sector of the sky (one pereson has binoculars* and one person eyes their naked eyes). Yes, it requires the coordination of others, but again, it's relatively cheap and adds no weight to your rocket.
* If using binoculars, I'm sure there's a sweetspot for magnification, but a wide enough view of the sky to be useful. Also, it might be a good idea for at least one person in the pair to have a good quality compass, with a bezel of at least 2 degrees of resolution. This can be VERY helpful if only 1 person was able to track the rocket until it hit the ground and they can reference a specific tree or landmark. Then they shoot a bearing with their compass and you'll have a a specific line of where the rocket is located; you just need to figure out how far away the rocket it. If you have 2 other compass users that also saw where the rocket fell, you can triangulate the position of the rocket. But if that many people saw where the rocket fell, it probably won't be too difficult to find it the old fashion way (assuming the landing spot is accessible on foot).
