Fitting ocean models to altimeter sea surface height (SSH) measurements requires knowledge of instrument noise (radar noise, sea state bias, path delay corrections, and orbit errors) and “representation” errors related to SSH signals (e.g., tidal or pressure driven) not computed in the models. Comparisons between the independent Ocean Topography Experiment (TOPEX)/Poseidon and Jason-1 altimetric missions when they were in identical orbits show that point by point the data are consistent within the mission specifications of about 3-cm rms, but large-scale dependences exist in the data differences, and these are both poorly known and capable of introducing major errors into oceanic state estimates. Here the authors focus on the time-variable component of the spatially dependent errors. The analysis reveals errors ranging from 2 cm in the Tropics to 4 cm at mid- and high latitudes and roughly consistent with a dependence of instrument noise on significant wave height. Analysis of the representation errors suggests that, over the deep ocean, uncertainties associated with the simplifying assumption of an inverted barometer response to pressure loading are larger than the remaining errors in modeling the large-scale tides. Over extensive regions, however, errors associated with eddy signals missing in coarse resolution models dominate. Obtaining a more quantitative estimate of the latter errors remains a challenge.