Background: The stand productivity of fast-growing forest plantation varies across short distances depending on site and forest characteristics. This indicates that forest management would benefit from a site-specific approach. A tool to characterize such productivity variations are yield maps and a cost effective source of data is automatically collected by harvesters. To create such maps we need to understand the effect of geospatial accuracy of tree location recorded by the harvester. Methods: This study investigated data sets from four stands: two had very accurate tree location, and two were harvester data files that have inaccuracy associated with both the GNSS recording under forest canopy and the physical dislocation of the GNSS. The GNSS unit is on the cabin of the machine, but the tree is felled using a boom and could be up to 12 meters from the cabin. Results: We establish a spatial resolution for studying variations in stand productivty mean tree volume and stocking across stands to allow the development of forest yield maps from harvester data. Conclusions: Assessing variability across a range of cell sizes from 10 x10 m to 100 x 100 m, we conclude that a cell size between 30 and 40 m is suitable to use as a reference for calculating volume per hectare and mean stem volume, and 60 m cell is more suitable for evaluating stocking. The variability pattern is consistent for the various accuracy levels. When the trees' position is relatively inaccurate, using mean tree volume and stocking per cell might be a method for mapping productivity from harvester data.