Forecasting and planning using data is fast and easy. Even without data, there are ways to forecast better without story estimates using well-known techniques. This session teaches how to quickly model a proposed feature or project to achieve reliable forecasts that can be tested and improved over time.
The goal of this session is to teach how to use concrete tools and knowledge that allow software projects to be rapidly forecast using historical data when available, or range estimates when no relevant history is available. To do this, this session will demonstrate practical tools used in real engagements over the past 10 years, all of which are free to download. I'll explain the thinking behind these tools and the safeguards put in place to avoid the forecasting failures.
The session will cover:
By attending this session you will be well placed to build your own simple tools to perform probabilistic forecasting, and you'll also be pointed to free spreadsheets and websites that do this for you. It really is simpler than most people believe, requiring no advanced mathematics or statistical knowledge. YOU will get this!
Troy is the founder and consultant for Focused Objective LLC, the leading quantitative analysis and forecasting vendor for the software industry. He regularly speaks at industry conferences and promotes better use of historical data, metrics and quantitative techniques to anyone who listens.
Troy has worked at all levels of the technology industry for the last 30 years, from entry level positions to vice president for 1,000+ person technology organisations (Sabre Airlines, Travelocity, LastMinute.com, Corbis). He now consults for major companies like Tableau Software, Microsoft, Skype, Walmart, Siemens, LeanKit. Follow his work on Twitter: @t_magennis or email him a question at firstname.lastname@example.org.