The Necessity of Estimation in Software Development Projects

Abstract

Today most Software Development Companies use Agile Project Management to manage their project, here the decisions are made as the project progresses to keep the agility in the project. This has been a successful transition with a lot of benefits, but many agree that estimation remains a great area of concern. The most important reasons for this are the unknown parts of the project that make it hard to calculate accurate estimates. Many teams are unable to deliver what they have committed to, because their estimates are often way off. This results in large project overruns, which prevent the organizations from reaching the market early enough and can result in bad market performance.

In this thesis, the necessity of estimation in Software Development Projects is discussed. Interviews with experienced project managers where held to find out how theory coincides with the Software Development Industry, Also, some of the main goals of these interviews were to find out more about how estimates in Software Development Projects are created within the industry today, and how necessary it is to have accurate estimates or estimates at all, for project managers.

This thesis shows that there are a lot of benefits and often necessary to have good estimates. The thesis also shows that it is important to have early specification of parts of the project, where the decisions about the software should be made as early as possible to create the most accurate estimates. But only the parts of the project which are soon going into development must be heavily specified, otherwise agility will be lost. For less specified parts of the project rough estimates should be enough to make the right decisions.