In last decades, farmland biodiversity came under large threat. To counteract farmland biodiversity loss and other environmental impacts of intensive agriculture, European farmers can apply Agri-environmental schemes. One of these is the creation of flower strips, a part of the cropping field where flowers are sown or naturally settled. Flower strips are known to increase biodiversity in the agricultural landscape, notably attracting specific insects groups, such as pollinators and natural enemies that can provide valuable pollination and biocontrol services to the crop. However, the plant species composition and management of the strips can have a large influence on the identity and amount of useful insects present in the strips, suggesting the need to develop tailored flower strips to maximize the services delivered. Functional diversity (FD) is sometimes proposed as a promising approach, focusing on plant functional traits rather than plant species itself. Yet, it is not certain that sowing a set of plant species results in the desired vegetation with the desired functional trait composition. Species from soil seed bank or dispersing from neighboring vegetation can settle in the strip, while sown species might not always be equally adapted to local conditions. To test this, we developed seed mixtures with four different levels of FD, based on flower traits, and sew them as flower strips in a conventional arable field. We monitored the vegetation to calculate the FD of the realized vegetation. While the absolute FD values of the realized vegetation were lower than the expected FD values, the realized vegetation showed the same FD gradient as expected from the sown mixtures, indicating that it is possible to manipulate FD in flower strips.