Engineering-to-order is a production approach that aims to develop each customer order from scratch based on specific requirements, and is therefore inevitably characterised by a high product variety. It is clear that this strategy is costly, but previous research is lacking in terms of providing real-life data regarding how a high product variety affects costs. Therefore, this thesis aims to analyse the cost-variety relationship through a single-case study consisting of observations, interviews as well as calculations, with respect to the value-creating process of a Swedish engineering-to-order business. The results reveal that while the cost of variety may not be influenced at the product level, it can be influenced at the component level. Furthermore, the three aspects that drives component variety cost the most are the number of new components in a product, the number of different components at the company as a whole, as well as the number of different suppliers. Of these, the most influential aspect is the time spent working with new components, which consumes roughly 10% of the total working time throughout the value-creating process. Moreover, the effects of these aspects differ across different parts of the process as well as component types. The results show that the largest costs are driven upstream in the value-creating process and for advanced components. It is however difficult to decrease the variety of advanced components as a rather high variety is required to meet customer needs. Instead, less advanced components offer more easily obtained savings. Finally, the thesis also suggests three ways to mitigate the negative effects of high product variety. Firstly, improving the accessibility and spread of information to ensure internal alignment in the management and maintenance of components. Secondly, identifying common synergies across different projects and products to create economies of scale. And thirdly, explore to what extent processes and components can be standardised.