A software development project may be considered a failure because it is late. In order to be able to assess this, a project start date should be known. The purpose of the paper is to study software development projects in a business context and especially project start from the supplier's perspective. In our research, we observed different ways of assigning project start but did not observe any company-level instructions for defining it. We raise questions regarding knowledge loss, project profitability, and having a project running late even before the project has started. We provide definitions for project start and project start date, and define boundaries for software supplier's project start-up. With this paper, we emphasise the need to study software development projects in a business context. The paper contributes to research on project management success, the software project business, and the management of a software project in a business context. (C) 2015 Elsevier Inc. All rights reserved.