OBJECTIVES To present an alternative implementation for the conventional Markov models with area under the curve (AUC) approach: the continuous model (CM). To present how the CM avoids the need of determining cycles in theory and to compare the traditional and the CM approach in terms of results and complexity in an oncology model example. METHODS The AUC model assumes that the survival function is known at any timepoint not only at the beginning and end of model cycle. The CM calculates the model outcomes for the whole timehorizon by using the values of the survival function in every timepoint, instead of the discrete timepoints defined by cycle length. The CM approach overcomes the issue of the artificial characterization of time using cycles, that is often criticized in Markov models. Using CM can also lead to more precise estimates. A simple oncology AUC model with three health states (progression free survival, progression and death) and four-weekly cycles was built and converted to a CM model in Excel®, using user defined Visual Basic functions. Results, generalizability and user friendliness were compared. RESULTS The results of the two models were similar: for health outcomes differences were around 1%, for costs and incremental cost-effectiveness ratios around 0.5%. Calculations were done in a single cell/outcome instead of a column of 100-200 cells depending on cycle length and time horizon, giving less scope for bugs and facilitating easier debugging. As a result the implementation of the CM model was faster and technical validation easier. CONCLUSIONS The CM approach requires more technical background from the developer; custom functions have to be built even for point estimates. However, results of a CM, requires smaller spreadsheet space, and provides more transparency and easier debugging, while providing similar or potentially more precise estimates compared to the AUC model results.