Current classifications of carbonate platforms use depositional gradient as the main criterion for separating systems into two end-member types, ramps and flat-topped platforms (FTPs). However, many examples do not conform to this simple classification. To investigate why this is and to better understand probable controls on platform development, we have used a series of 2D numerical forward model runs to investigate how sediment production, diffusional sediment transport, and other controls such as tectonic subsidence, antecedent topography, and relative sea-level oscillation interact to determine platform geometry. Modeling results reaffirm that rates of down-dip sediment transport relative to rates of autochthonous production are a critical factor in maintaining a ramp profile in stable cratonic settings under a constant rate of relative sea-level rise. Type of carbonate production versus water-depth curve, for example euphotic versus oligophotic, is not a significant control in our model cases. Both euphotic and oligophotic production profiles produce FTPs when diffusion coefficients are low relative to production rates, and ramps when diffusion coefficients are relatively high. These results suggest a continuum of platform types, ranging from transport-dominated, low-gradient systems at one end of the spectrum, to in situ accumulation dominated systems at the other. A system may be transport-dominated because high-energy processes are able to break down and transport even bound sediment, or because carbonate factories produce only sediment that is easily transportable under even low-energy conditions. Time evolution is also probably important. Initially low gradient systems will, in the absence of sufficiently high sediment transport rates, tend to evolve towards high-gradient flat-topped steep-margined platforms. Many observed or inferred platform geometries are therefore likely to be transient forms, and this could complicate interpretation. Investigating how basin bathymetry and style of subsidence control platform geometry suggests that, in transport-dominated systems, strata simply drape the underlying topography, and that pre-existing breaks of slope and differential fault subsidence are a stronger control on platform geometry in in situ accumulation dominated systems. Rotational subsidence tends to create transport-dominated systems during rotation as the topographic gradient increases and transport rate increases and outpaces in situ production rate. Relative sea-level oscillations tend to move the locus of sediment production laterally along any slope present on the platform, distributing the sediment accumulation across the whole width of the platform, suppressing progradation and steepening, and so favoring development of low-gradient systems. Based on all these results, we suggest that a simple cutoff classification into ramp and flat-topped platform types can still be useful in some circumstances, but a more meaningful approach may be to describe and predict platform strata in terms of a multiple-dimension platform parameter space containing a continuum of geometries controlled by sediment production, sediment diffusion coefficient, antecedent topography, differential subsidence effects, relative sea-level oscillations and perhaps other as yet unappreciated controls.