In recent years, the demand for results of business surveys broken down by region and content has increased significantly. Often, however, the sampling design of a survey is only designed for a reliable design-based estimation at the state or federal level due to maximum permissible sample sizes. If additional estimated values are to be determined for regional or content-related subpopulations, insufficient sample sizes might lead to unacceptably high variances of the estimates. A subpopulation in which the sample size is not large enough for a direct design-based estimation of sufficient precision is called a small area. So-called small area estimation methods can be used to calculate precise estimates for the respective subpopulations. These mostly model-based approaches rest on the supportive inclusion of additional auxiliary information from other subpopulations using a statistical model. However, the lack of timeliness of business registers usually used as a sampling frame for the design of business surveys or the time lag between the sampling design and the data collection itself lead to inconsistencies also referred to as frame errors. As a result, the variables contained in the business register might be obsolete. In addition, the register and the target population differ in terms of their composition and the number of businesses. Nevertheless, the register is a potential source of auxiliary variables for small area estimation methods. This however may create problems, as erroneous or obsolete auxiliary variables may cause small area estimation methods being even worse than classic design-based estimators. Furthermore, the sampling design of business statistics usually includes a stratification by industry groups and size classes. Due to the lack of topicality of the frame population and the strong dynamics of business populations, however, industry-specific and size-specific stratum jumpers may result. These are businesses that would have been assigned to a different design stratum if the correct design information had been available at the design stage. Accordingly, the assumptions under which the original design weights were determined within the sampling design are no longer applicable. Building on the challenges of business surveys elaborated above, the potential to improve the estimations for small areas in business statistics is analysed. In this context, the inconsistencies between the available frame population and the target population referred to as frame errors are considered in particular. On the one hand, various small area estimation methods are implemented and compared with one another with regard to their ability to improve the estimation quality despite outdated auxiliary information. On the other hand, as the assumptions under which the original design weights were determined within the sampling design no longer apply, various reweighting approaches are developed and evaluated. The comparison is made taking into account different frame error scenarios. Therefore, we examine the extent to which small area estimation approaches using outdated auxiliary information and various reweighting approaches can achieve an improvement compared to the classic design-based Horvitz-Thompson-Estimator in various realistic scenarios of stratum jumpers.