Organisations manage large amount of data scattered in multiple data sources disseminated within the organisation. In this context, decision-making having a complete vision of the available data to make the right decisions is a challenge. Choosing the right strategy to bypass this issue may be impactful on the enterprise data management policy. A good alternative to limit data integration issues can be found in Knowledge Graphs. These Knowledge Graphs allow decision makers to understand which data of interest are existing in the enterprise. They are integrating the semantic of data rather than data itself. In a previous research, we proposed a decentralised enterprise knowledge graph (DEKG) architecture to facilitate decision making and consider the scattered data sources of interest. Due to the decentralised dimension of this architecture and the separation of the data that is scattered in remote sources, but also to the "centralised" aspect of knowledge graphs, one of the remaining challenges is the decentralised query management in order to answer a decision-maker needs. As a result we define in this paper a four step approach to manage queries and to build results based on a DEKG. We illustrate the proposed approach through examples and detail its implementation. We also show the results of preliminary experiments before giving some future work.