This paper presents the issues that arose when using observation as a method to investigate the friendship experiences of skilled migrants to Australia and will draw on one case study by way of example.  Spencer and Pahl (2006) note that during interviews participants tend speak about their friends in terms of idealised concepts that differ from the enactment of face-to-face friendship.  Despite this, the scholarship on friendship research favours interviews as a method.  This paper takes Spencer and Pahl’s (2006) concern to avoid idealised concepts by using an alternative method - observation of a migrant meeting with their friend - to understand friendship in action and to supplement data gathered during semi-structured interviews about migrants’ friendship experiences. 
The findings add to the literature on the use of methods appropriate for friendship research by providing valuable insights about the effectiveness and challenges of observation, which were: the participant’s understanding of the method; a tension between the role of the researcher as observer or participant and the social norms of friends meeting; the spatial context of the observation, and language issues between the researcher and participants.  In conclusion suggestions will be made about ways to manage these challenges in future friendship research, such as provision of clearer information about the method to participants; and a need for the participants to have an opportunity to articulate their expectations of the method prior to the observation.  Nonetheless, this method is found to be useful in supplementing interview data and in generating a more holistic understanding of a migrant and his friend.