This paper examines challenges to evidence-based decision-making in the design and implementation of rural broadband investment programs. Our focus is on Canada, and the apparent need for further intra-rural broadband research and better data and mapping for informing public investment decisions, but similar challenges are evident in the international literature. Based on proprietary telecommunication provider datasets, the Canadian Radio-television and Telecommunication Commission (CRTC) estimates that broadband services with advertised speeds that meet its basic universal service targets (50 Mbps download and 10 Mbps upload) are available to 87.4 percent of households in Canada. In rural areas however, services that meet CRTC’s speed targets are available to 45.6 percent of households. Moreover, effective speeds and service quality levels that suppliers deliver and users experience tend to fall well below the government’s aspirational targets. In response to demand for better broadband, a variety of initiatives are directing public investment to the deployment of regional and rural broadband networks, which are typically owned and operated by private companies. There remains a serious lack of relevant data and its effective use in creating rural broadband strategies and managing public investment projects. Evidence from the literature suggest that this affects the degree and quality of geo-spatial and econometric analysis that results in a limited empirical basis to allocate scarce public investments, aggregate demand of consumers/communities, and assess the outcomes of rural broadband initiatives ex post. This paper provides a historical overview of rural broadband development in Canada and questions if the body of knowledge to inform public investment initiatives has grown sufficiently to ensure their effectiveness and sustainability. With a regional case from southwestern Ontario, Canada, we discuss the findings of the literature review, characterize the broadband data challenge, and discuss the importance of proprietary provider data cross-referenced with Internet user experience data.