There is no standard for GPS data. As a result, the unit you buy-- old or new or recently updated-- may be wrong, as in very wrong.
The first event to mention was held in Carmel Indiana at Carmel's new Monon Center. It's a wonderful facility, and probably cost too much. The roads to it have just been built, but have been planned for a long time. The Center was dedicated some months ago, so updated GPS units ought to be able to find it.
The problem is: no one updates them. Even though there are signs pointing to where it ought to be, and everyone received printed directions, they relied on their GPS units to get them there, as they rely on them to get them everywhere they're personally unfamiliar with. The damage was that many people were not only late, but fuming when they got there. We're talking angry. Yes, most subsequently relied on their maps or directions from a gas station nearby. But this is after their GPS units sent them all over the small burg of Carmel.
The second event was much more minor in nature, at a facility on the Indiana University campus. It's well-known to the 40,000+ students on campus, a building that's been around for years, Two out-of-towners were sent down 7th Street, which has been blocked-off on campus for almost a decade or more. I had a hard time finding people in Bloomington, the home of IU, that remembered it going through campus. But the GPS was sure it did. The duo were deposited at a dead-end. It took a while before they gave up and asked for directions. Just what a GPS user loathes to do.
What's needed? A method that allows civil engineers to update maps to a spec, nation, perhaps worldwide. Not a Google satellite view, rather the real platted map. Then there needs to be a downloading method that's both convenient and easily understood for users. Not all GPS vendors let you do this, but the better ones, like Garmin and Tom Tom do. Then, make the GPS turn off if it's not updated every 90 days or so. Don't let users get lost. Don't let them become so enraged that they want to dismount them and toss them across the parking lot, swearing your name.
There is no method, no standard, no reference point that we can use to make sure the reference data is good. It needs QC checks. It needs user feedback (hopefully minus the swear words). And finally, GPS systems need audit integrity and a means to show off that integrity. Otherwise, they make interesting noises when flung against something hard, like a concrete wall. In some cases, it's a satisfying sound, I'm told.