A multimeter will in general be close, but yes - since it does not place a load on the battery, you don't get a true measure of it. You're simply measuring the terminal voltage. As the amp-hours of charge in the battery decay there is a voltage depression which moves along a curve - this curve varies with the type of battery (and, even brand).
Think of it like this, as I'm sure we've all done this a time of two. You have a device with a battery that is nearly dead. The device can't be powered by the remaining juice in the battery, so it turns off. If you let it sit powered off for a time, you can almost always at least get it to somewhat turn back on.
For real life, using a multimeter gets you enough in the ballpark for what we do. When I want to really know, I bust out this:
[Invalid or Expired Link Removed]
And usually test a battery for at least 10 seconds to see what the final readout is. I've seen some pretty big swings on batteries.
TL;DR - Multimeter is likely close enough for our application, but I've seen enough deltas of multimeter vs load test to only really trust load testing.