Text visualisation is gaining popularity. Researchers have investi- gated methods that display the structure of documents, show how books have evolved over many editions, how words are distributed in text. But few researchers have investigated how visualisation techniques could help someone write better text documents. The goal of the writer is to create texts that are not just grammatically correct, but also idiomatic, and fluent to read. Our focus is to in- vestigate collocations, i.e., words that have become conventional to use together (fast food vs. *quick food, collect data vs. *pick data, strong tea vs. *powerful tea, etc.) While researchers are starting to use visualisation techniques to help users write better computer programs, musicians create music, or artists be more creative, there is less research in writing. In this short paper we not only make a call for action, but we look at related work, and discuss some the main challenges and opportunities for research in this area.