Most Knowledge Graph Embeddings, like RDF2vec, are designed to be trained on a fixed knowledge graph (KG). When that KG is updated, they usually need to be retrained from scratch, which takes quite a bit of time. In this paper, we introduce a method of incrementally updating an RDF2vec embedding instead of retraining it. We conduct an experiment using different snapshots of DBpedia, demonstrating that this is a competitive, yet faster method to obtain embedding vectors of an updated knowledge graph, which sometimes even yields better results than retraining from scratch.