Paleomagnetism is the only quantitative method available to test for lateral motions by tectonic plates across the surface of ancient Earth. Here, we present several analyses of such motions using strict quality criteria from the global paleomagnetic database of pre–800 Ma rocks. Extensive surface motion of cratons can be documented confidently to older than ca. 2775 Ma, but considering only the most reliable Archean data, we cannot discern differential motion from true polar wander (which can also generate surface motions relative to the geomagnetic reference frame). In order to find evidence for differential motions between pairs of Precambrian cratons, we compared distances between paleomagnetic poles through precisely isochronous intervals for pairs of cratons. The existing database yields several such comparisons with ages ranging from ca. 1110 to ca. 2775 Ma. Only one pair of these ages, 1110–1880 Ma, brackets significantly different apparent polar wander path lengths between the same two cratons and thus demonstrates differential surface motions. If slightly less reliable paleomagnetic results are considered, however, the number of comparisons increases dramatically, and an example is illustrated for which a single additional pole could constrain differential cratonic motion into the earliest Paleoproterozoic and late Neoarchean (in the interval 2445–2680 Ma). In a separate analysis based in part upon moderately reliable paleomagnetic poles, if a specific reconstruction is chosen for Laurentia and Baltica between ca. 1265 and 1750 Ma, then those cratons’ rotated apparent polar wander paths show convergence and divergence patterns that accord with regional tectonics and appear to be remarkably similar to predictions from a platetectonic conceptual model. Carefully targeted and executed future paleomagnetic studies of the increasingly well-dated Precambrian rock record can imminently extend these tests to ca. 2700 Ma, and with substantially more effort, to perhaps as old as ca. 3500 Ma.